===============================================================
Tag name: cam5_4_125
Originator(s): fvitt, joemci
Date: 12 May 2017
One-line Summary: Provide capability to send nitrogen deposition fluxes to surface models and WACCMX compset updates
Purpose of changes:
Provide capability to send nitrogen deposition fluxes to surface models.
This can be turned on by setting ndep_list in drv_flds_in which is read by
the driver.
Update WACCMX compsets.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/bld/build-namelist
. add ndep_inparm namelist group to drv_flds_in
M components/cam/bld/namelist_files/namelist_definition.xml
. new namelist variable "ndep_list" which will be in ndep_inparm namelist group to drv_flds_in
M components/cam/src/cpl/cam_cpl_indices.F90
M components/cam/src/control/camsrfexch.F90
. add noy and nhx nitrogen deposition flux arrays to cam_out object
M components/cam/src/cpl/atm_import_export.F90
. export noy and nhx nitrogen deposition fluxes
M components/cam/src/chemistry/mozart/mo_chm_diags.F90
. set the noy and nhx nitrogen deposition fluxes in cam_out object
. WD_NOY --> wet_deposition_NOy_as_N
. WD_NHX --> wet_deposition_NHx_as_N
. DF_NOY --> dry_deposition_NOy_as_N
. DF_NHX --> dry_deposition_NHx_as_N
M components/cam/src/chemistry/mozart/chemistry.F90
M components/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
. set the noy and nhx nitrogen deposition fluxes in cam_out object
M components/cam/src/chemistry/mozart/mo_neu_wetdep.F90
. change comment
WD_NOY --> wet_deposition_NOy_as_N
WD_NHX --> wet_deposition_NHx_as_N
M components/cam/bld/namelist_files/use_cases/2000_trop_strat_vbs_cam6.xml
M components/cam/bld/namelist_files/use_cases/2010_trop_strat_vbs_cam6.xml
M components/cam/bld/namelist_files/use_cases/hist_trop_strat_vbs_cam6.xml
M components/cam/bld/namelist_files/use_cases/sd_trop_strat_vbs_cam6.xml
M components/cam/bld/namelist_files/use_cases/sd_waccm_tsmlt_cam6.xml
M components/cam/bld/namelist_files/use_cases/waccm_ma_1850_cam6.xml
M components/cam/bld/namelist_files/use_cases/waccm_ma_2000_cam6.xml
M components/cam/bld/namelist_files/use_cases/waccm_ma_hist_cam6.xml
M components/cam/bld/namelist_files/use_cases/waccm_tsmlt_1850_cam6.xml
M components/cam/bld/namelist_files/use_cases/waccm_tsmlt_2000_cam6.xml
M components/cam/bld/namelist_files/use_cases/waccm_tsmlt_hist_cam6.xml
. WD_NOY --> wet_deposition_NOy_as_N
. WD_NHX --> wet_deposition_NHx_as_N
. DF_NOY --> dry_deposition_NOy_as_N
. DF_NHX --> dry_deposition_NHx_as_N
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
M components/cam/bld/namelist_files/use_cases/1950-2010_ccmi_refc1_waccmx_ma.xml
M components/cam/bld/namelist_files/use_cases/sd_waccmx_ma_cam4.xml
M components/cam/bld/namelist_files/use_cases/waccmxie_ma_2000_cam4.xml
M components/cam/cime_config/config_compsets.xml
M components/cam/cime_config/testdefs/testlist_cam.xml
- updates for waccmx compsets
M components/cam/src/ionosphere/waccmx/edyn_init.F90
M components/cam/src/ionosphere/waccmx/edynamo.F90
M components/cam/src/physics/cam/cam_diagnostics.F90
M components/cam/src/physics/cam/tidal_diag.F90
M components/cam/src/physics/waccm/radheat.F90
M components/cam/src/physics/waccmx/ion_electron_temp.F90
- added diagnostics
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all pass
yellowstone/intel/aux_cam:
ERP_Ld3.f09_f09.FWHIST.yellowstone_intel.cam-reduced_hist1d (Overall: NLFAIL) details:
FAIL ERP_Ld3.f09_f09.FWHIST.yellowstone_intel.cam-reduced_hist1d NLCOMP
ERS_Ln9.f10_f10.FSPCAMS.yellowstone_intel.cam-outfrq9s (Overall: FAIL) details:
FAIL ERS_Ln9.f10_f10.FSPCAMS.yellowstone_intel.cam-outfrq9s RUN time=16
SMS_D_Ln9.f19_f19.FXHIST.yellowstone_intel.cam-outfrq9s (Overall: FAIL) details:
FAIL SMS_D_Ln9.f19_f19.FXHIST.yellowstone_intel.cam-outfrq9s NLCOMP
FAIL SMS_D_Ln9.f19_f19.FXHIST.yellowstone_intel.cam-outfrq9s RUN time=118
SMS_Ld1.f09_f09.FW1850.yellowstone_intel.cam-reduced_hist1d (Overall: NLFAIŧL) details:
FAIL SMS_Ld1.f09_f09.FW1850.yellowstone_intel.cam-reduced_hist1d NLCOMP
Run failures for SPCAM and WACCMX are pre-existing due to CLM version
Change in electron_file is the cause of namelist compare failures for WACCM,
otherwise waccm is bit-for-bit unchanged.
hobart/nag: all pass
hobart/pgi: all pass
===============================================================
===============================================================
Tag name: cam5_4_124
Originator(s): fvitt
Date: 9 May 2017
One-line Summary: Misc WACCM bug fixes and compset updates
Purpose of changes:
WACCM updates:
- bug fix for EUV photolysis rates -- computed photolysis and heating rates
should not be dependent on the order of EUV reactions in chemistry mechanism file
- explicitly account for this NH3 loss as an NH4 gain in modal gas/aerosol exchange
- chemistry preprocessor changes
. increase the length of the reaction tag names - this fixes a bug where some of
the reaction tag names were truncated effectively disabling the corresponding
user reactions
. fix bug in vectorized solver -- change in vector length should not change results
- updated chemistry mechanisms for WACCM and CAM-Chem
- updated WACCM and cam-chem compsets -- update emissions, IC files, history streams, etc.
Bug fix for WACCM-SC:
- fix issue where prescribed H was being confused with NO
Bug fix for WACCM-X:
- initialize waccmx ionosphere sooner to avoid restart problem when history includes
fields decomposed on geo-magnetic grid -- history need to know about the geo-magnetic
grid earlier in a restarted run
Misc changes:
- added history flags
. history_chemspecies_srf
. history_cesm_forcing
. history_scwaccm_forcing
- units change in wet deposition flux (WD_*) diagnostics
- misc code clean up
- do not include tgcm_ubc_file and snoe_ubc_file namelist vars for WACCM-X
configurations -- this need for WACCM only
- regression tests updates
Bugs fixed (include bugzilla ID): see above
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: cacraig, eaton
List all subroutines eliminated:
D components/cam/src/chemistry/modal_aero/bam_nh4no3.F90
- remove this dead code
D components/cam/bld/namelist_files/use_cases/waccm_tsmlt_amip_cam6.xml
D components/cam/bld/namelist_files/use_cases/waccm_ma_amip_cam6.xml
D components/cam/bld/namelist_files/use_cases/waccm_sc_amip_cam6.xml
- replaced 'amip' with 'hist' in the file name
D components/cam/test/system/config_files/f4c5.4wtclbdm
D components/cam/test/system/config_files/f4c4prgspcdm
D components/cam/test/system/config_files/f1.9c4bamm
- replaced by aquaplanet configuration files
List all subroutines added and what they do:
A components/cam/bld/namelist_files/use_cases/waccm_ma_hist_cam6.xml
A components/cam/bld/namelist_files/use_cases/waccm_tsmlt_hist_cam6.xml
A components/cam/bld/namelist_files/use_cases/waccm_sc_hist_cam6.xml
- replaced 'amip' with 'hist' in the file name
A components/cam/bld/namelist_files/use_cases/hist_trop_strat_vbs_cam6.xml
A components/cam/bld/namelist_files/use_cases/2010_trop_strat_vbs_cam6.xml
A components/cam/bld/namelist_files/use_cases/sd_trop_strat_vbs_cam6.xml
- added for new cam-chem compsets
A components/cam/test/system/config_files/f4c5.5aqwmtdm
A components/cam/test/system/config_files/f1.9c4aqbamm
A components/cam/test/system/config_files/f4c4aqprgspcdm
- these include aquaplanet config option
List all existing files that have been modified, and describe the changes:
M components/cam/SVN_EXTERNAL_DIRECTORIES
- chemistry preprocessor changes
. increase the length of the reaction tag names
. fix bug in vectorized solver -- change in vector length should not change results
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/chem_mech.doc
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/chem_mech.in
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_adjrxt.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_rxt_rates_conv.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_imp_sol.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/chem_mods.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_prod_loss.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_lin_matrix.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_exp_sol.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/m_rxt_id.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_phtadj.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_nln_matrix.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_lu_factor.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_sim_dat.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_indprd.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/m_spc_id.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_lu_solve.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_setrxt.F90
- updated chemistry mechanism
- increase the length of the reaction tag names
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/chem_mech.doc
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/chem_mech.in
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/m_rxt_id.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_nln_matrix.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_lu_factor.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_sim_dat.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_indprd.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/m_spc_id.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_lu_solve.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_setrxt.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_adjrxt.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_rxt_rates_conv.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_imp_sol.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/chem_mods.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_prod_loss.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_lin_matrix.F90
- updated chemistry mechanism
- increase the length of the reaction tag names
M components/cam/src/chemistry/pp_waccm_mad_mam4/chem_mods.F90
M components/cam/src/chemistry/pp_waccm_mad_mam4/mo_sim_dat.F90
M components/cam/src/chemistry/pp_super_fast_llnl/mo_sim_dat.F90
M components/cam/src/chemistry/pp_super_fast_llnl/chem_mods.F90
M components/cam/src/chemistry/pp_waccm_sc_mam4/chem_mods.F90
M components/cam/src/chemistry/pp_waccm_sc_mam4/mo_sim_dat.F90
M components/cam/src/chemistry/pp_trop_mam3/mo_sim_dat.F90
M components/cam/src/chemistry/pp_trop_mam3/chem_mods.F90
M components/cam/src/chemistry/pp_trop_mam4/chem_mods.F90
M components/cam/src/chemistry/pp_trop_mam4/mo_sim_dat.F90
M components/cam/src/chemistry/pp_trop_bam/chem_mods.F90
M components/cam/src/chemistry/pp_trop_bam/mo_sim_dat.F90
M components/cam/src/chemistry/pp_trop_mam7/mo_sim_dat.F90
M components/cam/src/chemistry/pp_trop_mam7/chem_mods.F90
M components/cam/src/chemistry/pp_waccm_ma/chem_mods.F90
M components/cam/src/chemistry/pp_waccm_ma/mo_sim_dat.F90
M components/cam/src/chemistry/pp_waccm_ma_sulfur/chem_mods.F90
M components/cam/src/chemistry/pp_waccm_ma_sulfur/mo_sim_dat.F90
M components/cam/src/chemistry/pp_super_fast_llnl_mam3/chem_mods.F90
M components/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_sim_dat.F90
M components/cam/src/chemistry/pp_trop_ghg/mo_sim_dat.F90
M components/cam/src/chemistry/pp_trop_ghg/chem_mods.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_sulfur/m_rxt_id.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_sim_dat.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_sulfur/chem_mods.F90
M components/cam/src/chemistry/pp_waccm_tsmlt/chem_mods.F90
M components/cam/src/chemistry/pp_waccm_tsmlt/m_rxt_id.F90
M components/cam/src/chemistry/pp_waccm_tsmlt/mo_sim_dat.F90
M components/cam/src/chemistry/pp_waccm_sc/chem_mods.F90
M components/cam/src/chemistry/pp_waccm_sc/mo_sim_dat.F90
M components/cam/src/chemistry/pp_trop_mozart/chem_mods.F90
M components/cam/src/chemistry/pp_trop_mozart/mo_sim_dat.F90
M components/cam/src/chemistry/pp_terminator/chem_mods.F90
M components/cam/src/chemistry/pp_terminator/mo_sim_dat.F90
M components/cam/src/chemistry/pp_waccm_ma_mam4/mo_sim_dat.F90
M components/cam/src/chemistry/pp_waccm_ma_mam4/chem_mods.F90
- increase the length of the reaction tag names, otherwise the chemistry has not changed
M components/cam/src/chemistry/mozart/chemistry.F90
- invoke add_default for history_cesm_forcing flag
M components/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
- invoke add_default for history_cesm_forcing flag
- code cleanup
M components/cam/src/chemistry/mozart/mo_chm_diags.F90
- correction in NOy diagnostic to include 'PBZNIT'
- invoke add_default for history flags
. history_chemspecies_srf
. history_cesm_forcing
. history_scwaccm_forcing
M components/cam/src/chemistry/mozart/mo_extfrc.F90
- invoke add_default for history_cesm_forcing flag
- add vertically integrated external forcing diagnostic (kg/m2/sec)
M components/cam/src/chemistry/mozart/mo_lightning.F90
- invoke add_default for history_cesm_forcing flag
M components/cam/src/chemistry/mozart/mo_neu_wetdep.F90
- units change in wet deposition flux (WD_*) diagnostics
M components/cam/src/chemistry/mozart/mo_waccm_hrates.F90
- code clean up
M components/cam/src/chemistry/mozart/mo_usrrxt.F90
- add rates for 'TS1' chemistry
- code clean up
M components/cam/src/chemistry/mozart/mo_photo.F90
M components/cam/src/chemistry/mozart/mo_jeuv.F90
- bug fix for EUV photolysis rates -- computed photolysis and heating rates
should not be dependent on the order of EUV reactions in chemsitry mechanism file
M components/cam/src/chemistry/modal_aero/aero_model.F90
- invoke add_default for history_cesm_forcing flag
- explicitly account for this NH3 loss as an NH4 gain.
Because the NH3->NH4 on aerosols is actually already
taken into account that way, we don't need to have the call to
bam_nh4no3_formation.
M components/cam/src/chemistry/utils/mo_solar_parms.F90
- fix problem when solar data includes leap day and the calendar is set to noleap
M components/cam/src/dynamics/fv/ctem.F90
- invoke add_default for history_waccm flag
M components/cam/src/control/cam_comp.F90
- initialize waccmx ionosphere sooner to aviod restart
problem when history includes fields decomposed on
geo-magnetic grid -- history need to know about the
geo-magnetic grid sooner in a restarted run
M components/cam/src/physics/waccm/waccm_forcing.F90
- code clean up
- sc-waccm bug fix -- prescribed H was being confused with NO
M components/cam/src/physics/waccm/nlte_lw.F90
- code clean up
M components/cam/src/physics/waccm/radheat.F90
- invoke add_default for history_scwaccm_forcing flag
M components/cam/src/physics/cam/phys_control.F90
- added history flags
. history_chemspecies_srf
. history_cesm_forcing
. history_scwaccm_forcing
M components/cam/src/physics/cam/modal_aer_opt.F90
- change defaults for history_amwg flag
M components/cam/src/physics/cam/cam_diagnostics.F90
M components/cam/src/physics/cam/gw_drag.F90
- change defaults for history_waccm flag
M components/cam/src/physics/cam/nucleate_ice_cam.F90
- invoke add_default for history_cesm_forcing flag
M components/cam/src/physics/cam/tidal_diag.F90
- invoke add_default for history_waccm flag
M components/cam/src/physics/cam/check_energy.F90
- invoke add_default for history_waccm flag
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
- add default setting for history_chemspecies_srf flag
M components/cam/bld/build-namelist
- camchem trop-strat VBS LBC and emissions the same as waccm TSMLT
- do not include tgcm_ubc_file and snoe_ubc_file namelist vars for WACCM-X
configurations -- this need for WACCM only
- add default setting for history_chemspecies_srf flag
M components/cam/bld/namelist_files/namelist_definition.xml
- added history flags
. history_chemspecies_srf
. history_cesm_forcing
. history_scwaccm_forcing
M components/cam/bld/namelist_files/use_cases/waccm_tsmlt_2000_cam6.xml
M components/cam/bld/namelist_files/use_cases/2000_trop_strat_vbs_cam6.xml
M components/cam/bld/namelist_files/use_cases/waccm_ma_1850_cam6.xml
M components/cam/bld/namelist_files/use_cases/sd_waccm_tsmlt_cam6.xml
M components/cam/bld/namelist_files/use_cases/waccm_tsmlt_1850_cam6.xml
M components/cam/bld/namelist_files/use_cases/waccm_ma_2000_cam6.xml
M components/cam/cime_config/config_component.xml
M components/cam/cime_config/config_pes.xml
M components/cam/cime_config/config_compsets.xml
- compset updates
M components/cam/cime_config/testdefs/testlist_cam.xml
M components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq1d_newyear/user_nl_cam
- add geo-magnetic grid decomposed history fields
M components/cam/test/system/tests_waccm_mpi
M components/cam/test/system/input_tests_master
M components/cam/test/system/tests_chem_mpi
- regression tests updates
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
017 bl370 TBL.sh f1.9c6aqtsvbsdh outfrq3s 9s ..................................................FAIL! rc= 7 at Tue May 9 12:49:24 MDT 2017
- expected failure due to updates to trop_strat_mam4_vbs chemistry updates
020 bl371 TBL.sh f1.9c6aqwscdh outfrq3s_f19c6aqwsc+waccm_sc_hist_cam6 9s ......................FAIL! rc= 7 at Tue May 9 13:26:58 MDT 2017
- expected failure due to bug fix in sc-waccm (see above)
013 bl353 TBL.sh f1.9c4aqmozdh outfrq3s+2000_cam4_trop_chem 9s ................................FAIL! rc= 7 at Tue May 9 12:10:29 MDT 2017
015 bl360 TBL.sh f1.9c5aqm volc+1850-2005_cam5 2d .............................................FAIL! rc= 7 at Tue May 9 12:12:45 MDT 2017
023 bl380 TBL.sh f1.9c6aqcdh atrain 9s ........................................................FAIL! rc= 7 at Tue May 9 13:38:03 MDT 2017
026 bl390 TBL.sh f1.9c4aqwmxdh outfrq3s+waccmx_ma_2000_cam4 9s ................................FAIL! rc= 7 at Tue May 9 16:35:17 MDT 2017
029 bl391 TBL.sh f1.9c4aqwmxidh outfrq3s 9s ...................................................FAIL! rc= 7 at Tue May 9 16:42:17 MDT 2017
032 bl735 TBL.sh h16c5aqdm outfrq3s+1850-2005_cam5 9s .........................................FAIL! rc= 7 at Tue May 9 16:42:24 MDT 2017
- climate is bit-for-bit unchanged -- baseline failures are due to units change in wet deposition flux (WD_*) diagnostics
yellowstone/intel/aux_cam: CAM is B4B
ERP_Ld3.f09_f09.FWHIST.yellowstone_intel.cam-reduced_hist1d (Overall: DIFF) details:
FAIL ERP_Ld3.f09_f09.FWHIST.yellowstone_intel.cam-reduced_hist1d NLCOMP
FAIL ERP_Ld3.f09_f09.FWHIST.yellowstone_intel.cam-reduced_hist1d BASELINE ERROR BFAIL baseline directory '/glade/p/cesmdata/cseg/ccsm_baselines/cam5_4_123/ERP_Ld3.f09_f09.FWHIST.yellowstone_intel.cam-reduced_hist1d' does not exist
- new aux_cam test
ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s_clm5 (Overall: DIFF) details:
FAIL ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s_clm5 NLCOMP
FAIL ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s_clm5 BASELINE
- NLCOMP failed due to added namelist variable "history_chemspecies_srf"
- BASELINE failed due to units change in wet deposition flux (WD_*) diagnostics
otherwise this is bit-for-bit unchanged
ERP_Ln9.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq9s_clm5 (Overall: DIFF) details:
FAIL ERP_Ln9.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq9s_clm5 NLCOMP
FAIL ERP_Ln9.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq9s_clm5 BASELINE
- NLCOMP failed due to added namelist variable "history_chemspecies_srf"
- BASELINE failed due to units change in wet deposition flux (WD_*) diagnostics
otherwise this is bit-for-bit unchanged
ERS_Ld3.f19_f19.FXREFC1.yellowstone_intel.cam-outfrq1d_newyear (Overall: FAIL) details:
FAIL ERS_Ld3.f19_f19.FXREFC1.yellowstone_intel.cam-outfrq1d_newyear NLCOMP
FAIL ERS_Ld3.f19_f19.FXREFC1.yellowstone_intel.cam-outfrq1d_newyear RUN time=42
- fails due to issues with clm45 -- hopefully this will be resolved in future clm tag
ERS_Ln9.f10_f10.FSPCAMS.yellowstone_intel.cam-outfrq9s (Overall: FAIL) details:
FAIL ERS_Ln9.f10_f10.FSPCAMS.yellowstone_intel.cam-outfrq9s NLCOMP
FAIL ERS_Ln9.f10_f10.FSPCAMS.yellowstone_intel.cam-outfrq9s RUN time=16
- pre-existing failure due to issues with clm
SMS_Ld1.f09_f09.FW1850.yellowstone_intel.cam-reduced_hist1d (Overall: DIFF) details:
FAIL SMS_Ld1.f09_f09.FW1850.yellowstone_intel.cam-reduced_hist1d NLCOMP
FAIL SMS_Ld1.f09_f09.FW1850.yellowstone_intel.cam-reduced_hist1d BASELINE
- expected failure due to waccm compset updates
SMS_Lm13.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq1m_clm5 (Overall: DIFF) details:
FAIL SMS_Lm13.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq1m_clm5 NLCOMP
FAIL SMS_Lm13.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq1m_clm5 BASELINE
- NLCOMP failed due to added namelist variable "history_chemspecies_srf"
- BASELINE failed due to units change in wet deposition flux (WD_*) diagnostics
otherwise this is bit-for-bit unchanged
hobart/nag:
005 bl111 TBL.sh e8c5aqt5mdm ghgrmp_e8 9s .....................................................FAIL! rc= 7 at Tue May 9 09:19:32 MDT 2017
017 bl311 TBL.sh f10c5aqt5mdm ttrac 9s ........................................................FAIL! rc= 7 at Tue May 9 09:37:41 MDT 2017
027 bl325 TBL.sh f10c5aqscdm outfrq3s_subcol 9s ...............................................FAIL! rc= 7 at Tue May 9 09:53:15 MDT 2017
033 bl338 TBL.sh f10c5aqdm rad_diag_mam 9s ....................................................FAIL! rc= 7 at Tue May 9 09:57:30 MDT 2017
036 bl367 TBL.sh f10c6aqdm outfrq3s_convmic 9s ................................................FAIL! rc= 7 at Tue May 9 10:03:00 MDT 2017
039 bl368 TBL.sh f10c6aqdm outfrq3s 9s ........................................................FAIL! rc= 7 at Tue May 9 10:05:57 MDT 2017
042 bl369 TBL.sh f10c6aqdm outfrq3s_am 9s .....................................................FAIL! rc= 7 at Tue May 9 10:09:02 MDT 2017
045 bl426 TBL.sh f10c4aqwmxdm outfrq3s_newyear 9s .............................................FAIL! rc= 7 at Tue May 9 10:17:10 MDT 2017
048 bl430 TBL.sh f10c5aqcmtt1dm outfrq3s_carma 9s .............................................FAIL! rc= 7 at Tue May 9 10:27:15 MDT 2017
051 bl471 TBL.sh f10c6aqwmadm outfrq3s 9s .....................................................FAIL! rc= 7 at Tue May 9 10:37:53 MDT 2017
- climate is bit-for-bit unchanged -- baseline failures are due to units change in wet deposition flux (WD_*) diagnostics
hobart/pgi:
014 bl314 TBL.sh f10c4aqwscdm outfrq3s_diags 9s ...............................................FAIL! rc= 7 at Tue May 9 10:13:41 MDT 2017
- expected failure due to bug fix in sc-waccm (see above)
017 bl317 TBL.sh f10c5aqcdm outfrq3s_diags 9s .................................................FAIL! rc= 7 at Tue May 9 10:25:05 MDT 2017
023 bl321 TBL.sh f10c5aqcdm atrain 9s .........................................................FAIL! rc= 7 at Tue May 9 10:34:54 MDT 2017
026 bl334 TBL.sh f10c5aqudm outfrq3s 9s .......................................................FAIL! rc= 7 at Tue May 9 10:43:24 MDT 2017
032 bl511 TBL.sh s8c5aqt5mdm ttrac 9s .........................................................FAIL! rc= 7 at Tue May 9 10:56:06 MDT 2017
- climate is bit-for-bit unchanged -- baseline failures are due to units change in wet deposition flux (WD_*) diagnostics
===============================================================
===============================================================
Tag name: cam5_4_123
Originator(s): thomas.toniazzo@uni.no, cacraig
Date: May 8, 2017
One-line Summary: Angular momentum mods and additional dynamic wind increment diagnostics
Purpose of changes:
Purpose of changes:
- Angular momentum mods:
.correction or fixer increments no longer applied in first (top) four levels
.dycore correction: improved 2nd-order in time
.AM fixer: level fixer set as default, and tapering for global fixer
- Additional diagnostics of the dynamics wind increments
Bugs fixed (include bugzilla ID):
- corrected check for disallowing SPCAM being threaded to include sam1mom
- added initialization of cap3vc when high_alt is false
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
D components/cam/src/dynamics/fv/angular_momentum.F90
- removed unused routine
List all subroutines added and what they do:
A + components/cam/test/system/nl_files/outfrq3s_am
- added test turning on am_correction
List all existing files that have been modified, and describe the changes:
M components/cam/bld/configure
- Test for SPCAM threading was testing m2005 twice. Modified one
of these checks to be sam1mom
M components/cam/cime_config/testdefs/testlist_cam.xml
- Removed most of the yellowstone tests. Kept the ones needed for CAM's
regression testing until we move testing over completely to cheyenne
- changed ERP test to ERS for SPCAM since it is not doing threading and
failed on yellowstone for an unknown reason (ERP passed on cheyenne)
M components/cam/src/dynamics/fv/benergy.F90
- add use for par_vecsum
M components/cam/src/dynamics/fv/d2a3dikj.F90
- made a module
M components/cam/src/dynamics/fv/par_vecsum.F90
- put into a module and added an optional return summation variable
M components/cam/src/dynamics/fv/tp_core.F90
M components/cam/src/dynamics/fv/trac2d.F90
- removed obsolete INNER_OMP cpp directive
M components/cam/src/dynamics/fv/cd_core.F90
M components/cam/src/dynamics/fv/ctem.F90
M components/cam/src/dynamics/fv/dp_coupling.F90
M components/cam/src/dynamics/fv/dyn_comp.F90
M components/cam/src/dynamics/fv/sw_core.F90
M components/cam/src/dynamics/fv/te_map.F90
- modifications as described in Purpose section
M components/cam/test/system/input_tests_master
M components/cam/test/system/tests_pretag_hobart_nag
- added test for am_correction
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB
yellowstone/intel/aux_cam: all BFB except continue failing until update CLM external:
ERS_Ln9.f10_f10.FSPCAMS.yellowstone_intel.cam-outfrq9s (Overall: FAIL) details:
FAIL ERS_Ln9.f10_f10.FSPCAMS.yellowstone_intel.cam-outfrq9s NLCOMP
FAIL ERS_Ln9.f10_f10.FSPCAMS.yellowstone_intel.cam-outfrq9s RUN time=16
hobart/nag: all BFB except:
042 bl369 TBL.sh f10c6aqdm outfrq3s_am 9s .....................................................FAIL! rc= 7 at Fri May 5 19:09:07 MDT 2017
-- New test - no baseline for comparison
hobart/pgi: all BFB
===============================================================
===============================================================
Tag name: cam5_4_122
Originator(s): jet
Date: May 5, 2017
One-line Summary: Update to SCAM code - refactor forecast routine, bug fixes
Purpose of changes:
- Refactor the forecast routine of SCAM, better organized and more flexible. Misc bug fixes.
- added test for use of uninitialized variables
- updated scam defaults
- added more options for vertical advection scheme
- fixed bug in cam_pio_utils error in reading scam initial files.
- updated default testing to use ARM97 IOP.
- fixed errors when running adiabatically
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
Cleaned up namelist, removed non working and added new namelist variables to give user
more control over relaxation, what type of vertical advection is used for the prognostics
and which observed IOP variables are used in the forecast.
Removed:
scm_diurnal_avg
scm_iop_srf_prop
scm_clubb_iop_name
Added
scm_iop_lhflxshflxTg -- use iop latent/sensible/ground temp fluxes
scm_iop_Tg, -- use just ground temp fluxes
scm_cambfb_mode, -- set when using a cam generated IOP for bfb testing
scm_relax_top_p, -- set upper relaxation pressure limit
scm_relax_bot_p, -- set lower relaxation pressure limit
scm_relax_tau_sec, -- set relaxation constant
scm_relax_tau_top_sec, -- relaxation time at upper boundary
scm_relax_tau_bot_sec, -- relaxation time at lower boundary
scm_relax_linear, -- use a linear interpolation of relaxation time
between the upper and lower boundary relaxation time
scm_zadv_uv, -- vertical advection type for uv (valid values iop,eul,off)
scm_zadv_T, -- vertical advection type for T (valid values iop,eul,off)
scm_zadv_q, -- vertical advection type for q (valid values iop,eul,slt,off)
scm_use_obs_T, -- use IOP values of T for forecast
scm_use_obs_uv, -- use IOP values of uv for forecast
scm_use_obs_qv, -- use IOP values of qv for forecast
scm_force_latlon -- force scm to use given values of
lat/lon not closest from datasets
namelist relaxation example:
relax the atmosphere to obs between 20mb and 900mb
using a linear interpolation of the relaxation time throughout the
atmosphere ranging from 3hrs at the top to 1 day at the bottom
====================================
scm_relaxation = .true.
scm_relax_linear = .true.
scm_relax_bot_p = 90000.
scm_relax_tau_bot_sec = 86400.
scm_relax_tau_top_sec = 10800.
scm_relax_top_p = 2000.
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:goldy
List all subroutines eliminated:
D components/cam/src/physics/cam/iop_surf.F90
D components/cam/src/dynamics/eul/forecast.F90
List all subroutines added and what they do:
A components/cam/src/physics/cam/iop_forcing.F90 replaced iop_surf.F90
A components/cam/src/dynamics/eul/scmforecast.F90 modularized and
refactor of old forecast subroutine.
List all existing files that have been modified, and describe the changes:
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
M components/cam/bld/namelist_files/namelist_definition.xml
M components/cam/cime_config/config_compsets.xml
M components/cam/cime_config/config_pes.xml
M components/cam/doc/ChangeLog
M components/cam/src/chemistry/mozart/mo_drydep.F90
M components/cam/src/control/history_defaults.F90
M components/cam/src/control/history_scam.F90
M components/cam/src/control/scamMod.F90
M components/cam/src/dynamics/eul/dyn_comp.F90
M components/cam/src/dynamics/eul/dynpkg.F90
M components/cam/src/dynamics/eul/iop.F90
M components/cam/src/dynamics/eul/restart_dynamics.F90
M components/cam/src/dynamics/eul/scan2.F90
M components/cam/src/dynamics/eul/scanslt.F90
M components/cam/src/dynamics/eul/tfilt_massfix.F90
M components/cam/src/physics/cam/cam_diagnostics.F90
M components/cam/src/physics/cam/micro_mg_cam.F90
M components/cam/src/physics/cam/physpkg.F90
M components/cam/src/utils/cam_pio_utils.F90
M components/cam/test/system/nl_files/scm_b4b_o1
M components/cam/test/system/nl_files/scmarm
M components/cam/test/system/tests_pretag_hobart_nag
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB
yellowstone/intel/aux_cam: two known failures all others pass
ERP_Ln9.f10_f10.FSPCAMS.yellowstone_intel.cam-outfrq9s (Overall: FAIL) details:
FAIL ERP_Ln9.f10_f10.FSPCAMS.yellowstone_intel.cam-outfrq9s NLCOMP
FAIL ERP_Ln9.f10_f10.FSPCAMS.yellowstone_intel.cam-outfrq9s RUN time=28
hobart/nag: all BFB
hobart/pgi: all BFB
===============================================================
Tag name: cam5_4_121
Originator(s): cacraig
Date: May 4, 2017
One-line Summary: Update externals to match alpha06j except using CLM branch
Purpose of changes:
- Brought all externals up to match external used in cesm2_0_alpha06j
- CLM external is branch tag which is being used in Cecile's science runs (and contains
mods which are not yet on the CLM trunk)
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M .
M SVN_EXTERNAL_DIRECTORIES
- update externals to match alpha06j and use CLM branch which has
additional mods beyond the CLM trunk
M components/cam/test/system/TCB_ccsm.sh
M components/cam/test/system/TER_ccsm.sh
M components/cam/test/system/TSM_ccsm.sh
- add double dashes for cesm flags
- use new xmlchange structure
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB
yellowstone/intel/aux_cam: all have namelist changes and fail baselines due to
changes in the external components
ERP_Ln9.f10_f10.FSPCAMS.yellowstone_intel.cam-outfrq9s (Overall: FAIL) details:
FAIL ERP_Ln9.f10_f10.FSPCAMS.yellowstone_intel.cam-outfrq9s NLCOMP
FAIL ERP_Ln9.f10_f10.FSPCAMS.yellowstone_intel.cam-outfrq9s RUN time=25
- this requires a version of CLM at r236 or beyond
hobart/nag: all BFB
hobart/pgi: all BFB
===============================================================
===============================================================
Tag name: cam5_4_120
Originator(s): fvitt, joemci, liuh
Date: 24 Apr 2017
One-line Summary: Initial incorporation of a more complete ionosphere in WACCM-X
Purpose of changes:
- Add electrodynamics for the purpose of transporting O-plus in the ionosphere and
providing more realistic ion drift velocities
- Add transport of ions (just O+ in this commit) instead of transport by FV dynamical core
- Add energetics (temperature tendency) through solver for electron and ion temperature for
more realistic thermosphere temperature
- Major WACCM-X code reorganization, revision, and clean up for better compatibility with
existing infrastructure and better maintainability
- Add and revise compsets and namelist variables for WACCM-X
- A number additions to handle the fact that a few constants in the lower atmosphere are no
longer constant in the upper atmosphere where diffusion is dominant
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: CAM review team
List all subroutines eliminated:
D components/cam/bld/namelist_files/use_cases/waccmx_2000_cam4.xml
D components/cam/bld/namelist_files/use_cases/waccmxi_2000_cam4.xml
D components/cam/bld/namelist_files/use_cases/waccmx_1996_cam4.xml
- removed these obsolete use cases
D components/cam/src/physics/cam/ionosphere.F90
D components/cam/src/physics/cam/majorsp_diffusion.F90
- removed these stubs
D components/cam/src/physics/waccmx/ionosphere.F90
- renamed to ion_electron_temp
D components/cam/src/chemistry/mozart/wei96.F90
D components/cam/src/chemistry/mozart/efield.F90
D components/cam/src/chemistry/mozart/mag_parms.F90
D components/cam/src/chemistry/mozart/exbdrift.F90
- moved to physcis/waccm
List all subroutines added and what they do:
A components/cam/src/ionosphere
A components/cam/src/ionosphere/waccmx/dpie_coupling.F90
A components/cam/src/ionosphere/waccmx/edyn_init.F90
A components/cam/src/ionosphere/waccmx/filter.F90
A components/cam/src/ionosphere/waccmx/edyn_maggrid.F90
A components/cam/src/ionosphere/waccmx/ionosphere_interface.F90
A components/cam/src/ionosphere/waccmx/edynamo.F90
A components/cam/src/ionosphere/waccmx/getapex.F90
A components/cam/src/ionosphere/waccmx/edyn_geogrid.F90
A components/cam/src/ionosphere/waccmx/oplus.F90
A components/cam/src/ionosphere/waccmx/edyn_params.F90
A components/cam/src/ionosphere/waccmx/edyn_mudcom.F90
A components/cam/src/ionosphere/waccmx/edyn_mpi.F90
A components/cam/src/ionosphere/waccmx/edyn_mud.F90
A components/cam/src/ionosphere/waccmx/edyn_mudmod.F90
A components/cam/src/ionosphere/waccmx/edyn_solve.F90
A components/cam/src/ionosphere/waccmx/edyn_esmf.F90
A components/cam/src/ionosphere/waccmx/heelis.F90
A components/cam/src/ionosphere/waccmx/savefield_waccm.F90
A components/cam/src/ionosphere/waccmx/edyn_muh2cr.F90
A components/cam/src/ionosphere/waccmx/wei05sc.F90
A components/cam/src/ionosphere/waccmx
- waccmx ionosphere added -- electrodynamcis and o-plus ion transport
A components/cam/src/ionosphere/ionosphere_interface.F90
- stub interface
A components/cam/src/physics/waccm/exbdrift.F90
A components/cam/src/physics/waccm/efield.F90
A components/cam/src/physics/waccm/wei05sc.F90
A components/cam/src/physics/waccm/wei96.F90
A components/cam/src/physics/waccm/mo_aurora.F90
A components/cam/src/physics/waccm/mag_parms.F90
- moved from chemistry/mozart
A components/cam/src/physics/cam/waccmx_phys_intr.F90
- provides a general interface to waccm physics modules
A components/cam/src/physics/waccmx/ion_electron_temp.F90
- previously named "ionosphere.F90"
A components/cam/bld/config_files/defaults_waccmxie.xml
- default configuration settings for waccmx ionosphere edynamo
A components/cam/test/system/config_files/f1.9c4aqwmxidh
A components/cam/test/system/config_files/f4c6aqwmadm
A components/cam/test/system/config_files/f4c4aqwmxdm
A components/cam/test/system/config_files/f4c4aqwmxidm
A components/cam/test/system/config_files/f4c4aqwmxiedm
- new test configuration used in waccmx testing
A components/cam/test/system/nl_files/outfrq3s_newyear
A components/cam/bld/namelist_files/use_cases/1950-2010_ccmi_refc1_waccmx_ma.xml
A components/cam/bld/namelist_files/use_cases/waccmxie_ma_2000_cam4.xml
A components/cam/bld/namelist_files/use_cases/sd_waccmx_ma_cam4.xml
A components/cam/bld/namelist_files/use_cases/waccmx_ma_2000_cam4.xml
A components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq1d_newyear/user_nl_clm
A components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq1d_newyear/user_nl_cam
A components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq1d_newyear/shell_commands
A components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq1d_newyear
List all existing files that have been modified, and describe the changes:
M components/cam/test/system/input_tests_master
M components/cam/test/system/nl_files/outfrq3s_epp
M components/cam/test/system/nl_files/outfrq3s_ionos
M components/cam/test/system/tests_waccm_hybrid
- updates for waccmx
M components/cam/test/system/TR8.sh
- add r8 check for code in src/ionosphere
M components/cam/bld/build-namelist
- add SF6 to default flbc_list if in the mechanism
- set default waccmx ionosphere options
- set default fv_high_altitude for waccmx
M components/cam/bld/configure
- added ionosphere configure option
M components/cam/bld/Makefile.in
- add mpi_alltoallv to nag_mismatch_flag
M components/cam/bld/config_files/definition.xml
- added ionosphere config option
M components/cam/bld/namelist_files/use_cases/waccm_ma_2000_cam6.xml
M components/cam/bld/namelist_files/use_cases/sd_waccm5_mad_geos5.xml
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
- adjust waccmx default settings
M components/cam/bld/namelist_files/namelist_definition.xml
- add namelist options:
. fv_high_altitude
. oplus_adiff_limiter
. oplus_shapiro_const
. oplus_enforce_floor
. ionos_xport_active
. ionos_edyn_active
. emirical_ion_velocities
. dadadj_niter
M components/cam/src/physics/cam/constituents.F90
- use gas constant from shr_const_mod rather physconst
M components/cam/src/physics/cam/dadadj_cam.F90
M components/cam/src/physics/cam/dadadj.F90
- use variable Kappa
- add diagnostic
- add dadadj_niter namelist option
M components/cam/src/physics/cam/physics_types.F90
- physconst_update interface change
M components/cam/src/physics/cam/vertical_diffusion.F90
M components/cam/src/physics/cam/diffusion_solver.F90
- modify ubc_flux to be 2-dimensional
M components/cam/src/physics/cam/physpkg.F90
- use waccm_phys_inter to interface with waccm physics
M components/cam/src/physics/cam/iondrag.F90
- added timestep_init interface to this stub
- misc clean up
M components/cam/src/physics/waccm/qbo.F90
M components/cam/src/physics/waccm/radheat.F90
M components/cam/src/physics/waccm/nlte_fomichev.F90
M components/cam/src/physics/waccm/nlte_lw.F90
- misc clean up
M components/cam/src/physics/waccm/iondrag.F90
- add empirical_ion_velocities namelist option
M components/cam/src/physics/waccmx/majorsp_diffusion.F90
- mods for 2-dimensional fluxes at upper boundary
- misc clean up
- add diagnostic MBARV
M components/cam/src/chemistry/mozart/mo_aurora.F90
- now a stub version
M components/cam/src/chemistry/mozart/mo_waccm_hrates.F90
M components/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
- pbuf names of Te and Ti changed
M components/cam/src/chemistry/mozart/mo_apex.F90
- provide ability to update mag parameters when mag year changes
M components/cam/src/chemistry/mozart/euvac.F90
- misc clean up
M components/cam/src/chemistry/mozart/chemistry.F90
- corrected 'O1D' species name lookup in chem_register
- misc clean up
M components/cam/src/chemistry/mozart/upper_bc.F90
- mods for 2-dimensional fluxes at upper boundary
- misc clean up
M components/cam/src/chemistry/utils/apex.F90
- misc clean up
M components/cam/src/chemistry/utils/mo_solar_parms.F90
- changes to handle solar params which are more frequent than daily
M components/cam/src/chemistry/pp_waccm_ma/mo_sim_dat.F90
M components/cam/src/chemistry/pp_waccm_ma/mo_indprd.F90
M components/cam/src/chemistry/pp_waccm_ma/m_spc_id.F90
M components/cam/src/chemistry/pp_waccm_ma/chem_mech.doc
M components/cam/src/chemistry/pp_waccm_ma/chem_mech.in
M components/cam/src/chemistry/pp_waccm_ma/mo_setrxt.F90
M components/cam/src/chemistry/pp_waccm_ma/mo_adjrxt.F90
M components/cam/src/chemistry/pp_waccm_ma/mo_rxt_rates_conv.F90
M components/cam/src/chemistry/pp_waccm_ma/chem_mods.F90
M components/cam/src/chemistry/pp_waccm_ma/mo_prod_loss.F90
M components/cam/src/chemistry/pp_waccm_ma/mo_lin_matrix.F90
M components/cam/src/chemistry/pp_waccm_ma/m_rxt_id.F90
M components/cam/src/chemistry/pp_waccm_ma/mo_phtadj.F90
M components/cam/src/chemistry/pp_waccm_ma/mo_nln_matrix.F90
- include SF6
M components/cam/src/dynamics/fv/metdata.F90
M components/cam/src/dynamics/fv/benergy.F90
M components/cam/src/dynamics/fv/dp_coupling.F90
M components/cam/src/dynamics/fv/dyn_comp.F90
M components/cam/src/dynamics/fv/pkez.F90
M components/cam/src/dynamics/fv/p_d_adjust.F90
M components/cam/src/dynamics/fv/cd_core.F90
M components/cam/src/dynamics/fv/dynamics_vars.F90
M components/cam/src/dynamics/fv/diag_dynvar_ic.F90
M components/cam/src/dynamics/fv/geopk.F90
M components/cam/src/dynamics/fv/te_map.F90
M components/cam/src/dynamics/fv/stepon.F90
- changes for variable Kappa needed for the upper regions of waccmx
M components/cam/src/control/cam_restart.F90
M components/cam/src/control/cam_comp.F90
M components/cam/src/control/runtime_opts.F90
- interface with ionosphere
M components/cam/src/utils/physconst.F90
- add routine to calculate variable Kappa
M components/cam/cime_config/config_component.xml
M components/cam/cime_config/config_pes.xml
M components/cam/cime_config/config_compsets.xml
M components/cam/cime_config/testdefs/testlist_cam.xml
- waccmx compsets:
. FXN2000
. FXREFC1
. FX2000
. FSDX
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
026 bl390 TBL.sh f1.9c4aqwmxdh outfrq3s+waccmx_ma_2000_cam4 9s ................................FAIL! rc= 7 at Fri Apr 21 21:29:44 MDT 2017
029 bl391 TBL.sh f1.9c4aqwmxidh outfrq3s 9s ...................................................FAIL! rc= 7 at Fri Apr 21 22:11:11 MDT 2017
- expected failures due to waccmx updates
040 sm995 TSM_ccsm.sh T42_T42 FDABIP04 9s .....................................................FAIL! rc= 4 at Fri Apr 21 22:27:29 MDT 2017
041 eq995 TEQ_ccsm.sh T42_T42 FDABIP04 e64adh fcase+dabi_p2004 9s .............................FAIL! rc= 4 at Fri Apr 21 22:27:29 MDT 2017
- pre-existing failures
yellowstone/intel/aux_cam: B4B
ERP_Ld3.f09_f09.FWAMIP.yellowstone_intel.cam-reduced_hist1d (Overall: NLFAIL), details:
FAIL ERP_Ld3.f09_f09.FWAMIP.yellowstone_intel.cam-reduced_hist1d NLCOMP
SMS_Ld1.f09_f09.FW1850.yellowstone_intel.cam-reduced_hist1d (Overall: NLFAIL), details:
FAIL SMS_Ld1.f09_f09.FW1850.yellowstone_intel.cam-reduced_hist1d NLCOMP
namelist compare failure due to updated euvac_file, otherwise bit-for-bit unchanged
hobart/nag:
042 bl426 TBL.sh f10c4aqwmxdm outfrq3s_newyear 9s .............................................FAIL! rc= 7 at Fri Apr 21 15:24:14 MDT 2017
- expected failure due to waccmx updates
hobart/pgi: all pass
===============================================================
===============================================================
Tag name: cam5_4_119
Originator(s): cacraig
Date: April 19, 2017
One-line Summary: Add preliminary support for cheyenne regression testing and misc bug fixes
Purpose of changes:
- Started support for cheyenne regression testing. This is just a first step in the process,
and is far from being complete. It is being saved to use as a starting point for further
development.
- Removed illegal characters from namelist_definition.xml file
- Remove SPCAM branch test as it was unnecessary and took a long time
- Remove CAM version information from FSPCAM compsets
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
A components/cam/test/system/tests_pretag_cheyenne
- copy of yellowstone pretag tests for cheyenne runs
List all existing files that have been modified, and describe the changes:
M components/cam/bld/namelist_files/namelist_definition.xml
- removed illegal characters from xml file
M components/cam/cime_config/config_compsets.xml
- removed CAM version information from SPCAM compsets (it was incorrectly setting -phys twice)
M components/cam/cime_config/testdefs/testlist_cam.xml
- added aux_cam_test1 to test a single CESM compset
- duplicated all yellowstone runs to cheyenne runs
M components/cam/test/system/CAM_runcmnd.sh
- add capability to run cheyenne jobs -- note the compute node names start with "r"
M components/cam/test/system/test_driver.sh
- add the ability to run CAM regression testing on cheyenne nodes
- This is just a first step, and there are still many issues to resolve
M components/cam/test/system/tests_pretag_hobart_pgi
- Removed br222 (branch test for SPCAMM) as it was unnecessary and took a lot of time
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All pass except: (see below)
040 sm995 TSM_ccsm.sh T42_T42 FDABIP04 9s .....................................................FAIL! rc= 4 at Tue Apr 18 21:47:08 MDT 2017
041 eq995 TEQ_ccsm.sh T42_T42 FDABIP04 e64adh fcase+dabi_p2004 9s .............................FAIL! rc= 4 at Tue Apr 18 21:47:09 MDT 2017
yellowstone/intel/aux_cam: all PASS except:
ERP_Ln9.f10_f10.FSPCAMS.yellowstone_intel.cam-outfrq9s (Overall: DIFF), details:
FAIL ERP_Ln9.f10_f10.FSPCAMS.yellowstone_intel.cam-outfrq9s NLCOMP
FAIL ERP_Ln9.f10_f10.FSPCAMS.yellowstone_intel.cam-outfrq9s BASELINE
- This is due to the change in no longer supplying -phys cam4 in addition to -phys sam1mom. The theory is that
the answer changes are due to cice no longer being triggered with cam4 for some of its namelist settings. SPCAMS
will be checked with these new settings and if it is important to maintain the cam4 settings for cice, that
namelist can be set with user_mods when a newer version of cime is brought into CAM.
- Note that a special 9 timestep run of SPCAMM was run offline and its answers were identical, so it is not
impacted by the removal of -phys cam5 from the CAM configure
hobart/nag: all BFB
hobart/pgi: all BFB
cheyenne tests summary:
- All CESM tests fail due to internal compiler error based on needing HAS_F2008_CONTIGUOUS to be set to FALSE. This is set
in a more recent cime tag (Will need to temporarily set this in the CAM Makefile as well for the CAM CESM tests)
- Default time for running CAM regression tests may be way too low (jobs are being terminated without explanation) at test
#12 for first pass, and #16 during rerun
- test er331, er353 and sm360 are exiting with a message: ERROR: spmd_readnl: ERROR: incorrect yz domain decomposition
===============================================================
===============================================================
Tag name: cam5_4_118
Originator(s): andrew, cacraig, Guang Zhang
Date: March 28, 2017
One-line Summary: Add option for MG microphysics in ZM deep convection
Purpose of changes:
- Add Song and Zhang 2012 version of MG 2-moment microphysics in ZM
convective scheme as an option
- Add dlfzm to the pbuf
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
- Add new zmconv_microp namelist field which, if true, turns on the
ZM microphysics. By default, it is currently set to false
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
A + components/cam/src/physics/cam/zm_microphysics.F90
- added new routine for ZM convective microphysics, linked to MG
microphysics scheme
A + components/cam/test/system/nl_files/outfrq3s_convmic
- added namelist file for testing convective microphysics
List all existing files that have been modified, and describe the changes:
M components/cam/bld/build-namelist
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
M components/cam/bld/namelist_files/namelist_definition.xml
- add zmconv_microp namelist - defaults to false
M components/cam/src/physics/cam/clubb_intr.F90
M components/cam/src/physics/cam/macrop_driver.F90
M components/cam/src/physics/cam/ndrop.F90
M components/cam/src/physics/cam/nucleate_ice.F90
M components/cam/src/physics/cam/zm_conv.F90
M components/cam/src/physics/cam/zm_conv_intr.F90
- added appropriate hooks into ZM scheme for the new
ZM convective microphysics, linked to MG microphysics
M components/cam/src/physics/cam/convect_deep.F90
- dlf is now a member of pbuf
- added rice (reserved ice for energy integrals)
M components/cam/src/physics/cam/physpkg.F90
- dlf is now a member of pbuf
- added rice (reserved ice for energy integrals)
- add rice temporarily into snow_dp before calling check_energy_change
M components/cam/src/physics/spcam/crmclouds_camaerosols.F90
- added ztodt to convtran call
M components/cam/test/system/input_tests_master
M components/cam/test/system/tests_pretag_hobart_nag
- added test for convective microphysics
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB except:
040 sm995 TSM_ccsm.sh T42_T42 FDABIP04 9s .....................................................FAIL! rc= 4 at Mon Mar 27 22:29:03 MDT 2017
041 eq995 TEQ_ccsm.sh T42_T42 FDABIP04 e64adh fcase+dabi_p2004 9s .............................FAIL! rc= 4 at Mon Mar 27 22:29:03 MDT 2017
- expected failures due to pre-existing problems (see below)
yellowstone/intel/aux_cam: all PASS except:
ERP_Ld3.f09_f09.FWAMIP.yellowstone_intel.cam-reduced_hist1d (Overall: NLFAIL), details:
FAIL ERP_Ld3.f09_f09.FWAMIP.yellowstone_intel.cam-reduced_hist1d NLCOMP
ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s_clm5 (Overall: NLFAIL), details:
FAIL ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s_clm5 NLCOMP
ERP_Ln9.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq9s_clm5 (Overall: NLFAIL), details:
FAIL ERP_Ln9.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq9s_clm5 NLCOMP
ERP_Ln9.f10_f10.FSPCAMS.yellowstone_intel.cam-outfrq9s (Overall: NLFAIL), details:
FAIL ERP_Ln9.f10_f10.FSPCAMS.yellowstone_intel.cam-outfrq9s NLCOMP
SMS_Ld1.f09_f09.FW1850.yellowstone_intel.cam-reduced_hist1d (Overall: NLFAIL), details:
FAIL SMS_Ld1.f09_f09.FW1850.yellowstone_intel.cam-reduced_hist1d NLCOMP
SMS_Lm13.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq1m_clm5 (Overall: NLFAIL), details:
FAIL SMS_Lm13.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq1m_clm5 NLCOMP
- expected failure due to addition of zmconv_microp
hobart/nag: all BFB except:
036 bl367 TBL.sh f10c6aqdm outfrq3s_convmic 9s ................................................FAIL! rc= 7 at Mon Mar 27 17:54:55 MDT 2017
- new test so missing baseline
hobart/pgi: all BFB
===============================================================
===============================================================
Tag name: cam5_4_117
Originator(s): mmills, rneely, fvitt
Date: 23 Mar 2017
One-line Summary: Allow 3-mode specification of prescribed volcanic aerosols
Purpose of changes:
. Provide the ability to prescribe stratospheric sulfate aerosols in 3 modes input
to radiation transfer calculations. The old 1-mode specification is preserved for
backwards compatibility.
Bugs fixed (include bugzilla ID):
. Fix a bug in regression tests archive script so that all aux_cam test cases
are archived regardless of baseline compare status.
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/bld/namelist_files/namelist_definition.xml
- new namelist option "prescribed_strataero_3modes"
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
M components/cam/bld/build-namelist
- provide the correct default radiation properties files for the 3 sulfate modes
- provide a working prescribed stratospheric sulfate aerosols input data set in 3 modes
M components/cam/src/chemistry/utils/prescribed_strataero.F90
- expanded to handle 3 modes of prescribed stratospheric aerosols
M components/cam/src/physics/cam/phys_prop.F90
M components/cam/src/physics/cam/aer_rad_props.F90
- expanded for 3 radii of volcanic aerosols
M components/cam/test/system/archive_baseline.sh
- added "-p" option to the bless_test_results invocation
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
040 sm995 TSM_ccsm.sh T42_T42 FDABIP04 9s .....................................................FAIL! rc= 4 at Thu Mar 23 01:35:12 MDT 2017
041 eq995 TEQ_ccsm.sh T42_T42 FDABIP04 e64adh fcase+dabi_p2004 9s .............................FAIL! rc= 4 at Thu Mar 23 01:35:12 MDT 2017
- expected failures due to pre-existing problems (see below)
yellowstone/intel/aux_cam: all pass
hobart/nag: all pass
hobart/pgi: all pass
===============================================================
===============================================================
Tag name: cam5_4_116
Originator(s): fvitt, mickelso
Date: 21 Mar 2017
One-line Summary: Performance improvements to vectorized chemistry solver
Purpose of changes:
Improve performance of chemistry solver used in WACCM6 on yellowstone. This is
done by selecting a vector length of 64. This vector length is controlled by a
parameter which was added to the chemistry preprocessor input file.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/SVN_EXTERNAL_DIRECTORIES
- update chemisry preprocessor external
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/chem_mech.in
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/chem_mods.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_imp_sol.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_lin_matrix.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_lu_factor.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_lu_solve.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_nln_matrix.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_prod_loss.F90
- use vector length of 64
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
040 sm995 TSM_ccsm.sh T42_T42 FDABIP04 9s .....................................................FAIL! rc= 4 at Wed Mar 22 00:54:25 MDT 2017
041 eq995 TEQ_ccsm.sh T42_T42 FDABIP04 e64adh fcase+dabi_p2004 9s .............................FAIL! rc= 4 at Wed Mar 22 00:54:25 MDT 2017
- expected failures due to pre-existing problems (see below)
yellowstone/intel/aux_cam:
ERP_Ld3.f09_f09.FWAMIP.yellowstone_intel.cam-reduced_hist1d (Overall: NLFAIL), details:
FAIL ERP_Ld3.f09_f09.FWAMIP.yellowstone_intel.cam-reduced_hist1d NLCOMP
SMS_Ld1.f09_f09.FW1850.yellowstone_intel.cam-reduced_hist1d (Overall: NLFAIL), details:
FAIL SMS_Ld1.f09_f09.FW1850.yellowstone_intel.cam-reduced_hist1d NLCOMP
- namelist compare failures due to change in chemistry preprocessor input file chem_mech.in
hobart/nag: all pass
hobart/pgi: all pass
===============================================================
===============================================================
Tag name: cam5_4_115
Originator(s): cacraig, sacks@ucar.edu
Date: March 17, 2017
One-line Summary: Update externals to match cesm2_0_alpha06f and some fixes to unit tests
Purpose of changes:
- Update externals to match those used in cesm2_0_alpha06f
- Fix some of the broken unit tests (cime has an issue in unit testing
which breaks a couple of unit tests, but it does not indicate a
problem with CAM or the running of jobs in CESM)
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
A components/cam/test/unit/README.txt
- Added a README for running unit tests -- Was copied from CLM and slightly modified
List all existing files that have been modified, and describe the changes:
M .
M SVN_EXTERNAL_DIRECTORIES
- update to cesm2_0_alpha06f externals
- update cime and CLM (to corresponding branch needed by this version of cime)
- added new cime_config (which was removed from cime and is now in its own github repository)
M components/cam/bld/build-namelist
M components/cam/bld/namelist_files/namelist_definition.xml
- gust_fac does not have a defaut value, so CAM standalone needs to supply it
M components/cam/bld/Makefile.in
M components/cam/bld/configure
- directory reorganization in cime
M components/cam/bld/namelist_files/use_cases/hist_cam6.xml
- add missing micro_do_sb_physics set to true
M components/cam/cime_config/config_component.xml
- Added missing descriptions for %SPCAMCLBS and %SPCAMCLBM
M components/cam/test/unit/CMakeLists.txt
- changed variable identifiers from CESM to CIME
- cime directory reorganization changes
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB except:
040 sm995 TSM_ccsm.sh T42_T42 FDABIP04 9s .....................................................FAIL! rc= 4 at Wed Mar 15 21:34:01 MDT 2017
041 eq995 TEQ_ccsm.sh T42_T42 FDABIP04 e64adh fcase+dabi_p2004 9s .............................FAIL! rc= 4 at Wed Mar 15 21:34:01 MDT 2017
- This test fails due to the incomplete removal of atm_adiabatic and atm_ideal
which still remain in cime. Reran these tests with the head of cime (which
is not tagged yet) and these tests pass, so they should pass the next time
the CAM trunk updates its cime external
yellowstone/intel/aux_cam: all Baselines fail due to answer changes in CLM
all Namelist comparisons fail due to bad saving of namelists in previous tag
-- NOTES -- Namelists should be stored properly with this tag and PASS in the next tag
-- Baselines will be rechecked against Cecile's runs when the CLM branch that
she is using is moved to the CLM trunk. There was no way to check the answers
with this CLM configuration (but it was needed due to changes in cime)
hobart/nag: all BFB
hobart/pgi: all BFB
===============================================================
===============================================================
Tag name: cam5_4_114
Originator(s): cacraig, mark@atmos.colostate.edu, minghuai.wang@nju.edu.cn
Date: March 10, 2017
One-line Summary: Bug fixes to SPCAM
Purpose of changes:
- Scale the local ECPP tendencies by necpp in crm_physics_tend
- add optional variables to return vertical values of aqso4_h2o2 and aqso4_o3_
- Bugfix for cnst_species_class n modal_aero_data (only caused issues for SPCAM)
- Now use bulk prescribed aerosols for sam1mom
Bugs fixed (include bugzilla ID):
- The consituents had a couple of bugs in modal_aero_data that were causing problems
for m2005 SPCAM runs. After running all the CAM regression tests, this bug only
seems to impact m2005 SPCAM.
Describe any changes made to build system:
- FSPCAMS is now using bulk prescribed aerosols
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/bld/build-namelist
M components/cam/bld/configure
- SPCAM sam1mom now uses bulk prescribed aerosols
M components/cam/cime_config/config_component.xml
- added FSPCAMCLBS and FSPCAMCLBM to allow SPCAM m2005 and sam1mom to run with CLUBB
M components/cam/cime_config/config_compsets.xml
- update to using CLM50
- add compsets FSPCAMCLBS and FSPCAMCLBM
M components/cam/cime_config/testdefs/testlist_cam.xml
- add prealpha and prebeta tests for SPCAM
M components/cam/src/chemistry/aerosol/mo_setsox.F90
M components/cam/src/chemistry/bulk_aero/sox_cldaero_mod.F90
M components/cam/src/chemistry/modal_aero/sox_cldaero_mod.F90
- add optional variables to return the vertical values of aqso4_h2o2 and aqso4_o3
M components/cam/src/chemistry/modal_aero/modal_aero_data.F90
- removed redundant setting of constituents and use the correct index for the
call which is used
M components/cam/src/physics/spcam/crm_physics.F90
- modifications to account for ecpp timestep not equaling the CAM timestep
M components/cam/src/physics/spcam/ecpp/ecpp_modal_cloudchem.F90
- use the values of aqso4_h2o2 and aqso4_o3 at each vertical level
M components/cam/src/physics/spcam/ecpp/module_data_ecpp1.F90
- removed unused declarations
M components/cam/src/physics/spcam/ecpp/module_ecpp_ppdriver2.F90
- limit the fields which are output to only the ones who have active constituents
M components/cam/src/physics/spcam/ecpp/module_ecpp_td2clm.F90
- cleanup of diagnostics
M components/cam/test/system/config_files/f10spsaqdm
- change chem to "none" for sam1mom
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB except:
029 bl391 TBL.sh f1.9c4aqwmxdh outfrq3s+waccmxi_2000_cam4 9s ..................................FAIL! rc= 7 at Mon Mar 13 10:30:24 MDT 2017
- For some reason this baseline passed. Three more identical runs were made
and five tests were run with different number of tasks and all of them passed baseline.
It is assumed that something on caldera/yellowstone caused a momentary glitch in the
original run. If this happens again on this test, further investigation may be warranted.
041 eq995 TEQ_ccsm.sh T42_T42 FDABIP04 e64adh fcase+dabi_p2004 9s .............................FAIL! rc= 7 at Mon Mar 13 10:30:25 MDT 2017
- continuing failure due to issue with cime (first reported in cam5_4_105)
yellowstone/intel/aux_cam: all BFB except:
- namelists still have issues due to cime
ERP_Ln9.f10_f10.FSPCAMS.yellowstone_intel.cam-outfrq9s (Overall: DIFF), details:
FAIL ERP_Ln9.f10_f10.FSPCAMS.yellowstone_intel.cam-outfrq9s BASELINE
- answer changing tag for SPCAM
hobart/nag: all BFB except:
013 bl221 TBL.sh f10spsaqdm outfrq3s 9s .......................................................FAIL! rc= 7 at Thu Mar 9 17:06:00 MST 2017
- answer changing tag for SPCAM
hobart/pgi: all BFB except:
012 bl222 TBL.sh f10spmaqdm outfrq3s 9s .......................................................FAIL! rc= 7 at Thu Mar 9 18:01:01 MST 2017
- answer changing tag for SPCAM
===============================================================
===============================================================
Tag name: cam5_4_113
Originator(s): cacraig, fvitt
Date: March 6, 2017
One-line Summary: Initialize all variables before they are used
Purpose of changes:
- Use the "-nan" compiler flag in NAG to find uninitialized variables
which are used before they are set
- CAM-SE has issues, but since it has a major rewrite coming soon to
the CAM trunk, temporarily removed the CAM-SE tests from the NAG test suite
- SCAM also has issues and will have an upcoming tag which will include fixes
for uninitialized values
- remove the remaining references to atm_adiabatic and atm_ideal_phys namelist options
Bugs fixed (include bugzilla ID):
- gw_front.F90 had a bug in the midpoint_interp call where it was sending an array
subsection which contained unset memory. This is an answer-changing bug fix
for WACCM and WACCMX.
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton, fvitt, joemic, hanli
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/bld/Makefile.in
- add NAN initialization for Intel and NAG debug modes (note NAG
testing is much more robust than Intel)
M components/cam/bld/build-namelist
M components/cam/bld/namelist_files/namelist_definition.xml
- remove atm_adiabatic and atm_ideal_phys references
M components/cam/src/chemistry/mozart/mo_aurora.F90
- Initialize byloc
M components/cam/src/chemistry/utils/apex.F90
- a number of variables need to be intent(inout) to keep the
initializations that occurred before these routines are called
M components/cam/src/physics/cam/clubb_intr.F90
- the pbuf field FICE needs to be initialized on restarts as well
- limit comparisons to just the array subsection which has been set
M components/cam/src/physics/cam/convect_shallow.F90
- initialize the pbuf fields SH_FLXPRC and SH_FLXSNW
M components/cam/src/physics/cam/gw_front.F90
- Fixes a bug that was introduced in cam5_4_33. The function
midpoint_interp was including values outside the range that had
been set -- This change is what causes answer changes
for WACCM and WACCMX
M components/cam/src/physics/cam/macrop_driver.F90
- initialize FICE and CMELIQ on restarts as well
M components/cam/src/physics/cam/micro_mg_cam.F90
- limit cldmax calculation to array subsection which has been set
M components/cam/src/physics/cam/micro_mg_utils.F90
- initialize am_evp_st
M components/cam/src/physics/cam/ndrop.F90
- limit calculation of lcldo and lcldn to array subsections
which have been set
M components/cam/src/physics/cam/physpkg.F90
- limit adding of CLDICE with CLDLIQ to just those sections
which have been set
- only pass the array subsection of flx_heat which has been
set into check_energy_chng
M components/cam/src/physics/cam/sslt_rebin.F90
- remove unused routines from a use statement
M components/cam/src/physics/camrt/radiation.F90
- set fsnr and flnr to zero if they are not active
M components/cam/src/physics/rrtmg/radiation.F90
- initialize pbuf field QRL
M components/cam/src/physics/spcam/crm_physics.F90
- initialize pbuf field AST only on the first time step
M components/cam/src/physics/waccmx/majorsp_diffusion.F90
- tubc was only being set on the initialization step and
it needed to be set every time mspdiff is called -- Fix
provided by Francis Vitt
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
020 bl371 TBL.sh f1.9c6aqwscdh outfrq3s_f19c6aqwsc+waccm_sc_amip_cam6 9s ......................FAIL! rc= 7 at Fri Mar 3 18:59:29 MST 2017
026 bl390 TBL.sh f1.9c4aqwmxdh outfrq3s+waccmx_2000_cam4 9s ...................................FAIL! rc= 7 at Fri Mar 3 19:57:01 MST 2017
029 bl391 TBL.sh f1.9c4aqwmxdh outfrq3s+waccmxi_2000_cam4 9s ..................................FAIL! rc= 7 at Fri Mar 3 20:41:10 MST 2017
- Answer changes as expected for WACCM/WACCMX
041 eq995 TEQ_ccsm.sh T42_T42 FDABIP04 e64adh fcase+dabi_p2004 9s .............................FAIL! rc= 7 at Fri Mar 3 21:00:39 MST 2017
- continuing failure due to issue with cime (first reported in cam5_4_105)
yellowstone/intel/aux_cam:
ERP_Ld3.f09_f09.FWAMIP.yellowstone_intel.cam-reduced_hist1d (Overall: DIFF), details:
FAIL ERP_Ld3.f09_f09.FWAMIP.yellowstone_intel.cam-reduced_hist1d BASELINE
SMS_Ld1.f09_f09.FW1850.yellowstone_intel.cam-reduced_hist1d (Overall: DIFF), details:
FAIL SMS_Ld1.f09_f09.FW1850.yellowstone_intel.cam-reduced_hist1d BASELINE
- In addition to namelist failures due to problems in cime, answer changes as expected for WACCM/WACCMX
hobart/nag:
039 bl426 TBL.sh f10c4aqwmxdm outfrq3s_ionos 9s ...............................................FAIL! rc= 7 at Fri Mar 3 13:11:04 MST 2017
045 bl471 TBL.sh f10c6aqwmadm outfrq3s 9s .....................................................FAIL! rc= 7 at Fri Mar 3 13:35:05 MST 2017
- Answer changes as expected for WACCM/WACCMX
046 sm711 TSM.sh h5c5aqt5mdm ttrac 9s .........................................................FAIL! rc= 8 at Fri Mar 3 13:37:09 MST 2017
047 er711 TER.sh h5c5aqt5mdm ttrac 4+5s .......................................................FAIL! rc= 5 at Fri Mar 3 13:37:10 MST 2017
048 br711 TBR.sh h5c5aqt5mdm ttrac 6+3s .......................................................FAIL! rc= 5 at Fri Mar 3 13:37:17 MST 2017
049 bl711 TBL.sh h5c5aqt5mdm ttrac 9s .........................................................FAIL! rc= 4 at Fri Mar 3 13:37:17 MST 2017
050 mc711 TMC.sh h5c5aqt5mdm ttrac 9s .........................................................FAIL! rc= 4 at Fri Mar 3 13:37:17 MST 2017
051 eq701 TEQ.sh h5c5aqt5mdm ttrac h5c5aqdm ttrac_lb0 9s ......................................FAIL! rc= 4 at Fri Mar 3 13:37:17 MST 2017
052 eq702 TEQ.sh h5c5aqt5mdm ttrac h5c5aqdm ttrac_lb1 9s ......................................FAIL! rc= 4 at Fri Mar 3 13:37:17 MST 2017
053 eq703 TEQ.sh h5c5aqt5mdm ttrac h5c5aqdm ttrac_lb3 9s ......................................FAIL! rc= 4 at Fri Mar 3 13:37:17 MST 2017
054 sm741 TSM.sh h16c6aqdm ghgrmp_unstruct 9s .................................................FAIL! rc= 8 at Fri Mar 3 13:39:29 MST 2017
055 er741 TER.sh h16c6aqdm ghgrmp_unstruct 4+5s ...............................................FAIL! rc= 5 at Fri Mar 3 13:39:29 MST 2017
056 bl741 TBL.sh h16c6aqdm ghgrmp_unstruct 9s .................................................FAIL! rc= 4 at Fri Mar 3 13:39:29 MST 2017
- There are known initialization issues with CAM-SE and since a major update is coming soon, these tests
are being removed until that tag is made
059 sc003 TSM.sh scmc5aqds scmarm 7s ..........................................................FAIL! rc= 8 at Fri Mar 3 13:46:14 MST 2017
- There were a good number of initialization issues with SCAM. These will be fixed when the upcoming
SCAM tag is made on the CAM trunk. Until then, this test is being removed.
hobart/pgi:
015 bl314 TBL.sh f10c4aqwscdm outfrq3s_diags 9s ...............................................FAIL! rc= 7 at Fri Mar 3 13:35:24 MST 2017
- Answer changes as expected for WACCM/WACCMX
Summarize any changes to answers, i.e.,
- what code configurations: all WACCM/WACCMX
- what platforms/compilers: all
- nature of change (roundoff; larger than roundoff but same climate; new
climate): larger than roundoff, but same climate
URL for AMWG diagnostics output used to validate the WACCM answer change:
https://acomstaff.acom.ucar.edu/mmills/diagnostics/WACCM6/FW2000.gwfront_7y/
===============================================================
===============================================================
Tag name: cam5_4_112
Originator(s): juliob, cacraig
Date: Feb 27, 2017
One-line Summary: Low level and gravity wave changes and several minor mods
Purpose of changes:
- Namelist control variables added to gw_rdg.F90
- Allow code to reproduce Scinocca&McFarlane 2000 downslope wind and flow splitting
regimes, or modified regime definition using separate dividing streamlines
for downslope wind and flow splitting
- Optional smooth transitions to downslope wind regime implemented
- Code cleanup of unused variables in gw_rdg module
- Rename all CESM tests which are not routinely run from "aux_" to "test_"
- Update default topo file for ne30np4
- Implement bug fix in fv/spmd_dyn which was causing CAM to endrun when CAM tasks > npes_yz
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
- Added the following namelist variables for gw_rdg
gw_rdg_do_divstream -- If .true. use separate dividing streamlines for downslope wind and
flow splitting regimes (“DS” configuration). If .false. use single
dividing streamline as in Scinocca&McFarlane 2000 (“SM” configuration).
gw_rdg_C_BetaMax_DS -- Enhancement factor for downslope wind stress in DS configuration
gw_rdg_C_GammaMax -- Enhancement factor for depth of downslope wind regime in DS configuration
gw_rdg_Frx0 -- Lower inverse Froude number limits on linear ramp terminating downslope wind
regime for high mountains in DS configuration
gw_rdg_Frx1 -- Upper inverse Froude number limits on linear ramp terminating downslope wind
regime for high mountains in DS configuration
gw_rdg_C_BetaMax_SM -- Enhancement factor for downslope wind stress in DS configuration
gw_rdg_Fr_c -- Critical inverse4 Froude number
gw_rdg_do_smooth_regimes -- If true, then use smooth regimes
gw_rdg_do_adjust_tauoro -- If true, then adujust tauoro
gw_rdg_do_backward_compat -- If true, then adjust for bit-for-bit answers with the ("N5") configuration
gw_rdg_orohmin -- minimum surface displacement height for orographic waves (m)
gw_rdg_orovmin -- Minimum wind speed for orographic waves
gw_rdg_orostratmin -- Minimum stratification allowing wave behavior
gw_rdg_orom2min -- Minimum stratification allowing wave behavior
List any changes to the defaults for the boundary datasets:
Topo file for ne30np4 has been updated
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/bld/build-namelist
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
M components/cam/bld/namelist_files/namelist_definition.xml
- Add gravity wave ridge namelist settings
- Update ne30np4 default topo file
M components/cam/cime_config/testdefs/testlist_cam.xml
- Rename all CESM aux_XXX tests other than aux_cam and aux_cam_short to test_XXX
M components/cam/src/dynamics/fv/spmd_dyn.F90
- two sections of code should only be executed by master, otherwise the code
was calling endrun when CAM tasks > npex_yz
M components/cam/src/physics/cam/gw_drag.F90
- cleanup of unused variables
M components/cam/src/physics/cam/gw_rdg.F90
- Namelist control variables added to gw_rdg.F90
- Allow code to reproduce Scinocca&McFarlane 2000 downslope wind and flow splitting
regimes, or modified regime definition using separate dividing streamlines
for downslope wind and flow splitting
- Optional smooth transitions to downslope wind regime implemented
- Code cleanup of unused variables in gw_rdg module
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB except:
041 eq995 TEQ_ccsm.sh T42_T42 FDABIP04 e64adh fcase+dabi_p2004 9s .............................FAIL! rc= 7 at Fri Feb 24 21:10:36 MST 2017
- continuing failure due to issue with cime (first reported in cam5_4_105)
yellowstone/intel/aux_cam: all PASS except still issues with namelst testing in cime
hobart/nag: all BFB
hobart/pgi: all BFB
===============================================================
===============================================================
Tag name: cam5_4_111
Originator(s): fvitt
Date: 14 Feb 2017
One-line Summary: Update emissions input data sets used by WACCM6 compsets
Purpose of changes:
Previous CAM trunk tags (cam5_4_107 through cam5_4_110) used emissions input
data sets which have been lost. Here WACCM6 build namelist use case files
point to updated emissions data sets.
Reduce history data saved in the baselines of the aux_cam WACCM tests.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
components/cam/cime_config/testdefs/testmods_dirs/cam/reduced_hist1d
components/cam/cime_config/testdefs/testmods_dirs/cam/reduced_hist1d/user_nl_cam
- added for the purpose of reducing the amount of data saved in the baselines
List all existing files that have been modified, and describe the changes:
components/cam/bld/namelist_files/use_cases/waccm_tsmlt_2000_cam6.xml
components/cam/bld/namelist_files/use_cases/waccm_tsmlt_amip_cam6.xml
components/cam/bld/namelist_files/use_cases/waccm_tsmlt_1850_cam6.xml
- updated 1-degree emissions data sets
components/cam/bld/namelist_files/use_cases/waccmxi_2000_cam4.xml
- replace full path to IC with path relative to inputdata
components/cam/cime_config/testdefs/testlist_cam.xml
- WACCM aux_cam test cases use reduced_hist1d modifier
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
041 eq995 TEQ_ccsm.sh T42_T42 FDABIP04 e64adh fcase+dabi_p2004 9s .............................FAIL! rc= 7 at Mon Feb 13 23:40:46 MST 2017
- pre-existing failure due to issue with cime (first reported in cam5_4_105)
yellowstone/intel/aux_cam:
ERP_Ld3.f09_f09.FWAMIP.yellowstone_intel.cam-reduced_hist1d (Overall: DIFF), details:
FAIL ERP_Ld3.f09_f09.FWAMIP.yellowstone_intel.cam-reduced_hist1d NLCOMP
FAIL ERP_Ld3.f09_f09.FWAMIP.yellowstone_intel.cam-reduced_hist1d BASELINE ERROR BFAIL baseline directory '/glade/p/cesmdata/cseg/ccsm_baselines/cam5_4_110/ERP_Ld3.f09_f09.FWAMIP.yellowstone_intel.cam-reduced_hist1d' does not exist
ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s_clm5 (Overall: NLFAIL), details:
FAIL ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s_clm5 NLCOMP
ERP_Ln9.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq9s_clm5 (Overall: NLFAIL), details:
FAIL ERP_Ln9.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq9s_clm5 NLCOMP
ERP_Ln9.f10_f10.FSPCAMS.yellowstone_intel.cam-outfrq9s (Overall: NLFAIL), details:
FAIL ERP_Ln9.f10_f10.FSPCAMS.yellowstone_intel.cam-outfrq9s NLCOMP
SMS_Ld1.f09_f09.FW1850.yellowstone_intel.cam-reduced_hist1d (Overall: DIFF), details:
FAIL SMS_Ld1.f09_f09.FW1850.yellowstone_intel.cam-reduced_hist1d NLCOMP
FAIL SMS_Ld1.f09_f09.FW1850.yellowstone_intel.cam-reduced_hist1d BASELINE ERROR BFAIL baseline directory '/glade/p/cesmdata/cseg/ccsm_baselines/cam5_4_110/SMS_Ld1.f09_f09.FW1850.yellowstone_intel.cam-reduced_hist1d' does not exist
- waccm compsets are expected to give different results due to changes in emssions input data
- namelist comparisons seem to be broken
hobart/nag: All Pass
hobart/pgi: All Pass
===============================================================
===============================================================
Tag name: cam5_4_110
Originator(s): goldy, pel
Date: 2017-02-12
One-line Summary: Add Kessler 'simple' physics option
Purpose of changes: Add a new, 'simple' physcs option, Kessler
Bugs fixed (include bugzilla ID): NA
Describe any changes made to build system: NA
Describe any changes made to the namelist: NA
List any changes to the defaults for the boundary datasets: NA
Describe any substantial timing or memory changes: NA
Code reviewed by: goldy
List all subroutines eliminated: NA
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/bld/config_files/definition.xml
M components/cam/bld/configure
- Add Kessler as a physics option
A + components/cam/bld/namelist_files/use_cases/dctest_baro_kessler.xml
M components/cam/cime_config/config_component.xml
M components/cam/cime_config/config_compsets.xml
- Compset and use case for Kessler physics with baroclinic wave initial conditions
M components/cam/cime_config/testdefs/testlist_cam.xml
- prealpha test for Kessler physics option
M components/cam/doc/ChangeLog
- Used the GNU ChangeBot to write this entry
M components/cam/src/control/cam_control_mod.F90
- Added Kessler physics setting
M components/cam/src/dynamics/eul/dyn_comp.F90
- Make use of pertlim namelist parameter repeatable (bug fix)
M components/cam/src/dynamics/fv/stepon.F90
- Do not do dry air mass check and set if using analytic initial conditions
M components/cam/src/dynamics/se/dp_coupling.F90
M components/cam/src/dynamics/se/dyn_comp.F90
- Initialize state%PHIS in unused columns to avoid warnings
M components/cam/src/dynamics/sld/dyn_comp.F90
- Make use of pertlim namelist parameter repeatable (bug fix)
M components/cam/src/physics/cam/cam_diagnostics.F90
- Do not assume that any moist physics package has all cloud and precipitation pbuf and tracer fields
M components/cam/src/physics/cam/physics_types.F90
- Dry mass pressure adjustments
M components/cam/src/physics/simple/held_suarez_cam.F90
- Code cleanup
A + components/cam/src/physics/simple/kessler_cam.F90
A + components/cam/src/physics/simple/kessler_mod.F90
- New Kessler physics interface and module
M components/cam/src/physics/simple/physpkg.F90
- Call Kessler physics when appropriate
- Move diagnostics write to after physics parameterizations (apparently,
people like to look at results after they have been calculated, who knew?)
M components/cam/src/cam/simple/restart_physics.F90
A + components/cam/src/physics/simple/restart_physics.F90
- Add minimal restart_physics for simple options allowing less clutter in CAM version
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
006 bl151 TBL.sh e64hsdh outfrq3s+held_suarez_1994 9s .........................................FAIL! rc= 7 at Sun Feb 12 01:52:12 MST 2017
035 bl736 TBL.sh h16adtermdh terminator 9s ....................................................FAIL! rc= 7 at Sun Feb 12 05:08:58 MST 2017
- Expected failures due to change in diagnostics
041 eq995 TEQ_ccsm.sh T42_T42 FDABIP04 e64adh fcase+dabi_p2004 9s .............................FAIL! rc= 7 at Sun Feb 12 05:12:01 MST 2017
- failure due to issue with cime (first reported in cam5_4_105)
yellowstone/intel/aux_cam: Namelist failures due to inadequate bless-test-results
ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s_clm5 (Overall: NLFAIL), details:
FAIL ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s_clm5 NLCOMP
ERP_Ln9.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq9s_clm5 (Overall: NLFAIL), details:
FAIL ERP_Ln9.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq9s_clm5 NLCOMP
ERP_Ln9.f10_f10.FSPCAMS.yellowstone_intel.cam-outfrq9s (Overall: NLFAIL), details:
FAIL ERP_Ln9.f10_f10.FSPCAMS.yellowstone_intel.cam-outfrq9s NLCOMP
SMS_Lm13.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq1m_clm5 (Overall: NLFAIL), details:
FAIL SMS_Lm13.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq1m_clm5 NLCOMP
hobart/nag:
031 bl335 TBL.sh f10idm idphys 9s .............................................................FAIL! rc= 7 at Sun Feb 12 12:53:07 MST 2017
- Expected failure due to change in diagnostics
hobart/pgi:
030 bl339 TBL.sh f10adhterm terminator 9s .....................................................FAIL! rc= 7 at Sun Feb 12 13:53:17 MST 2017
- Expected failure due to change in diagnostics
CAM tag used for the baseline comparison tests if different than previous
tag: NA
Summarize any changes to answers, i.e.,
- what code configurations: NA
- what platforms/compilers: NA
- nature of change: NA
If bitwise differences were observed, how did you show they were no worse
than roundoff? NA
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository): NA
===============================================================
===============================================================
Tag name: cam5_4_109
Originator(s): cacraig
Date: Feb 8, 2017
One-line Summary: Fix topo file issues discovered during prealpha testing
Purpose of changes:
- Fixed a couple of problems that were discovered with the assignment
of topo files for different compsets/resolutions
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
- Update the topo default files for 1 and 2 degree FV to Julio's
current files which contain the GBXAR field
M components/cam/bld/namelist_files/use_cases/hist_cam6.xml
- removed the hardwired topo file as the defaults now supply it
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
041 eq995 TEQ_ccsm.sh T42_T42 FDABIP04 e64adh fcase+dabi_p2004 9s .............................FAIL! rc= 7 at Tue Feb 7 19:53:38 MST 2017
- failure due to issue with cime (first reported in cam5_4_105)
yellowstone/intel/aux_cam:
ERP_Ld3.f09_f09.FWAMIP.yellowstone_intel.cam-outfrq9s (Overall: NLFAIL), details:
FAIL ERP_Ld3.f09_f09.FWAMIP.yellowstone_intel.cam-outfrq9s NLCOMP
ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s_clm5 (Overall: NLFAIL), details:
FAIL ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s_clm5 NLCOMP
ERP_Ln9.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq9s_clm5 (Overall: NLFAIL), details:
FAIL ERP_Ln9.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq9s_clm5 NLCOMP
ERP_Ln9.f10_f10.FSPCAMS.yellowstone_intel.cam-outfrq9s (Overall: NLFAIL), details:
FAIL ERP_Ln9.f10_f10.FSPCAMS.yellowstone_intel.cam-outfrq9s NLCOMP
- Failures due to incorrect baseline management (bless_test_results?)
SMS_Ld1.f09_f09.FW1850.yellowstone_intel.cam-outfrq1d (Overall: NLFAIL), details:
FAIL SMS_Ld1.f09_f09.FW1850.yellowstone_intel.cam-outfrq1d NLCOMP
- Failure due to change in default topo file (bug fixed)
hobart/nag: not run (changes were limited to CESM tests which only run on yellowstone)
hobart/pgi: not run (changes were limited to CESM tests which only run on yellowstone)
===============================================================
===============================================================
Tag name: cam5_4_108
Originator(s): goldy
Date: 2017-02-05
One-line Summary: Remove CLM40 from CAM
Purpose of changes: Remove CLM40 compsets from CAM to allow CLM group cleanup
Bugs fixed (include bugzilla ID): NA
Describe any changes made to build system: NA
Describe any changes made to the namelist: NA
List any changes to the defaults for the boundary datasets: NA
Describe any substantial timing or memory changes: NA
Code reviewed by: goldy
List all subroutines eliminated: NA
List all subroutines added and what they do: NA
List all existing files that have been modified, and describe the
changes:
M SVN_EXTERNAL_DIRECTORIES
- Removed CLM40 external (no longer used by CAM)
M components/cam/cime_config/config_compsets.xml
- Grouped existing CLM45 compsets and converted existing CLM40 compsets to CLM45
M components/cam/doc/ChangeLog
- Rewrote this file from scratch (that's a true alternative fact).
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
041 eq995 TEQ_ccsm.sh T42_T42 FDABIP04 e64adh fcase+dabi_p2004 9s .............................FAIL! rc= 7 at Sat Feb 4 14:17:42 MST 2017
Roundoff level diffs in a few fields (including gw which is a constant),
probably due to different compiler settings.
yellowstone/intel/aux_cam:
FAIL ERP_Ld3.f09_f09.FWAMIP.yellowstone_intel.cam-outfrq9s NLCOMP
FAIL ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s_clm5 NLCOMP
FAIL ERP_Ln9.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq9s_clm5 NLCOMP
FAIL ERP_Ln9.f10_f10.FSPCAMS.yellowstone_intel.cam-outfrq9s NLCOMP
FAIL SMS_Ld1.f09_f09.FW1850.yellowstone_intel.cam-outfrq1d NLCOMP
FAIL SMS_Lm13.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq1m_clm5 NLCOMP
Failures due to incorrect baseline management (bless_test_results?)
FAIL ERP_Ln9.f10_f10.FSPCAMS.yellowstone_intel.cam-outfrq9s BASELINE
Expected failure due to change to CLM45
hobart/nag: ALL PASS
hobart/pgi: ALL PASS
CAM tag used for the baseline comparison tests if different than previous
tag: NA
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate): NA
===============================================================
===============================================================
Tag name: cam5_4_107
Originator(s): fvitt
Date: 2 Feb 2017
One-line Summary: Corrections to ammonium aerosols, update boundary datasets, switch to vectorized chemistry solver in WACCM6
Purpose of changes:
WACCM6 updates:
- include formation of bulk ammonium nitrate
- add wet removal of ammonium aerosols (as gas phase species)
- switch from scalar solver to vector solver
- update emissions and fixed lower boundary data
- set fixed geomagnetic year for perpetual year use cases
Misc fixes:
- correctly set arrays to avoid undefined values
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
A components/cam/src/chemistry/modal_aero/bam_nh4no3.F90
- for formation of bulk ammonium nitrate
List all existing files that have been modified, and describe the changes:
M components/cam/bld/namelist_files/master_aer_wetdep_list.xml
M components/cam/bld/namelist_files/master_gas_wetdep_list.xml
- move wet deposition of NH3 aerosols to gas phase species list
M components/cam/bld/namelist_files/use_cases/sd_waccm_tsmlt_cam6.xml
M components/cam/bld/namelist_files/use_cases/waccm_ma_1850_cam6.xml
M components/cam/bld/namelist_files/use_cases/waccm_ma_2000_cam6.xml
M components/cam/bld/namelist_files/use_cases/waccm_ma_amip_cam6.xml
M components/cam/bld/namelist_files/use_cases/waccm_sc_amip_cam6.xml
M components/cam/bld/namelist_files/use_cases/waccm_tsmlt_1850_cam6.xml
M components/cam/bld/namelist_files/use_cases/waccm_tsmlt_2000_cam6.xml
M components/cam/bld/namelist_files/use_cases/waccm_tsmlt_amip_cam6.xml
M components/cam/bld/namelist_files/use_cases/waccmx_2000_cam4.xml
M components/cam/bld/namelist_files/use_cases/waccmxi_2000_cam4.xml
- update emissions and fixed lower boundary data
- set fixed geo-magnetic year for perpetual year waccm use cases
M components/cam/src/chemistry/modal_aero/aero_model.F90
- include formation of bulk NH4 aeroeols
M components/cam/src/chemistry/mozart/chemistry.F90
M components/cam/src/chemistry/mozart/mo_chm_diags.F90
M components/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
M components/cam/src/chemistry/mozart/mo_neu_wetdep.F90
- pass wetdep fluxes into chm_diags for history output
- correction to NOy summation
- include NH4 and NH4NO3 with wet deposition of gas phase species
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/chem_mech.in
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_exp_sol.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_imp_sol.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_indprd.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_lin_matrix.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_lu_factor.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_lu_solve.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_nln_matrix.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_prod_loss.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_setrxt.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_sim_dat.F90
- switch from scalar solver to vector solver
M components/cam/src/chemistry/utils/mo_solar_parms.F90
M components/cam/src/dynamics/fv/dyn_comp.F90
M components/cam/src/physics/carma/cam/carma_intr.F90
- correctly set arrays to zero to avoid undefined values
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
013 bl353 TBL.sh f1.9c4aqmozdh outfrq3s+2000_cam4_trop_chem 9s ................................FAIL! rc= 7 at Thu Feb 2 14:14:39 MST 2017
- expected failure due to wet removal of ammonium aerosols as gas phase species
020 bl371 TBL.sh f1.9c6aqwscdh outfrq3s_f19c6aqwsc+waccm_sc_amip_cam6 9s ......................FAIL! rc= 7 at Thu Feb 2 15:26:47 MST 2017
- expected failures due to change in LBC file
026 bl390 TBL.sh f1.9c4aqwmxdh outfrq3s+waccmx_2000_cam4 9s ...................................FAIL! rc= 7 at Thu Feb 2 15:52:18 MST 2017
029 bl391 TBL.sh f1.9c4aqwmxdh outfrq3s+waccmxi_2000_cam4 9s ..................................FAIL! rc= 7 at Thu Feb 2 15:54:09 MST 2017
- expected failures due to change in geomagnetic year
041 eq995 TEQ_ccsm.sh T42_T42 FDABIP04 e64adh fcase+dabi_p2004 9s .............................FAIL! rc= 7 at Thu Feb 2 15:54:10 MST 2017
- failure due to issue with cime (first reported in cam5_4_105)
yellowstone/intel/aux_cam:
ERP_Ld3.f09_f09.FWAMIP.yellowstone_intel.cam-outfrq9s (Overall: DIFF), details:
FAIL ERP_Ld3.f09_f09.FWAMIP.yellowstone_intel.cam-outfrq9s NLCOMP
FAIL ERP_Ld3.f09_f09.FWAMIP.yellowstone_intel.cam-outfrq9s BASELINE
SMS_Ld1.f09_f09.FW1850.yellowstone_intel.cam-outfrq1d (Overall: DIFF), details:
FAIL SMS_Ld1.f09_f09.FW1850.yellowstone_intel.cam-outfrq1d NLCOMP
FAIL SMS_Ld1.f09_f09.FW1850.yellowstone_intel.cam-outfrq1d BASELINE
- expected failures due to change in boundary datasets
ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s_clm5 (Overall: NLFAIL), details:
FAIL ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s_clm5 NLCOMP
ERP_Ln9.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq9s_clm5 (Overall: NLFAIL), details:
FAIL ERP_Ln9.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq9s_clm5 NLCOMP
ERP_Ln9.f10_f10.FSPCAMS.yellowstone_intel.cam-outfrq9s (Overall: NLFAIL), details:
FAIL ERP_Ln9.f10_f10.FSPCAMS.yellowstone_intel.cam-outfrq9s NLCOMP
- these failures in namelist comparisons are due to differences in PE layouts,
otherwise the results are bit-for-bit unchanged
hobart/nag: all pass
hobart/pgi: all pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: WACCM6, bulk aerosol model
- what platforms/compilers: all
- nature of change: larger than roundoff but same climate
===============================================================
===============================================================
Tag name: cam5_4_106
Originator(s): cacraig
Date: Jan 26, 2017
One-line Summary: Update the svn externals to match cesm2_0_alpha06a
Purpose of changes:
- Update the CAM externals to match cesm2_0_alpha06
- In a couple of outfld calls in the CLUBB interface, there were
variables which had arithemetic being performed on them within the call.
Now perform the arithmetic prior to the outfld call and pass that variable
to outfld
-- CHANGES for CAM TESTING (only useful for developers) --
- Now that cime has been updated to allow the override of threading, add
the SPCAMS compset test to aux_cam. Also added testing the SPCAMM compset
to the aux_spcam testing
- Use the override capability for any CAM compsets which use CISM1 to be forced
to use one node and one thread for that component
- With the update to cime, bless_test_results will now copy the namelists, so
remove the --hist-only flag from the baseline archiving
- Modifications to test_driver.sh to support rerunning CESM tests and
having the CESM bld and run directories go into the case setup directory
when running the CAM regression tests
- remove the h5c4aqwmdh test as it was never set up to use aquaplanet
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
D components/cam/test/system/config_files/h5c4aqwmdh
- this test removed as it did not have aquaplanet in
it (despite its name) and creating the initialization
file was problematic. Since this was a WACCM4 test using
SE dycore (which is not supported), Brian Eaton decided
to drop the test
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M .
M SVN_EXTERNAL_DIRECTORIES
- update the externals to match those in cesm2_0_alpha06a
M components/cam/cime_config/config_compsets.xml
- remove obsolete single thread definitions (now in config_pes.xml)
M components/cam/cime_config/config_pes.xml
- add single threading for SPCAM compsets
- all compsets which use CISM1 need single node and thread for
that component
M components/cam/cime_config/testdefs/testlist_cam.xml
- add compset for SPCAM m2005 testing with aux_spcam and add
FSPCAMS to aux_cam testing using the 10deg grid
M components/cam/src/physics/cam/clubb_intr.F90
- outfld calls with arithmetic in the calls were replaced
with temporary variables and the arithmetic is performed
before the call
M components/cam/test/system/TCB_ccsm.sh
- add "--run-unsupported" to create_newcase to always test compsets, even
when they do not appear in the CESM testlist
M components/cam/test/system/archive_baseline.sh
- remove "--hist-only" which was required due to bug in bless_test_results
which has now been fixed
M components/cam/test/system/test_driver.sh
- "--rerun-cesm" now requires the test_id to be supplied as well
M components/cam/test/system/input_tests_master
M components/cam/test/system/tests_pretag_yellowstone
- removed test "760" (see the information about the deleted
config_file above for more information)
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB except:
044 eq995 TEQ_ccsm.sh T42_T42 FDABIP04 e64adh fcase+dabi_p2004 9s .............................FAIL! rc= 7 at Wed Jan 25 18:56:06 MST 2017
-- Failure due to issue with cime (first reported in cam5_4_105)
yellowstone/intel/aux_cam:
ERP_Ld3.f09_f09.FWAMIP.yellowstone_intel.cam-outfrq9s (Overall: DIFF), details:
FAIL ERP_Ld3.f09_f09.FWAMIP.yellowstone_intel.cam-outfrq9s NLCOMP
FAIL ERP_Ld3.f09_f09.FWAMIP.yellowstone_intel.cam-outfrq9s BASELINE
ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s_clm5 (Overall: DIFF), details:
FAIL ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s_clm5 NLCOMP
FAIL ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s_clm5 BASELINE
ERP_Ln9.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq9s_clm5 (Overall: DIFF), details:
FAIL ERP_Ln9.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq9s_clm5 NLCOMP
FAIL ERP_Ln9.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq9s_clm5 BASELINE
SMS_Ld1.f09_f09.FW1850.yellowstone_intel.cam-outfrq1d (Overall: DIFF), details:
FAIL SMS_Ld1.f09_f09.FW1850.yellowstone_intel.cam-outfrq1d NLCOMP
FAIL SMS_Ld1.f09_f09.FW1850.yellowstone_intel.cam-outfrq1d BASELINE
SMS_Lm13.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq1m_clm5 (Overall: DIFF), details:
FAIL SMS_Lm13.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq1m_clm5 NLCOMP
FAIL SMS_Lm13.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq1m_clm5 BASELINE
-- namelists fail due to error with previous bless_test_results -- this failure
should not occur in future tags
-- answer changes due to changes in external components
ERP_Ln9.f10_f10.FSPCAMS.yellowstone_intel.cam-outfrq9s (Overall: DIFF), details:
FAIL ERP_Ln9.f10_f10.FSPCAMS.yellowstone_intel.cam-outfrq9s NLCOMP
FAIL ERP_Ln9.f10_f10.FSPCAMS.yellowstone_intel.cam-outfrq9s BASELINE ERROR BFAIL baseline directory '/glade/p/cesmdata/cseg/ccsm_baselines/cam5_4_105/ERP_Ln9.f10_f10.FSPCAMS.yellowstone_intel.cam-outfrq9s' does not exist
-- new test, no baseline for comparison
hobart/nag: all BFB
hobart/pgi: all BFB
===============================================================
===============================================================
Tag name: cam5_4_105
Originator(s): eaton
Date: Wed Jan 25 09:50:25 MST 2017
One-line Summary: refactor standalone regression testing
Purpose of changes:
. Recent and upcoming changes to the CLM and CICE components will require
significant changes to CAM's configure and build-namelist utilities in
order to continue using those components when running from CAM standalone
scripts. The support for this is not available. Instead we have chosen
to maintain the CAM standalone scripts only for the purpose of running
configurations that don't require the use of CLM or CICE, e.g.,
adiabatic, ideal physics, and aquaplanet. Therefore the part of CAM
regression testing which relies on the standalone build has been
refactored to only use adiabatic, ideal physics, and aquaplanet
configurations. F-compset configurations which require an active CLM
and/or CICE component are now tested using the cime test harness.
Bugs fixed (include bugzilla ID):
. fixed SLD failure in MPI mode (some compilers are getting more strict
about not aliasing input and output arguments in mpiallgatherv call)
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
components/cam/test/system/config_files/e48c4dh
components/cam/test/system/config_files/e48c5h
components/cam/test/system/config_files/e48c5m
components/cam/test/system/config_files/e64am
components/cam/test/system/config_files/e64c4bfbiop
components/cam/test/system/config_files/e64c5bfbiop
components/cam/test/system/config_files/e64c5m
components/cam/test/system/config_files/e64c5paqm
components/cam/test/system/config_files/e8c5dm
components/cam/test/system/config_files/e8c5paqdm
components/cam/test/system/config_files/e8c5t5mdm
components/cam/test/system/config_files/f0.9c4m
components/cam/test/system/config_files/f0.9c5.5mam4vbsh
components/cam/test/system/config_files/f1.9c4carmsuldh
components/cam/test/system/config_files/f1.9c4carmwtsuldh
components/cam/test/system/config_files/f1.9c4cm
components/cam/test/system/config_files/f1.9c4dm
components/cam/test/system/config_files/f1.9c4fastdh
components/cam/test/system/config_files/f1.9c4ghgdm
components/cam/test/system/config_files/f1.9c4m
components/cam/test/system/config_files/f1.9c4mozdh
components/cam/test/system/config_files/f1.9c4mozm
components/cam/test/system/config_files/f1.9c4wcarmbc_m
components/cam/test/system/config_files/f1.9c4wmxdh
components/cam/test/system/config_files/f1.9c4wmxh
components/cam/test/system/config_files/f1.9c4wscdh
components/cam/test/system/config_files/f1.9c5.4dh
components/cam/test/system/config_files/f1.9c5.4dm
components/cam/test/system/config_files/f1.9c5.5h
components/cam/test/system/config_files/f1.9c5.5wmam4dh
components/cam/test/system/config_files/f1.9c5.5wmam4h
components/cam/test/system/config_files/f1.9c5.5wscmam4dh
components/cam/test/system/config_files/f1.9c5.5wscmam4dm
components/cam/test/system/config_files/f1.9c5.5wtmam4h
components/cam/test/system/config_files/f1.9c5carmc_ddm
components/cam/test/system/config_files/f1.9c5carmcirm
components/cam/test/system/config_files/f1.9c5carmmetm
components/cam/test/system/config_files/f1.9c5carmpmcm
components/cam/test/system/config_files/f1.9c5carmseam
components/cam/test/system/config_files/f1.9c5cdm
components/cam/test/system/config_files/f1.9c5clbdh
components/cam/test/system/config_files/f1.9c5clbdm
components/cam/test/system/config_files/f1.9c5cm
components/cam/test/system/config_files/f1.9c5dm
components/cam/test/system/config_files/f1.9c5fstmam3dh
components/cam/test/system/config_files/f1.9c5m
components/cam/test/system/config_files/f1.9c5mam4dh
components/cam/test/system/config_files/f1.9c5mam4vbsdh
components/cam/test/system/config_files/f1.9c5mam7dh
components/cam/test/system/config_files/f1.9c5mg1_5dm
components/cam/test/system/config_files/f1.9c5mg2dm
components/cam/test/system/config_files/f1.9c5nonedh
components/cam/test/system/config_files/f10c3dm
components/cam/test/system/config_files/f10c4carmwtsuldm
components/cam/test/system/config_files/f10c4cdm
components/cam/test/system/config_files/f10c4wmdm
components/cam/test/system/config_files/f10c4wmxdm
components/cam/test/system/config_files/f10c4wscdm
components/cam/test/system/config_files/f10c4wtsmltdm
components/cam/test/system/config_files/f10c5.5wmam4dm
components/cam/test/system/config_files/f10c5.5wscmam4dm
components/cam/test/system/config_files/f10c5cdm
components/cam/test/system/config_files/f10c5dm
components/cam/test/system/config_files/f10c5mam4vbsdm
components/cam/test/system/config_files/f10c5nonedm
components/cam/test/system/config_files/f10c5paqdm
components/cam/test/system/config_files/f10c5spcammm
components/cam/test/system/config_files/f10c5spcamsm
components/cam/test/system/config_files/f10c5t5mdm
components/cam/test/system/config_files/f10c5t5mdmsc
components/cam/test/system/config_files/f10c5udm
components/cam/test/system/config_files/f10c5wmdiondm
components/cam/test/system/config_files/f10c5wtmam4dm
components/cam/test/system/config_files/f2.5c5dm
components/cam/test/system/config_files/f4c4dh
components/cam/test/system/config_files/f4c4fastdm
components/cam/test/system/config_files/f4c4rrtmgdm
components/cam/test/system/config_files/f4c4wmdh
components/cam/test/system/config_files/f4c4wmh
components/cam/test/system/config_files/f4c4wscdm
components/cam/test/system/config_files/f4c5carmm_idm
components/cam/test/system/config_files/f4c5carmm_sdm
components/cam/test/system/config_files/f4c5carmp_sdm
components/cam/test/system/config_files/f4c5carmthodm
components/cam/test/system/config_files/f4c5carmtt1dm
components/cam/test/system/config_files/f4c5carmtt2dm
components/cam/test/system/config_files/f4c5dm
components/cam/test/system/config_files/f4c5fstmam3dm
components/cam/test/system/config_files/f4c5mam4dm
components/cam/test/system/config_files/f4c5mam7dm
components/cam/test/system/config_files/f4c5paqdh
components/cam/test/system/config_files/fsd0.9c5.5wtsmlth
components/cam/test/system/config_files/fsd1.9c4wcarmsuldm
components/cam/test/system/config_files/fsd1.9c5.4wmdionh
components/cam/test/system/config_files/fsd1.9c5.4wtclbh
components/cam/test/system/config_files/fsd1.9c5clbdh
components/cam/test/system/config_files/fsd1.9c5clbdm
components/cam/test/system/config_files/h30c4wscdm
components/cam/test/system/config_files/h30c5h
components/cam/test/system/config_files/h5c4wmdm
components/cam/test/system/config_files/s32c5dh
components/cam/test/system/config_files/s32c5paqdh
components/cam/test/system/config_files/s64c5dh
components/cam/test/system/config_files/s64c5h
components/cam/test/system/config_files/s64c5m
components/cam/test/system/config_files/s64c5paqdh
components/cam/test/system/config_files/s8c5dm
components/cam/test/system/config_files/s8c5paqdm
components/cam/test/system/config_files/s8c5t5mdm
components/cam/test/system/config_files/scm64c4bfbiop
components/cam/test/system/config_files/scm64c5bfbiop
components/cam/test/system/config_files/scmc5armiop
components/cam/test/system/nl_files/no_ttrac
components/cam/test/system/nl_files/off1.9x2.5
components/cam/test/system/nl_files/outfrq3s_clubb
components/cam/test/system/nl_files/outfrq3s_macmic2_classnuc
components/cam/test/system/nl_files/outfrq3s_npryz
components/cam/test/system/tests_edison_intel
components/cam/test/system/tests_hopper_pgi
components/cam/test/system/tests_posttag_yellowstone
components/cam/test/system/tests_titan
components/cam/test/system/tests_titan_pgi
components/cam/test/system/tests_waccm_phys
. remove old tests
List all subroutines added and what they do:
components/cam/test/system/config_files/e64addh
components/cam/test/system/config_files/e64adh
components/cam/test/system/config_files/e64c4aqiopdm
components/cam/test/system/config_files/e64c5aqiopdm
components/cam/test/system/config_files/e64hsdh
components/cam/test/system/config_files/e8c3aqdm
components/cam/test/system/config_files/e8c4aqdm
components/cam/test/system/config_files/e8c5aqt5mdm
components/cam/test/system/config_files/f1.9c4aqh
components/cam/test/system/config_files/f1.9c4aqmozdh
components/cam/test/system/config_files/f1.9c4aqwmxdh
components/cam/test/system/config_files/f1.9c5aqm
components/cam/test/system/config_files/f1.9c6aqcdh
components/cam/test/system/config_files/f1.9c6aqtsvbsdh
components/cam/test/system/config_files/f1.9c6aqwscdh
components/cam/test/system/config_files/f10c4aqcdm
components/cam/test/system/config_files/f10c4aqwmxdm
components/cam/test/system/config_files/f10c4aqwscdm
components/cam/test/system/config_files/f10c5aqcdm
components/cam/test/system/config_files/f10c5aqcmtt1dm
components/cam/test/system/config_files/f10c5aqdm
components/cam/test/system/config_files/f10c5aqpbadm
components/cam/test/system/config_files/f10c5aqscdm
components/cam/test/system/config_files/f10c5aqt5mdm
components/cam/test/system/config_files/f10c5aqudm
components/cam/test/system/config_files/f10c6aqdm
components/cam/test/system/config_files/f10c6aqt5mdm
components/cam/test/system/config_files/f10c6aqwmadm
components/cam/test/system/config_files/f10spmaqdm
components/cam/test/system/config_files/f10spsaqdm
components/cam/test/system/config_files/f4c4aqdh
components/cam/test/system/config_files/h16adtermdh
components/cam/test/system/config_files/h16c5aqh
components/cam/test/system/config_files/h16c6aqdm
components/cam/test/system/config_files/h5c4aqwmdh
components/cam/test/system/config_files/h5c5aqbamdm
components/cam/test/system/config_files/h5c5aqdm
components/cam/test/system/config_files/h5c5aqt5mdm
components/cam/test/system/config_files/s8c5aqt5mdm
components/cam/test/system/config_files/scmc4aqds
components/cam/test/system/config_files/scmc5aqds
components/cam/test/system/nl_files/outfrq3s_diags
components/cam/test/system/nl_files/outfrq3s_f19c6aqwsc
components/cam/test/system/nl_files/ttrac_lb0
. for new aquaplanet regression tests
List all existing files that have been modified, and describe the changes:
components/cam/bld/namelist_files/namelist_defaults_cam.xml
. added new IC files for aquaplanet tests
components/cam/src/dynamics/sld/dyn_comp.F90
. add access to clat, clon for analytic ICs
components/cam/src/dynamics/sld/scanslt.F90
. fix illegal aliasing in mpiallgatherv call
components/cam/src/physics/camrt/radiation.F90
. remove undefined calculation when FSNR not output
components/cam/test/system/config_files/h16c5aqdm
. change '-ocn aquaplanet' to -aquaplanet
components/cam/test/system/input_tests_master
components/cam/test/system/nl_files/ghgrmp_e8
components/cam/test/system/nl_files/ghgrmp_unstruct
components/cam/test/system/nl_files/outfrq3s_carma
components/cam/test/system/nl_files/ttrac
components/cam/test/system/nl_files/ttrac_lb1
components/cam/test/system/nl_files/ttrac_lb2
components/cam/test/system/nl_files/ttrac_lb3
components/cam/test/system/test_driver.sh
components/cam/test/system/tests_cam_phys
components/cam/test/system/tests_carma
components/cam/test/system/tests_chem_hybrid
components/cam/test/system/tests_chem_mpi
components/cam/test/system/tests_pretag_hobart_nag
components/cam/test/system/tests_pretag_hobart_pgi
components/cam/test/system/tests_pretag_yellowstone
components/cam/test/system/tests_waccm_hybrid
components/cam/test/system/tests_waccm_mpi
. updates for new regression tests
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS except:
003 bl150 TBL.sh e64addh outfrq3s+dabi_p2004 9s ...............................................FAIL! rc= 7 at Tue Jan 24 11:32:59 MST 2017
006 bl151 TBL.sh e64hsdh outfrq3s+held_suarez_1994 9s .........................................FAIL! rc= 7 at Tue Jan 24 11:33:43 MST 2017
010 bl331 TBL.sh f4c4aqdh co2rmp+1850_cam4 9s .................................................FAIL! rc= 7 at Tue Jan 24 11:35:57 MST 2017
013 bl353 TBL.sh f1.9c4aqmozdh outfrq3s+2000_cam4_trop_chem 9s ................................FAIL! rc= 7 at Tue Jan 24 11:49:43 MST 2017
015 bl360 TBL.sh f1.9c5aqm volc+1850-2005_cam5 2d .............................................FAIL! rc= 7 at Tue Jan 24 11:52:10 MST 2017
017 bl370 TBL.sh f1.9c6aqtsvbsdh outfrq3s+1850_trop_strat_vbs_cam6 9s .........................FAIL! rc= 7 at Tue Jan 24 12:19:38 MST 2017
020 bl371 TBL.sh f1.9c6aqwscdh outfrq3s+waccm_sc_amip_cam6 9s .................................FAIL! rc= 4 at Tue Jan 24 12:31:44 MST 2017
023 bl380 TBL.sh f1.9c6aqcdh atrain 9s ........................................................FAIL! rc= 7 at Tue Jan 24 12:43:28 MST 2017
026 bl390 TBL.sh f1.9c4aqwmxdh outfrq3s+waccmx_2000_cam4 9s ...................................FAIL! rc= 7 at Tue Jan 24 13:26:22 MST 2017
029 bl391 TBL.sh f1.9c4aqwmxdh outfrq3s+waccmxi_2000_cam4 9s ..................................FAIL! rc= 7 at Tue Jan 24 14:08:17 MST 2017
032 bl735 TBL.sh h16c5aqdm outfrq3s+1850-2005_cam5 9s .........................................FAIL! rc= 7 at Tue Jan 24 14:23:08 MST 2017
035 bl736 TBL.sh h16adtermdh terminator 9s ....................................................FAIL! rc= 7 at Tue Jan 24 14:26:32 MST 2017
038 bl760 TBL.sh h5c4aqwmdh outfrq3s 9s .......................................................FAIL! rc= 7 at Tue Jan 24 14:44:29 MST 2017
044 eq995 TEQ_ccsm.sh T42_T42 FDABIP04 e64adh fcase+dabi_p2004 9s .............................FAIL! rc= 7 at Tue Jan 24 14:50:43 MST 2017
eq995 is failing due to bug in
cime/cime_config/cesm/machines/config_compilers.xml which defines NO_R16
for intel. Intel does support quad precision. Jim will remove in future
cime tag.
yellowstone/intel/aux_cam: All 5 tests PASS baselines. Namelist comparison
still fails.
hobart/nag: All PASS except:
005 bl111 TBL.sh e8c5aqt5mdm ghgrmp_e8 9s .....................................................FAIL! rc= 7 at Tue Jan 24 09:45:54 MST 2017
010 bl113 TBL.sh e8c4aqdm outfrq3s+aquaplanet_cam4 9s .........................................FAIL! rc= 7 at Tue Jan 24 09:48:14 MST 2017
013 bl221 TBL.sh f10spsaqdm outfrq3s 9s .......................................................FAIL! rc= 7 at Tue Jan 24 09:57:41 MST 2017
017 bl311 TBL.sh f10c5aqt5mdm ttrac 9s ........................................................FAIL! rc= 7 at Tue Jan 24 10:02:54 MST 2017
024 bl318 TBL.sh f10c4aqcdm sat_hist 9s .......................................................FAIL! rc= 7 at Tue Jan 24 10:12:14 MST 2017
027 bl325 TBL.sh f10c5aqscdm outfrq3s_subcol 9s ...............................................FAIL! rc= 7 at Tue Jan 24 10:16:01 MST 2017
033 bl338 TBL.sh f10c5aqdm rad_diag_mam 9s ....................................................FAIL! rc= 7 at Tue Jan 24 10:19:49 MST 2017
036 bl368 TBL.sh f10c6aqdm outfrq3s 9s ........................................................FAIL! rc= 7 at Tue Jan 24 10:24:47 MST 2017
039 bl426 TBL.sh f10c4aqwmxdm outfrq3s_ionos 9s ...............................................FAIL! rc= 7 at Tue Jan 24 10:31:39 MST 2017
042 bl430 TBL.sh f10c5aqcmtt1dm outfrq3s_carma 9s .............................................FAIL! rc= 7 at Tue Jan 24 10:40:13 MST 2017
045 bl471 TBL.sh f10c6aqwmadm outfrq3s 9s .....................................................FAIL! rc= 7 at Tue Jan 24 10:49:19 MST 2017
049 bl711 TBL.sh h5c5aqt5mdm ttrac 9s .........................................................FAIL! rc= 7 at Tue Jan 24 10:56:40 MST 2017
056 bl741 TBL.sh h16c6aqdm ghgrmp_unstruct 9s .................................................FAIL! rc= 7 at Tue Jan 24 11:32:12 MST 2017
hobart/pgi: All PASS except:
004 bl112 TBL.sh e8c3aqdm outfrq3s+aquaplanet_cam3 9s .........................................FAIL! rc= 7 at Tue Jan 24 09:46:00 MST 2017
008 bl114 TBL.sh e8c4aqdm co2rmp 9s ...........................................................FAIL! rc= 7 at Tue Jan 24 09:51:37 MST 2017
012 bl222 TBL.sh f10spmaqdm outfrq3s 9s .......................................................FAIL! rc= 7 at Tue Jan 24 10:49:31 MST 2017
015 bl314 TBL.sh f10c4aqwscdm outfrq3s_diags 9s ...............................................FAIL! rc= 7 at Tue Jan 24 10:56:46 MST 2017
018 bl317 TBL.sh f10c5aqcdm outfrq3s_diags 9s .................................................FAIL! rc= 7 at Tue Jan 24 11:07:12 MST 2017
021 bl320 TBL.sh f10c5aqpbadm rad_diag 9s .....................................................FAIL! rc= 7 at Tue Jan 24 11:13:49 MST 2017
024 bl321 TBL.sh f10c5aqcdm atrain 9s .........................................................FAIL! rc= 7 at Tue Jan 24 11:16:07 MST 2017
027 bl334 TBL.sh f10c5aqudm outfrq3s 9s .......................................................FAIL! rc= 7 at Tue Jan 24 11:24:23 MST 2017
030 bl339 TBL.sh f10adhterm terminator 9s .....................................................FAIL! rc= 7 at Tue Jan 24 11:29:02 MST 2017
033 bl511 TBL.sh s8c5aqt5mdm ttrac 9s .........................................................FAIL! rc= 7 at Tue Jan 24 11:36:14 MST 2017
037 bl712 TBL.sh h5c5aqbamdm outfrq3s_bam+aquaplanet_cam5 9s ..................................FAIL! rc= 7 at Tue Jan 24 11:43:02 MST 2017
All failing baseline comparisons due to new tests with no available
baseline.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_4_104
Originator(s): fvitt, mmills, bardeenc
Date: 23 Jan 2017
One-line Summary: Updates to chemical tropopause, add SF6 to WACCM6, and updates to compsets
Purpose of changes:
Updates to tropopause detection logic used in chemistry, clouds, and CLUBB and
to settings for stratospheric ice formation
Modify the chemical tropopause definition poleward of 50 degrees from using a
fixed 300 hPa level (see note for 5_4_89) to use the lapse rate tropopause at
P > 125 hPa and the climatology if the lapse rate tropopause is found at
P < 125 hPa or is not found. This eliminates “false” tropopause detection in the
75 hPa < P < 125 hPa range that are not physical and have a detrimental effect on
heterogeneous chemistry in the polar stratosphere, provides a diagnostic that can
respond to some climate change, and gives a smoother and more physical tropopause
detection across latitude and altitude ranges that improves some diagnostics.
Change settings for namelist variables that control the use of the tropopause
detection logic (see notes for cam5_4_90 and cam5_4_91) used in heterogeneous
chemistry, stratospheric aerosols, stratospheric clouds, and CLUBB stratospheric
adjustments to all use this new definition. Previous settings were used to
retain bit-for-bit backward compatibility, but these settings represent better
physics. Namelist variables affected are: prescribed_strataero_use_chemtrop,
chem_use_chemtrop, and nucleate_ice_use_troplev.
Change settings related to clouds and nucleation in the stratosphere to increase
nucleation and decrease dehydration. These include: nucleate_ice_strat, and
cldfrc2m_rhminis.
Add SF6 to WACCM6
Updates to WACCM6 and CAM-Chem build-namelist use case files
Add FW1850 compset
Add WACCM to CIME aux_cam test list (FWAMIP and FW1850)
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/bld/build-namelist
- add SF6 to flbc_list
- set uwshcu_rpen only when pbl scheme is 'uw'
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
- impose these defaults :
nucleate_ice_strat = 1.0
nucleate_ice_use_troplev = .true.
chem_use_chemtrop = .true.
prescribed_strataero_use_chemtrop = .true.
cldfrc2m_rhminis = 1.0
M components/cam/bld/namelist_files/use_cases/1850_trop_strat_vbs_cam6.xml
M components/cam/bld/namelist_files/use_cases/2000_trop_strat_vbs_cam6.xml
- remove the full path to ic files
M components/cam/bld/namelist_files/use_cases/sd_waccm_tsmlt_cam6.xml
M components/cam/bld/namelist_files/use_cases/waccm_ma_1850_cam6.xml
M components/cam/bld/namelist_files/use_cases/waccm_ma_2000_cam6.xml
M components/cam/bld/namelist_files/use_cases/waccm_ma_amip_cam6.xml
M components/cam/bld/namelist_files/use_cases/waccm_sc_amip_cam6.xml
M components/cam/bld/namelist_files/use_cases/waccm_tsmlt_1850_cam6.xml
M components/cam/bld/namelist_files/use_cases/waccm_tsmlt_2000_cam6.xml
M components/cam/bld/namelist_files/use_cases/waccm_tsmlt_amip_cam6.xml
- updates to build-namelist use cases
M components/cam/cime_config/testdefs/testlist_cam.xml
- added ERP_Ld3 FWAMIP test to aux_cam tests
- other adjustments to aux_wcm tests
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/chem_mech.doc
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/chem_mech.in
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/chem_mods.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/m_rxt_id.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/m_spc_id.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_adjrxt.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_exp_sol.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_imp_sol.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_indprd.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_lin_matrix.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_lu_factor.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_lu_solve.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_nln_matrix.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_phtadj.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_prod_loss.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_rxt_rates_conv.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_setrxt.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_sim_dat.F90
- added SF6 to this mechanism
M components/cam/src/physics/cam/clubb_intr.F90
- correction to the KVH_CLUBB diagnostic
M components/cam/src/physics/cam/tropopause.F90
- more realistic chemical tropopuase over the poles (see above)
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
008 bl222 TBL.sh f10c5spcammm outfrq3s 9s .....................................................FAIL! rc= 7 at Mon Jan 23 14:01:07 MST 2017
019 bl357 TBL.sh f1.9c5.5h outfrq3s+1850_cam55 9s .............................................FAIL! rc= 7 at Mon Jan 23 14:01:26 MST 2017
022 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Mon Jan 23 14:01:42 MST 2017
026 bl373 TBL.sh f1.9c4mozdh outfrq3s+2000_cam4_trop_chem 9s ..................................FAIL! rc= 7 at Mon Jan 23 14:09:23 MST 2017
028 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...............................................FAIL! rc= 7 at Mon Jan 23 14:12:10 MST 2017
031 bl456 TBL.sh f1.9c5.5wtmam4h outfrq3s_megan+waccm_tsmlt_2000_cam6 9s ......................FAIL! rc= 7 at Mon Jan 23 14:32:18 MST 2017
034 bl390 TBL.sh f1.9c4wmxh outfrq3s+waccmx_2000_cam4 9s ......................................FAIL! rc= 7 at Mon Jan 23 14:42:36 MST 2017
040 bl421 TBL.sh f1.9c5mam4dh outfrq3s 9s .....................................................FAIL! rc= 7 at Mon Jan 23 14:56:40 MST 2017
043 bl429 TBL.sh f1.9c4wmxh outfrq3s+waccmxi_2000_cam4 9s .....................................FAIL! rc= 7 at Mon Jan 23 15:05:49 MST 2017
045 bl492 TBL.sh f0.9c5.5mam4vbsh outfrq3s+1850_trop_strat_vbs_cam6 9s ........................FAIL! rc= 7 at Mon Jan 23 15:23:58 MST 2017
048 bl470 TBL.sh fsd1.9c5.4wtclbh outfrq3s_sd+sd_waccm5_geos5 9s ..............................FAIL! rc= 7 at Mon Jan 23 15:52:14 MST 2017
051 bl482 TBL.sh f1.9c5.5wscmam4dh outfrq3s+waccm_sc_amip_cam6 9s .............................FAIL! rc= 7 at Mon Jan 23 16:28:34 MST 2017
069 bl740 TBL.sh h30c5h outfrq3s_lb0 9s .......................................................FAIL! rc= 7 at Mon Jan 23 16:54:37 MST 2017
yellowstone/intel/aux_cam:
ERP_Ld3.f09_f09.FWAMIP.yellowstone_intel.cam-outfrq9s (Overall: DIFF), details:
FAIL ERP_Ld3.f09_f09.FWAMIP.yellowstone_intel.cam-outfrq9s NLCOMP
FAIL ERP_Ld3.f09_f09.FWAMIP.yellowstone_intel.cam-outfrq9s BASELINE ERROR BFAIL baseline directory '/glade/p/cesmdata/cseg/ccsm_baselines/cam5_4_103/ERP_Ld3.f09_f09.FWAMIP.yellowstone_intel.cam-outfrq9s' does not exist
ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s_clm5 (Overall: DIFF), details:
FAIL ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s_clm5 NLCOMP
FAIL ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s_clm5 BASELINE
ERP_Ln9.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq9s_clm5 (Overall: DIFF), details:
FAIL ERP_Ln9.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq9s_clm5 NLCOMP
FAIL ERP_Ln9.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq9s_clm5 BASELINE
SMS_Ld1.f09_f09.FW1850.yellowstone_intel.cam-outfrq1d (Overall: DIFF), details:
FAIL SMS_Ld1.f09_f09.FW1850.yellowstone_intel.cam-outfrq1d NLCOMP
FAIL SMS_Ld1.f09_f09.FW1850.yellowstone_intel.cam-outfrq1d BASELINE ERROR BFAIL baseline directory '/glade/p/cesmdata/cseg/ccsm_baselines/cam5_4_103/SMS_Ld1.f09_f09.FW1850.yellowstone_intel.cam-outfrq1d' does not exist
SMS_Lm13.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq1m_clm5 (Overall: DIFF), details:
FAIL SMS_Lm13.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq1m_clm5 NLCOMP
FAIL SMS_Lm13.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq1m_clm5 BASELINE
hobart/nag:
058 bl356 TBL.sh fsd1.9c5clbdm off1.9x2.5 9s ..................................................FAIL! rc= 7 at Mon Jan 23 11:39:23 MST 2017
061 bl368 TBL.sh f1.9c5mg2dm outfrq3s_macmic2_classnuc 9s .....................................FAIL! rc= 7 at Mon Jan 23 11:52:32 MST 2017
064 bl369 TBL.sh f1.9c5clbdm outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Mon Jan 23 12:10:36 MST 2017
073 bl426 TBL.sh f10c4wmxdm outfrq3s_ionos 9s .................................................FAIL! rc= 7 at Mon Jan 23 12:35:01 MST 2017
076 bl471 TBL.sh f10c5.5wmam4dm outfrq3s_modalstrat 9s ........................................FAIL! rc= 7 at Mon Jan 23 12:48:02 MST 2017
hobart/pgi: All pass
The baseline failures are expected due to changes in chemistry tropopause and nucleate_ice namelist settings
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: all chemistry configurations including cam5 and cam6
- what platforms/compilers: all
- nature of change : larger than roundoff but same climate
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
https://svn-ccsm-models.cgd.ucar.edu/cam1/branch_tags/tropopause_cam5_4_97_tags/tropopause02_cam5_4_99
- platform/compilers:
yellowstone/intel
- configure commandline:
create_newcase -compset F2000_DEV -res f09_f09 -project CESM0005 -mach yellowstone -case /glade/p/work/mmills/case/f.e20.F2000_DEV.f09_f09.tropopause02_cam5_4_99.02
- MSS location of output:
/home/mmills/csm/f.e20.F2000_DEV.f09_f09.tropopause02_cam5_4_99.02
MSS location of control simulations used to validate new climate:
/home/mmills/csm/f.e20.F2000_DEV.f09_f09.cam5_4_99.01
URL for AMWG diagnostics output used to validate new climate:
https://acomstaff.acom.ucar.edu/mmills/ccsm-diagnostics/CAM6_170117/tropopause02_cam5_4_99.02-cam5_4_99.01_y2-10/
===============================================================
===============================================================
Tag name: cam5_4_103
Originator(s): goldy
Date: 2017-01-22
One-line Summary: CAM history bug fixes and Cray directive removal
Purpose of changes: Fix incorrect coordinate output when using column (patch)
output with unstructured grids.
Fix bug with point processing near prime meridian.
Also, take out old, unused Cray directives which were causing trouble.
Bugs fixed (include bugzilla ID): NA
Describe any changes made to build system: NA
Describe any changes made to the namelist: NA
List any changes to the defaults for the boundary datasets: NA
Describe any substantial timing or memory changes: NA
Code reviewed by: goldy
List all subroutines eliminated: NA
List all subroutines added and what they do: NA
List all existing files that have been modified, and describe the changes:
M components/cam/doc/ChangeLog
- Brought this file over 100 lines closer to the magic 10^5 mark
M components/cam/src/control/cam_history.F90
- Fixed problems with column output and improved log messages.
- Corrected point coordinate processing near prime meridian.
M components/cam/src/control/cam_history_support.F90
M components/cam/src/utils/cam_grid_support.F90
M components/cam/src/utils/cam_map_utils.F90
- Fixed problem with column output for unstructured grids.
M components/cam/src/dynamics/eul/dyn.F90
M components/cam/src/dynamics/eul/grcalc.F90
M components/cam/src/dynamics/eul/hordif.F90
M components/cam/src/dynamics/eul/linemsdyn.F90
M components/cam/src/dynamics/eul/spegrd.F90
M components/cam/src/dynamics/fv/dp_coupling.F90
M components/cam/src/dynamics/fv/mapz_module.F90
M components/cam/src/dynamics/fv/tp_core.F90
M components/cam/src/dynamics/sld/dp_coupling.F90
M components/cam/src/dynamics/sld/realloc4.F90
M components/cam/src/dynamics/sld/realloc7.F90
M components/cam/src/physics/cam/check_energy.F90
M components/cam/src/physics/cam/cldwat.F90
M components/cam/src/physics/cam/hk_conv.F90
M components/cam/src/physics/cam/modal_aer_opt.F90
M components/cam/src/physics/cam/phys_grid.F90
M components/cam/src/physics/cam/qneg3.F90
M components/cam/src/physics/cam/zm_conv.F90
M components/cam/src/physics/camrt/radlw.F90
M components/cam/src/physics/camrt/radsw.F90
M components/cam/src/physics/rrtmg/aer_src/rrtmg_lw_rtrnmc.f90
M components/cam/src/physics/rrtmg/aer_src/rrtmg_lw_taumol.f90
M components/cam/src/physics/rrtmg/aer_src/rrtmg_sw_cldprmc.f90
M components/cam/src/physics/rrtmg/aer_src/rrtmg_sw_rad.f90
M components/cam/src/physics/rrtmg/aer_src/rrtmg_sw_reftra.f90
M components/cam/src/physics/rrtmg/aer_src/rrtmg_sw_spcvmc.f90
M components/cam/src/physics/rrtmg/aer_src/rrtmg_sw_vrtqdr.f90
M components/cam/src/utils/fft99.F90
M components/cam/src/utils/spmd_utils.F90
M components/cam/src/physics/icarus-scops/icarus.f90
- Took out old, unused Cray performance directives
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: Expected fail due to bad coordinate variables in baseline
004 bl132 TBL.sh e48c4dh ghgrmp_e8+1850_cam4 9s ...............................................FAIL! rc= 7 at Sat Jan 21 18:19:29 MST 2017
yellowstone/intel/aux_cam: Expected fail due to bug in bless_test_results
FAIL ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s_clm5 NLCOMP
FAIL ERP_Ln9.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq9s_clm5 NLCOMP
FAIL SMS_Lm13.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq1m_clm5 NLCOMP
hobart/nag: Expected fail due to bad coordinate variables in baseline
010 bl112 TBL.sh e8c5dm ghgrmp_e8 9s ..........................................................FAIL! rc= 7 at Sat Jan 21 11:18:10 MST 2017
hobart/pgi: Expected fail due to bad coordinate variables in baseline
010 bl112 TBL.sh e8c5dm ghgrmp_e8 9s ..........................................................FAIL! rc= 7 at Sat Jan 21 11:53:02 MST 2017
CAM tag used for the baseline comparison tests if different than previous
tag: NA
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate): NA (answer changes were in column output processing)
If bitwise differences were observed, how did you show they were no worse
than roundoff? NA
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository): NA
===============================================================
===============================================================
Tag name: cam5_4_102
Originator(s): cacraig
Date: Jan 20, 2017
One-line Summary: Update CLM external to match cesm2_0_beta05
Purpose of changes:
- Update the CLM external to match the version which is used
in cesm2_0_beta05. This version is needed for an upcoming CAM tag.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M .
M SVN_EXTERNAL_DIRECTORIES
- update CLM5 version
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB
yellowstone/intel/aux_cam:
ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s_clm5 (Overall: DIFF), details:
FAIL ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s_clm5 NLCOMP
FAIL ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s_clm5 BASELINE
ERP_Ln9.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq9s_clm5 (Overall: DIFF), details:
FAIL ERP_Ln9.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq9s_clm5 NLCOMP
FAIL ERP_Ln9.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq9s_clm5 BASELINE
SMS_Lm13.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq1m_clm5 (Overall: DIFF), details:
FAIL SMS_Lm13.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq1m_clm5 NLCOMP
FAIL SMS_Lm13.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq1m_clm5 BASELINE
- Namelists will continue to fail until update cime external
- Update of CLM external changed answers for all these runs which use CLM5
hobart/nag: all BFB
hobart/pgi: all BFB
-- Answer changes for all jobs which use CLM4.5 or CLM5 due to answer changes in
the CLM tag
Summarize any changes to answers, i.e.,
- what code configurations: all which use CLM4.5 or CLM5
- what platforms/compilers: all
===============================================================
===============================================================
Tag name: cam5_4_101
Originator(s): hannay, cacraig
Date: Jan 17, 2017
One-line Summary: Match Cecile's #125 configuration (with caveat)
Purpose of changes:
- This tag matches Cecile's #125 configuration for the atmosphere. The
land is different between 125 and this tag. We verified run 125 was
identical to this run with a two month test case (see the detais below
in the testing section)
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
Changed the namelist default settings for CAM6 (and 5.5 and 5.4):
clubb_do_liqsupersat = .false.
micro_mg_berg_eff_factor = 0.7
clubb_gamma_coef = 0.32
micro_mg_dcs = 540.D-6
clubb_c14 = 2.2D0
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
- Changed the default namelist settings as described above
- Removed a couple of settings which were set for MG2 without CLUBB
as these are now obsolete.
M components/cam/bld/namelist_files/use_cases/1850_cam6.xml
M components/cam/bld/namelist_files/use_cases/2000_cam6.xml
M components/cam/bld/namelist_files/use_cases/hist_cam6.xml
M components/cam/bld/namelist_files/use_cases/sd_waccm_tsmlt_cam6.xml
M components/cam/bld/namelist_files/use_cases/waccm_ma_1850_cam6.xml
M components/cam/bld/namelist_files/use_cases/waccm_ma_2000_cam6.xml
M components/cam/bld/namelist_files/use_cases/waccm_ma_amip_cam6.xml
M components/cam/bld/namelist_files/use_cases/waccm_sc_amip_cam6.xml
M components/cam/bld/namelist_files/use_cases/waccm_tsmlt_1850_cam6.xml
M components/cam/bld/namelist_files/use_cases/waccm_tsmlt_2000_cam6.xml
M components/cam/bld/namelist_files/use_cases/waccm_tsmlt_amip_cam6.xml
- add missing resolution qualifier to bnd_topo file
- remove clubb_do_liqsupersat and clubb_gamma_coef as they are now
set in the defaults
M components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq1m_clm5/user_nl_cam
- changed the output precision from double to single to match the resolution
of Cecile's runs to ease comparisons
M components/cam/src/physics/cam/micro_mg_utils.F90
- Process rate enhancement for relative variance is removed from the Siefert
and Behang 2001 (SB2001) autoconversion scheme.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB except cam5.4, cam5.5 tests due to changes in this tag
015 bl353 TBL.sh f1.9c5.4dh outfrq3s 9s .......................................................FAIL! rc= 7 at Sat Jan 14 09:29:19 MST 2017
019 bl357 TBL.sh f1.9c5.5h outfrq3s+1850_cam55 9s .............................................FAIL! rc= 7 at Sat Jan 14 09:34:47 MST 2017
022 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Sat Jan 14 09:47:02 MST 2017
031 bl456 TBL.sh f1.9c5.5wtmam4h outfrq3s_megan+waccm_tsmlt_2000_cam6 9s ......................FAIL! rc= 7 at Sat Jan 14 10:27:52 MST 2017
045 bl492 TBL.sh f0.9c5.5mam4vbsh outfrq3s+1850_trop_strat_vbs_cam6 9s ........................FAIL! rc= 7 at Sat Jan 14 11:18:05 MST 2017
048 bl470 TBL.sh fsd1.9c5.4wtclbh outfrq3s_sd+sd_waccm5_geos5 9s ..............................FAIL! rc= 7 at Sat Jan 14 11:47:12 MST 2017
051 bl482 TBL.sh f1.9c5.5wscmam4dh outfrq3s+waccm_sc_amip_cam6 9s .............................FAIL! rc= 7 at Sat Jan 14 12:22:32 MST 2017
yellowstone/intel/aux_cam:
ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s_clm5 (Overall: DIFF), details:
FAIL ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s_clm5 NLCOMP
FAIL ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s_clm5 BASELINE
ERP_Ln9.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq9s_clm5 (Overall: DIFF), details:
FAIL ERP_Ln9.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq9s_clm5 NLCOMP
FAIL ERP_Ln9.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq9s_clm5 BASELINE
SMS_Lm13.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq1m_clm5 (Overall: DIFF), details:
FAIL SMS_Lm13.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq1m_clm5 NLCOMP
FAIL SMS_Lm13.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq1m_clm5 BASELINE
- namelist failures are due to the namelist files still not be put into the baselines
via bless_test_results
- Baseline failures are due to changes in this tag
hobart/nag: all BFB except cam5.4, cam5.5 tests due to changes in this tag
055 bl354 TBL.sh f1.9c5.4dm outfrq3s_npryz 9s .................................................FAIL! rc= 7 at Fri Jan 13 18:32:31 MST 2017
058 bl356 TBL.sh fsd1.9c5clbdm off1.9x2.5 9s ..................................................FAIL! rc= 7 at Fri Jan 13 18:56:17 MST 2017
061 bl368 TBL.sh f1.9c5mg2dm outfrq3s_macmic2_classnuc 9s .....................................FAIL! rc= 7 at Fri Jan 13 19:10:11 MST 2017
064 bl369 TBL.sh f1.9c5clbdm outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Fri Jan 13 19:27:11 MST 2017
076 bl471 TBL.sh f10c5.5wmam4dm outfrq3s_modalstrat 9s ........................................FAIL! rc= 7 at Fri Jan 13 19:58:06 MST 2017
hobart/pgi: all BFB except cam5.4, cam5.5 tests due to changes in this tag
057 bl354 TBL.sh f1.9c5.4dm outfrq3s_npryz 9s .................................................FAIL! rc= 7 at Fri Jan 13 19:35:33 MST 2017
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: all CAM5.4, CAM5.5, CAM6.0
- what platforms/compilers: all
- nature of change (roundoff; larger than roundoff but same climate; new
climate): New climate
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository): cam5_4_101
- platform/compilers: yellowstone/intel
- CESM commandline: .//create_newcase --case /glade/scratch/cacraig/aux_cam_20170113152136/SMS_Lm13.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq1m_clm5.C.aux_cam_20170113152136 --res f09_f09 --mach yellowstone --compiler intel --compset F2000_DEV --test --project P93300606 --user-mods-dir /glade/u/home/cacraig/cam5_4_101-TRUNK/cime/../components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq1m_clm5 --queue=regular --walltime 3:00
- yellowstone location of output from 13 month run which compared with Cecile's 2 month run:
/glade/p/cesmdata/cseg/ccsm_baselines/cam5_4_101/SMS_Lm13.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq1m_clm5
URL for AMWG diagnostics output used to validate new climate:
Diags for F2000_DEV:
http://webext.cgd.ucar.edu/FCLIMO/f.e20.F2000_DEV.f09_f09.CLUBB_reorder_n05_cam5_4_96/atm/
Diags for 125 (for reference with the caveat above):
http://webext.cgd.ucar.edu/B1850/b.e20.B1850.f09_g16.pi_control.all.125/
===============================================================
===============================================================
Tag name: cam5_4_100
Originator(s): cacraig, Minghaui Wang, Mark Branson, fvitt
Date: Jan 12, 2017
One-line Summary: SPCAM: Bug fix for sam1mom and allow running of spcam_m2005
Purpose of changes:
- Super-parameterized CAM (SPCAM) now has the ability to run with the
double moment SAM microphysics (spcam_m2005) in addition to the already
released single moment SAM (spcam_sam1mom)
- Using the flags: "-nan -v -O0" in the NAG compiler, tracked down all
uninitialized values and fixed them either by initializing
the values, putting assignments within if blocks or limiting array bounds
when passing into subroutines
- Added FSPCAMM compset. Note that this compset along with FSPCAMS require
a fix in cime to override the threading setting of 2. These two compsets
do not work in this tag out-of-the-box. A fix for this is coming shortly.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
- Now allow super-parameterized CAM double moment SAM microphysics to
be run (spcam_m2005)
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
D components/cam/src/physics/spcam/crm/crmx_init.F90
- This routine was never called
List all subroutines added and what they do:
A + components/cam/test/system/config_files/f10c5spcammm
- add spcam_m2005 configuration for CAM regression testing
List all existing files that have been modified, and describe the changes:
M components/cam/bld/configure
- Add a check to prevent turning on threading if using SPCAM as
the underlying CRM model is not threadsafe
- Remove the block on configuring spcam_m2005
M components/cam/cime_config/config_component.xml
M components/cam/cime_config/config_compsets.xml
- introduce a compset for spcam_m2005
M components/cam/src/chemistry/modal_aero/aero_model.F90
- initialize reff_strat as the variable is intent(out) and
does not keep the initialization which occurred outside this routine
M components/cam/src/physics/cam/spcam_drivers.F90
- Bug fix from Mark Branson: Add in a call to sslt_rebin_adv to solve
problems with SSLTA and SSLTC
- Added initializations and if logic around assignments to eliminate
uninitialized variable issues
M components/cam/src/physics/spcam/crm/crmx_crm_module.F90
M components/cam/src/physics/spcam/crm/crmx_kurant.F90
M components/cam/src/physics/spcam/crm/crmx_module_ecpp_stats.F90
M components/cam/src/physics/spcam/crm/crmx_params.F90
M components/cam/src/physics/spcam/crmclouds_camaerosols.F90
- initialized variables which were used before they were set
M components/cam/src/physics/spcam/crm/crmx_diagnose.F90
- removed a line which was not initialized before it was used, and
the computed value was not subsequently used (dead code)
M components/cam/src/physics/spcam/crm_physics.F90
- removed pbuf fields, prer_evap, relvar and accre_enhan which are
not needed by SPCAM
- initialize pbuf fields which are used by subsequent parameterizations
- limit arrays to only the sections which have been set
M components/cam/src/physics/spcam/ecpp/ecpp_modal_cloudchem.F90
- bug fix from Minghuai Wang (from his colleage Guangxin Lin)
M components/cam/src/physics/spcam/ecpp/module_ecpp_td2clm.F90
- removed dead code which was having issues with uninitialized values
M components/cam/test/system/input_tests_master
M components/cam/test/system/tests_pretag_yellowstone
- add spcam_m2005 test to CAM regression testing
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB except:
008 bl222 TBL.sh f10c5spcammm outfrq3s 9s .....................................................FAIL! rc= 7 at Mon Jan 9 22:02:39 MST 2017
No baseline for comparison as this is a new test
yellowstone/intel/aux_cam: all PASS except namelist failure due to missing archived namelists
hobart/nag: all BFB except:
017 bl221 TBL.sh f10c5spcamsm outfrq3s 9s .....................................................FAIL! rc= 7 at Mon Jan 9 14:57:26 MST 2017
- The sam1mom version of SPCAM has answer changes due to the introduction of the call to sslt_rebin_adv (bug fix)
hobart/pgi: all BFB
===============================================================
===============================================================
Tag name: cam5_4_99
Originator(s): mvertens, goldy
Date: 2017-01-01
One-line Summary: Update CAM'S configure to be called from the new script, cime_config/buildcpp
Purpose of changes: CAM'S configure is now called from a new file,
cime_config/buildcpp. buildcpp is primarily now called from
cime_config/buildlib - but can also be called from cime_config/buildnml
for cases where namelist generation needs to occur but a
config_cache.xml file is either not available or is outdated.
- Removed references to CCSM_CO2_PPMV in user_nl_clm testmods files
- Also, fixed FV restart bug which affects PIO2 runs (2395).
Bugs fixed (include bugzilla ID): 2395
Describe any changes made to build system: configure infrastructure updated
Describe any changes made to the namelist: NA
List any changes to the defaults for the boundary datasets: NA
Describe any substantial timing or memory changes: NA
Code reviewed by: mvertens, goldy
List all subroutines eliminated: NA
List all subroutines added and what they do: NA
List all existing files that have been modified, and describe the changes:
A + components/cam/cime_config/buildcpp
- New CIME script to handle production of CPP variables from case configuration
M components/cam/cime_config/buildlib
M components/cam/cime_config/buildnml
- Added and modified code to be forwards compatible with upcoming CIME
namelist building changes
M components/cam/cime_config/config_component.xml
- Added entry for CAM_CPPDEFS needed by new configure utility script (buildcpp)
M components/cam/cime_config/testdefs/testmods_dirs/cam/fire_emis/user_nl_clm
M components/cam/cime_config/testdefs/testmods_dirs/cam/mam4_fire_emis/user_nl_clm
M components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq1d/user_nl_clm
M components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq1m_clm5/user_nl_clm
M components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq3d/user_nl_clm
M components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq9s/user_nl_clm
M components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq9s_clm5/user_nl_clm
M components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq9s_sat_hist/user_nl_clm
M components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq9s_wse/user_nl_clm
- Removed references to deprecated variable, CCSM_CO2_PPMV
M components/cam/doc/ChangeLog_template
- Removed jaguar as a test machine
M components/cam/src/dynamics/fv/restart_dynamics.F90
- Fixed location of pio_setframe calls which did not work with PIO2 (bugzilla
2395)
M components/cam/src/dynamics/se/dyn_comp.F90
- Slight code documentation
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: ALL PASS
yellowstone/intel/aux_cam: All PASS (note that namelist comparison still
broken)
hobart/nag: ALL PASS
hobart/pgi: ALL PASS
CAM tag used for the baseline comparison tests if different than previous
tag: NA
Summarize any changes to answers: NA
If bitwise differences were observed, how did you show they were no worse
than roundoff? NA
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository): NA
MSS location of control simulations used to validate new climate: NA
URL for AMWG diagnostics output used to validate new climate: NA
===============================================================
===============================================================
Tag name: cam5_4_98
Originator(s): eaton
Date: Tue Dec 27 11:35:16 MST 2016
One-line Summary: Remove dependence on CCSM_CO2_PPMV var in CESM scripts;
fix in hist_cam6 use case.
Purpose of changes:
. Remove the setting of CAM's co2vmr variable by cam/cime_config/buildnml.
This eliminates CAM's dependence on the CCSM_CO2_PPMV var in CESM scripts.
CAM's co2vmr namelist variable is only used when flbc_file is not
specified, or scenario_ghg is not set to RAMPED. flbc_file is used by
all CAM-Chem and WACCM configurations, and scenario_ghg is set to RAMPED
in all historical runs that are not CAM-Chem or WACCM. So basically the
only use of co2vmr is for climatology runs of that standard cam physics
packages, or for scam runs with which are short duration and use fixed
co2.
It turns out this dependency was redunant since the few cases where
co2vmr is used already had values that were consistent with the
CCSM_CO2_PPMV values set in use case files for in the namelist defaults
file.
. bug fix in hist_cam6 use case file
Bugs fixed (include bugzilla ID):
. use case file hist_cam6.xml needed updated srf_emis_specifier files to
allow starting on 1850-01-01
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
components/cam/bld/namelist_files/use_cases/2000_cam6.xml
components/cam/bld/namelist_files/use_cases/2000_cam4_trop_chem.xml
. add 367.0e-6
components/cam/bld/namelist_files/use_cases/1850_trop_strat_vbs_cam6.xml
components/cam/bld/namelist_files/use_cases/2000_trop_strat_vbs_cam6.xml
. add CYCLICAL
components/cam/bld/namelist_files/use_cases/scam_arm95.xml
. add 368.9e-6
components/cam/cime_config/buildnml
. remove setting co2vmr
components/cam/bld/namelist_files/use_cases/hist_cam6.xml
. updated srf_emis_specifier files for DMS and SO2 to
allow starting on 1850-01-01
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS
yellowstone/intel/aux_cam: All PASS (note that namelist comparison still
broken)
hobart/nag: All PASS
hobart/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_4_97
Originator(s): fvitt, mmills, tilmes
Date: 8 Dec 2016
One-line Summary: Update WACCM6 chemistry and remove obsolete compsets
Purpose of changes:
- use the latest TSMLT chemistry in WACCM6
- add specified dynamics compset for WACCM6 FSDW
- remove unsupported obsolete compsets and associated build namelist
use case files and regression tests
- add default fields to h0 history when history_chemistry is TRUE
- correction in how emissions scaling factor is applied
Bugs fixed (include bugzilla ID):
correction in emissions scaling factor
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
D components/cam/src/chemistry/pp_trop_mozart_mam3
D components/cam/src/chemistry/pp_trop_strat_soa
D components/cam/src/chemistry/pp_trop_strat_mam3
D components/cam/src/chemistry/pp_trop_mozart_soa
D components/cam/src/chemistry/pp_trop_strat_mam7
- remove obsolete chem packages
D components/cam/test/system/tests_ccmi_mpi
D components/cam/test/system/tests_ccmi_hybrid
- these no longer apply
D components/cam/test/system/config_files/f1.9c5tsmam7h
D components/cam/test/system/config_files/f4c5tsmam3dm
D components/cam/test/system/config_files/f4c5tsmam7dm
D components/cam/test/system/config_files/f4c5mzmam3dm
D components/cam/test/system/config_files/fsd1.9c4wtssoadh
D components/cam/test/system/config_files/f1.9c4tssoadh
D components/cam/test/system/config_files/f1.9c4wtssoadh
D components/cam/test/system/config_files/f4c4soadm
D components/cam/test/system/config_files/f1.9c4soadh
D components/cam/test/system/config_files/f1.9c5tsmam3dh
D components/cam/test/system/config_files/f4c4wtssoadh
D components/cam/test/system/config_files/f4c4wtssoadm
D components/cam/test/system/config_files/fsd1.9c4wtssoah
D components/cam/test/system/config_files/f1.9c5mzmam3dh
- remove these obsolete configure files
D components/cam/bld/namelist_files/use_cases/cam4_bam_radpsv_geos5.xml
D components/cam/bld/namelist_files/use_cases/waccm_sc_2005-2100_cam4_rcp26.xml
D components/cam/bld/namelist_files/use_cases/waccm_sc_1955-2005_cam4.xml
D components/cam/bld/namelist_files/use_cases/sd_2007-2011_htap_trop_strat_soa.xml
D components/cam/bld/namelist_files/use_cases/waccm_sc_1850_cam4.xml
D components/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp45.xml
D components/cam/bld/namelist_files/use_cases/1850-2005_cam4_super_fast_llnl.xml
D components/cam/bld/namelist_files/use_cases/cam4_chem_radpsv_geos5.xml
D components/cam/bld/namelist_files/use_cases/cam4_trop_strat_chem.xml
D components/cam/bld/namelist_files/use_cases/cam5_trop_strat_chem.xml
D components/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp85.xml
D components/cam/bld/namelist_files/use_cases/waccm_1955-2005_cam4.xml
D components/cam/bld/namelist_files/use_cases/waccm_1974-2005_cam4.xml
D components/cam/bld/namelist_files/use_cases/2005-2100_cam4_strataero_rcp45.xml
D components/cam/bld/namelist_files/use_cases/2000_cam4_trop_moz_soa.xml
D components/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_waccm_ma.xml
D components/cam/bld/namelist_files/use_cases/sd_waccm_geos5.xml
D components/cam/bld/namelist_files/use_cases/sd_1975-2010_ccmi_refc1_waccm_tsmlt.xml
D components/cam/bld/namelist_files/use_cases/1850_waccm_tsmlt_cam5.xml
D components/cam/bld/namelist_files/use_cases/mozart_megan_emis.xml
D components/cam/bld/namelist_files/use_cases/2007-2011_htap_trop_strat_soa.xml
D components/cam/bld/namelist_files/use_cases/sd_1975-2010_ccmi_refc1_trop_strat_soa.xml
D components/cam/bld/namelist_files/use_cases/waccm_1850_cam4.xml
D components/cam/bld/namelist_files/use_cases/volc_waccm_tsmlt_megan_cam5.xml
D components/cam/bld/namelist_files/use_cases/1850_cam4_super_fast_llnl.xml
D components/cam/bld/namelist_files/use_cases/2004-2100_ccmi_refc2_rcp45_trop_strat_soa.xml
D components/cam/bld/namelist_files/use_cases/waccm_sc_2005-2100_cam4_rcp45.xml
D components/cam/bld/namelist_files/use_cases/1950-2010_ccmi_refc1_waccm_tsmlt.xml
D components/cam/bld/namelist_files/use_cases/2000_ccmi_refc1_waccm_tsmlt.xml
D components/cam/bld/namelist_files/use_cases/waccm_sc_2005-2100_cam4_rcp85.xml
D components/cam/bld/namelist_files/use_cases/2000_cam5_trop_moz_mam3.xml
D components/cam/bld/namelist_files/use_cases/waccm_1850-2005_cam4.xml
D components/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp26.xml
D components/cam/bld/namelist_files/use_cases/1979-2010_sd_trop_strat_soa.xml
D components/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_rcp85_waccm_ma.xml
D components/cam/bld/namelist_files/use_cases/waccm_refb1.xml
D components/cam/bld/namelist_files/use_cases/volc_waccm_tsmlt_nomegan_cam5.xml
D components/cam/bld/namelist_files/use_cases/sd_1975-2010_ccmi_refc1_waccm_ma.xml
D components/cam/bld/namelist_files/use_cases/rcp6.0_waccm_tsmlt_nomegan_cam5.xml
D components/cam/bld/namelist_files/use_cases/1950-2010_ccmi_refc1_waccm_ma.xml
D components/cam/bld/namelist_files/use_cases/2000_cam5_trop_strat_mam3.xml
D components/cam/bld/namelist_files/use_cases/1850_cam4_trop_bam.xml
D components/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_waccm_tsmlt.xml
D components/cam/bld/namelist_files/use_cases/2000_cam5_trop_strat_mam7.xml
D components/cam/bld/namelist_files/use_cases/sd_cam4_trop_strat_soa.xml
D components/cam/bld/namelist_files/use_cases/volc_waccm_ma_cam5.xml
D components/cam/bld/namelist_files/use_cases/waccm_2000_cam4.xml
D components/cam/bld/namelist_files/use_cases/2000_cam4_super_fast_llnl.xml
D components/cam/bld/namelist_files/use_cases/2000_cam4_trop_strat_soa.xml
D components/cam/bld/namelist_files/use_cases/sd_cam5_trop_strat_mam3.xml
D components/cam/bld/namelist_files/use_cases/1950-2010_ccmi_refc1_trop_strat_soa.xml
D components/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_trop_strat_soa.xml
- remove obsolete use cases
D components/cam/bld/namelist_files/use_cases/2000_trop_strat_vbs_cam55.xml
D components/cam/bld/namelist_files/use_cases/1850_trop_strat_vbs_cam55.xml
D components/cam/bld/namelist_files/use_cases/waccm_tsmlt_2000_cam55.xml
D components/cam/bld/namelist_files/use_cases/waccm_tsmlt_amip_cam55.xml
D components/cam/bld/namelist_files/use_cases/waccm_tsmlt_1850_cam55.xml
D components/cam/bld/namelist_files/use_cases/waccm_sc_amip_cam55.xml
D components/cam/bld/namelist_files/use_cases/waccm_sc_2000_cam55.xml
D components/cam/bld/namelist_files/use_cases/waccm_ma_amip_cam55.xml
D components/cam/bld/namelist_files/use_cases/waccm_ma_1850_cam55.xml
D components/cam/bld/namelist_files/use_cases/waccm_ma_2000_cam55.xml
- most of these *cam55.xml use cases were renamed as *cam6.xml
List all subroutines added and what they do:
A components/cam/test/system/config_files/fsd0.9c5.5wtsmlth
- new test configure for 1-degree SD WACCM
A components/cam/bld/namelist_files/use_cases/sd_waccm_tsmlt_cam6.xml
A components/cam/bld/namelist_files/use_cases/2000_trop_strat_vbs_cam6.xml
A components/cam/bld/namelist_files/use_cases/1850_trop_strat_vbs_cam6.xml
A components/cam/bld/namelist_files/use_cases/waccm_tsmlt_amip_cam6.xml
A components/cam/bld/namelist_files/use_cases/waccm_tsmlt_1850_cam6.xml
A components/cam/bld/namelist_files/use_cases/waccm_tsmlt_2000_cam6.xml
A components/cam/bld/namelist_files/use_cases/waccm_sc_amip_cam6.xml
A components/cam/bld/namelist_files/use_cases/waccm_ma_amip_cam6.xml
A components/cam/bld/namelist_files/use_cases/waccm_ma_1850_cam6.xml
A components/cam/bld/namelist_files/use_cases/waccm_ma_2000_cam6.xml
- new use cases for cam6 physics
List all existing files that have been modified, and describe the changes:
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_lin_matrix.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_phtadj.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/m_rxt_id.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_nln_matrix.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_lu_factor.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_indprd.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_sim_dat.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/m_spc_id.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/chem_mech.doc
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/chem_mech.in
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_lu_solve.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_setrxt.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_adjrxt.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_rxt_rates_conv.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/chem_mods.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_prod_loss.F90
- use the latest TSMLT chemistry in WACCM
M components/cam/src/chemistry/mozart/mo_srf_emissions.F90
- correction in the scaling factor
M components/cam/src/chemistry/mozart/mo_extfrc.F90
- correction in the scaling factor
- added default history to h0 when history_chemistry is TRUE
M components/cam/src/chemistry/modal_aero/aero_model.F90
M components/cam/src/chemistry/mozart/mo_chm_diags.F90
M components/cam/src/chemistry/mozart/chemistry.F90
- added default history to h0 when history_chemistry is TRUE
M components/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
- remove unused history_aerosol variable
M components/cam/test/system/tests_posttag_yellowstone
M components/cam/test/system/tests_waccm_hybrid
M components/cam/test/system/tests_waccm_mpi
- obsolete test were removed
M components/cam/test/system/config_files/f4c5portdm
M components/cam/test/system/config_files/f4c5portdh
- use trop_mam3 chem -- trop_strat_mam3 no longer exists
M components/cam/test/system/tests_titan
M components/cam/test/system/tests_pretag_yellowstone
M components/cam/test/system/input_tests_master
M components/cam/test/system/tests_chem_hybrid
M components/cam/test/system/tests_chem_mpi
- obsolete test were removed
- waccm6 tests were adjusted
M components/cam/test/system/nl_files/outfrq3s_sd
- use megan emis factors file compatible with clm4
M components/cam/bld/config_files/definition.xml
- adjust list of allowed chem configure options
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
- new megan factors file
- remove defaults for obsolete chemistry packages
- increase default fv_nsplit from 12 to 16 for 1-degree waccm
M components/cam/bld/namelist_files/namelist_definition.xml
- adjust the list of possible chemistry packages
M components/cam/bld/namelist_files/use_cases/sd_waccm5_geos5.xml
- correction in rate_sums specifier
M components/cam/bld/build-namelist
- adjustment to the flbc_list for TSMLT chemistry
- removed references to obsolete chem packages
M components/cam/bld/configure
- adjust the list of possible chemistry packages
M components/cam/cime_config/config_compsets.xml
M components/cam/cime_config/config_component.xml
- clean out obsolete compsets
- add FSDW compset for waccm6
M components/cam/cime_config/testdefs/testlist_cam.xml
- test the supported waccm6 compsets
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
027 bl456 TBL.sh f1.9c5.5wtmam4h outfrq3s_megan+waccm_tsmlt_2000_cam6 9s ......................FAIL! rc= 7 at Wed Dec 7 23:47:31 MST 2016
044 bl470 TBL.sh fsd1.9c5.4wtclbh outfrq3s_sd+sd_waccm5_geos5 9s ..............................FAIL! rc= 7 at Thu Dec 8 00:59:34 MST 2016
- expected failures due to changes in waccm_tsmlt_mam4 chemistry mechanism
041 bl492 TBL.sh f0.9c5.5mam4vbsh outfrq3s+1850_trop_strat_vbs_cam6 9s ........................FAIL! rc= 7 at Thu Dec 8 00:32:58 MST 2016
- new test for yellowstone
047 bl482 TBL.sh f1.9c5.5wscmam4dh outfrq3s+waccm_sc_amip_cam6 9s .............................FAIL! rc= 7 at Thu Dec 8 01:34:46 MST 2016
- expected failure due to changes in waccm_sc_amip_cam6 usecase
yellowstone/intel/aux_cam: All PASS (note NLCOMP tests failed due to lack archived of namelist files from baseline runs)
hobart/nag:
082 bl011 TBL.sh f4c5portdm outfrq24h_port 2d .................................................FAIL! rc= 7 at Wed Dec 7 18:42:33 MST 2016
- expected failure due to the chemistry mechanism used in f4c5portdm changed from "trop_strat_mam3" to "trop_mam3"
-- trop_strat_mam3 chemistry package is removed in this commit
hobart/pgi:
068 bl011 TBL.sh f4c5portdm outfrq24h_port 2d .................................................FAIL! rc= 7 at Wed Dec 7 18:30:19 MST 2016
- expected failure due to the chemistry mechanism used in f4c5portdm changed from "trop_strat_mam3" to "trop_mam3"
-- trop_strat_mam3 chemistry package is removed in this commit
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: waccm_tsmlt_mam4 chemistry, waccm use cases
- nature of change: CAM is bit-for-bit unchanged while WACCM (waccm_tsmlt_mam4 chemistry) has changed
===============================================================
===============================================================
Tag name: cam5_4_96
Originator(s): goldy
Date: 2016-12-04
One-line Summary: Add analytic initial conditions options
Purpose of changes: Enable setting initial conditions to analytic
functions for testing purposes (e.g., dycore validation).
Bugs fixed (include bugzilla ID): NA
Describe any changes made to build system: NA
Describe any changes made to the namelist:
- New namelist item, analytic_ic_type (see below)
List any changes to the defaults for the boundary datasets: NA
Describe any substantial timing or memory changes: NA
Code reviewed by: eaton, cacraig, fvitt, goldy
List all subroutines eliminated: NA
List all subroutines added and what they do:
- analytic_ic_readnl ! Read dyn_test_nl namelist
- analytic_ic_active ! .true. if analytic IC should be set
- analytic_ic_set_ic ! Set analytic initial conditions
List all existing files that have been modified, and describe the changes:
M components/cam/bld/build-namelist
- Added new namelist item, analytic_ic_type, to determine which type of
analytic initial condition to use (or none).
- Make sure that use_topo_file is always expressed in namelist
M components/cam/bld/config_files/definition.xml
- Added new config switch, analytic_ic, to enable conditional compilation
of initial condition code.
M components/cam/bld/configure
- Added new config switch, analytic_ic, to enable conditional compilation
of initial condition code.
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
- Set default for analytic_ic_type (none)
- Set defaults for use_topo_file
M components/cam/bld/namelist_files/namelist_definition.xml
- Added new namelist item, analytic_ic_type, to determine which type of
analytic initial condition to use (or none).
M components/cam/bld/namelist_files/use_cases/dctest_baro_moist.xml
- Added an analytic_ic_type setting (baroclinic_wave)
M components/cam/cime_config/config_component.xml
- Added analytic_ic to the CAM%DCTBM compset.
M components/cam/doc/ChangeLog
- Made this file great again.
M components/cam/src/control/runtime_opts.F90
- Read the analytic IC namelist (analytic_ic_readnl)
M components/cam/src/dynamics/eul/dyn_comp.F90
- Add analytic initial condition reading to Eulerian dycore
M components/cam/src/dynamics/fv/dyn_comp.F90
- Add analytic initial condition reading to finite volume dycore
M components/cam/src/dynamics/se/dyn_comp.F90
- Add analytic initial condition reading to spectral element dycore
M components/cam/src/dynamics/sld/dyn_comp.F90
- Add analytic initial condition reading to SLD dycore
A + components/cam/src/dynamics/tests/dyn_tests_utils.F90
- Added parameters for vertical coordinate type
A + components/cam/src/dynamics/tests/inic_analytic.F90
- Main interface for analytic initial conditions.
A + components/cam/src/dynamics/tests/initial_conditions/ic_baroclinic.F90
- Implementation of analytic initial conditions for baroclinic wave tests.
A + components/cam/src/dynamics/tests/initial_conditions/ic_held_suarez.F90
- Implementation of analytic initial conditions for Held-Suarez tests.
M components/cam/src/physics/cam/const_init.F90
- Clarify when a variable is not found on initial dataset.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS
yellowstone/intel/aux_cam: All PASS - note that the comparison of namelist
file is broken.
hobart/nag: All PASS
hobart/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag: NA
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_4_95
Originator(s): toniazzo, eaton
Date: Fri Dec 2 09:09:05 MST 2016
One-line Summary: add option for angular momentum correction in FV dycore
Purpose of changes:
. Add angular momentum correction/fixer/diagnostic for the FV dycore from
Thomas Toniazzo . The new code is optional, off
by default, and controlled by the new namelist variables described below.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist:
. Add namelist vars to control AM corrections:
logical :: fv_am_correction ! apply correction for angular momentum (AM)
! conservation in SW eqns (default: .false.)
logical :: fv_am_fixer ! apply global fixer to conserve AM (default: .false.)
logical :: fv_am_fix_lbl ! apply AM fixer level by level (default: .false.)
logical :: fv_am_diag ! turn on an AM diagnostic calculation
! written to log file (default: .false.)
List any changes to the defaults for the boundary datasets:
. The default FV, 2deg, 30 level ncdata file is changed
From: cami_0000-01-01_1.9x2.5_L30_c070703.nc
To: cami-mam3_0000-01-01_1.9x2.5_L30_c090306.nc
- this was done to simplify the attributes for the standard cam5
configuration, which in turn allowed simplifying the attributes for the
cam5/aquaplanet ncdata file.
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do:
components/cam/src/dynamics/fv/angular_momentum.F90
. new subroutine for calculating the AM diagnostics which are output to the
log file.
List all existing files that have been modified, and describe the changes:
components/cam/bld/namelist_files/namelist_defaults_cam.xml
. Update attributes for the default FV 2deg L30 ncdata file. This was done
to fix a problem getting the correct default for the corresponding ncdata
file for aquaplanet.
components/cam/bld/namelist_files/namelist_definition.xml
. Add namelist vars to control AM corrections:
logical :: fv_am_correction ! apply correction for angular momentum (AM)
! conservation in SW eqns
logical :: fv_am_fixer ! apply global fixer to conserve AM
logical :: fv_am_fix_lbl ! apply AM fixer level by level
logical :: fv_am_diag ! turn on an AM diagnostic calculation written to log file
components/cam/src/dynamics/fv/cd_core.F90
. add control logicals for correction and fixer
. and output args for fixer
. remove obsolete !CSD$ directives
. pass am_correction to d2a2c_winds
. pass am_correction to d_sw, get ddpa and ddu back
components/cam/src/dynamics/fv/d2a3dikj.F90
. pass am_correction to d2a3dikj
. new code inside am_correction conditionals
components/cam/src/dynamics/fv/dp_coupling.F90
. access am_correction via dyn_state object, and pass to routines d2a3dikj
and uv3s_update
components/cam/src/dynamics/fv/dyn_comp.F90
. dyn_readnl
- add new namelist vars to namelist reader
- add check that fv_am_fixer must be true when fv_am_fix_lbl is true
- call phys_setopts to set fv_am_correction in the phys_control module
. dyn_run
- add AM conservation diagnostic
- add AM fixer
- pass am_correction to te_map
components/cam/src/dynamics/fv/dyn_grid.F90
. dyn_grid_init
- pass am_correction to grid_vars_init
components/cam/src/dynamics/fv/dynamics_vars.F90
. add AM control variables to the t_fvdycore_state derived type
. grid_vars_init
- add dummy arg for am_correction
- modified calc of cose, sinp, and fc inside conditionals
. remove unused routines
components/cam/src/dynamics/fv/sw_core.F90
. c_sw
- new calc of va
. d_sw
- add am_correction, ddp, duc to dummy args
. d2a2c_winds
- add am_correction to dummy args
- mods inside am_correction conditional
components/cam/src/dynamics/fv/te_map.F90
. add am_correction to dummy args
. mods inside am_correction conditional
components/cam/src/dynamics/fv/uv3s_update.F90
. add am_correction to uv3s_update dummy args
. mods inside am_correction conditional
components/cam/src/physics/cam/diffusion_solver.F90
. set module variable am_correction via init_vdiff method
. mod inside am_correction conditional
components/cam/src/physics/cam/phys_control.F90
. add fv_am_correction module data
. add phys_setopts method to allow setting fv_am_correction from the
dyn_readnl method
components/cam/src/physics/cam/vertical_diffusion.F90
. pass fv_am_correction to init_vdiff
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS except:
036 bl421 TBL.sh f1.9c5mam4dh outfrq3s 9s .....................................................FAIL! rc= 7 at Thu Dec 1 20:02:11 MST 2016
This failure is due to a change in the default IC files.
yellowstone/intel/aux_cam: All PASS - note that the comparison of namelist
file is broken.
hobart/nag: All PASS
hobart/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_4_94
Originator(s): cacraig
Date: Nov 28, 2016
One-line Summary: Update to cesm2_beta04 externals and bug fix cice5 tag
Purpose of changes:
- Update CAM externals to match those used in cesm2_beta04
- Using a more recent CICE tag which corrects an issue in F compsets when using the
tag in cesm2_beta04
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
- All compsets are now using CISM2
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M .
M SVN_EXTERNAL_DIRECTORIES
- update to cesm2_beta04 externals
- update to bug fix cice external to fix issue with F compsets
M components/cam/cime_config/config_compsets.xml
- have compsets now use CISM2
- add grid file for CISM2 as specified by Cecile
M components/cam/cime_config/testdefs/testlist_cam.xml
- add aux_short test for F1850_DONOTUSE
M components/cam/test/system/archive_baseline.sh
- added CESM_TESTDIR environment variable to archive_baseline.sh for yellowstone
to enable the archiving of the CESM tests
NOTE -- The CESM_TESTDIR can be found by looking in the file in test/system
and by default is called something like aux_cam_test_20161123172411.log
M components/cam/test/system/test_driver.sh
- removed generate capability from CESM tests - will generate the baselines in
the archive_baseline script
- removed the required test-id which was used for generating baselines
- added CESM_TESTDIR environment variable for yellowstone runs. User can specify
or by default it will create a dir called something like aux_cam_test_20161123172411
in /glade/scratch/$USER on yellowstone
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
004 bl132 TBL.sh e48c4dh ghgrmp_e8+1850_cam4 9s ...............................................FAIL! rc= 7 at Thu Nov 24 04:12:02 MST 2016
008 bl331 TBL.sh f4c4dh co2rmp+1850_cam4 9s ...................................................FAIL! rc= 7 at Thu Nov 24 04:15:05 MST 2016
011 bl353 TBL.sh f1.9c5.4dh outfrq3s 9s .......................................................FAIL! rc= 7 at Thu Nov 24 04:26:42 MST 2016
015 bl357 TBL.sh f1.9c5.5h outfrq3s+1850_cam55 9s .............................................FAIL! rc= 7 at Thu Nov 24 04:32:22 MST 2016
018 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Thu Nov 24 04:44:23 MST 2016
022 bl373 TBL.sh f1.9c4mozdh outfrq3s+2000_cam4_trop_chem 9s ..................................FAIL! rc= 7 at Thu Nov 24 05:00:11 MST 2016
024 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...............................................FAIL! rc= 7 at Thu Nov 24 05:02:39 MST 2016
027 bl485 TBL.sh f1.9c5.5wtmam4h outfrq3s_megan+waccm_tsmlt_2000_cam55 9s .....................FAIL! rc= 7 at Thu Nov 24 05:27:13 MST 2016
030 bl390 TBL.sh f1.9c4wmxh outfrq3s+waccmx_2000_cam4 9s ......................................FAIL! rc= 7 at Thu Nov 24 05:37:24 MST 2016
033 bl397 TBL.sh f1.9c4cm outfrq3s+1850-2005_cam4 9s ..........................................FAIL! rc= 7 at Thu Nov 24 05:40:18 MST 2016
036 bl421 TBL.sh f1.9c5mam4dh outfrq3s 9s .....................................................FAIL! rc= 7 at Thu Nov 24 05:51:30 MST 2016
039 bl429 TBL.sh f1.9c4wmxh outfrq3s+waccmxi_2000_cam4 9s .....................................FAIL! rc= 7 at Thu Nov 24 05:59:37 MST 2016
041 bl430 TBL.sh f1.9c5mzmam3dh outfrq3s_NEUwetdep 9s .........................................FAIL! rc= 7 at Thu Nov 24 06:08:55 MST 2016
044 bl440 TBL.sh f1.9c4soadh outfrq3s+soa_chem_megan_emis 9s ..................................FAIL! rc= 7 at Thu Nov 24 06:27:15 MST 2016
047 bl470 TBL.sh fsd1.9c5.4wtclbh outfrq3s_sd+sd_waccm5_geos5 9s ..............................FAIL! rc= 7 at Thu Nov 24 06:52:20 MST 2016
050 bl482 TBL.sh f1.9c5.5wscmam4dh outfrq3s+waccm_sc_amip_cam55 9s ............................FAIL! rc= 7 at Thu Nov 24 07:32:34 MST 2016
054 bl387 TBL.sh f1.9c4dm outfrq3s+1850-2005_cam4 9s ..........................................FAIL! rc= 7 at Thu Nov 24 07:39:05 MST 2016
057 bl388 TBL.sh f1.9c4m outfrq24h+1850_cam4 2d ...............................................FAIL! rc= 7 at Thu Nov 24 07:40:35 MST 2016
068 bl740 TBL.sh h30c5h outfrq3s_lb0 9s .......................................................FAIL! rc= 7 at Thu Nov 24 08:00:14 MST 2016
yellowstone/intel/aux_cam:
ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s_clm5 (Overall: DIFF), details:
FAIL ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s_clm5 NLCOMP
FAIL ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s_clm5 BASELINE
FAIL ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s_clm5 TPUTCOMP Error: Computation time increase > 25% from baseline
ERP_Ln9.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq9s_clm5 (Overall: DIFF), details:
FAIL ERP_Ln9.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq9s_clm5 NLCOMP
FAIL ERP_Ln9.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq9s_clm5 BASELINE
SMS_Lm13.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq1m_clm5 (Overall: DIFF), details:
FAIL SMS_Lm13.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq1m_clm5 NLCOMP
FAIL SMS_Lm13.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq1m_clm5 BASELINE
hobart/nag:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s ...........................................................FAIL! rc= 7 at Tue Nov 22 09:52:45 MST 2016
010 bl112 TBL.sh e8c5dm ghgrmp_e8 9s ..........................................................FAIL! rc= 7 at Tue Nov 22 09:56:48 MST 2016
014 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Tue Nov 22 10:00:03 MST 2016
017 bl221 TBL.sh f10c5spcamsm outfrq3s 9s .....................................................FAIL! rc= 7 at Tue Nov 22 10:07:16 MST 2016
021 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Tue Nov 22 10:12:26 MST 2016
031 bl314 TBL.sh f10c4wscdm outfrq3s 9s .......................................................FAIL! rc= 7 at Tue Nov 22 10:21:54 MST 2016
037 bl318 TBL.sh f10c4cdm sat_hist 9s .........................................................FAIL! rc= 7 at Tue Nov 22 10:30:28 MST 2016
040 bl324 TBL.sh f4c4rrtmgdm outfrq3s 9s ......................................................FAIL! rc= 7 at Tue Nov 22 10:33:31 MST 2016
043 bl325 TBL.sh f10c5t5mdmsc outfrq3s_subcol 9s ..............................................FAIL! rc= 7 at Tue Nov 22 10:37:42 MST 2016
049 bl338 TBL.sh f10c5dm rad_diag_mam 9s ......................................................FAIL! rc= 7 at Tue Nov 22 10:41:09 MST 2016
055 bl354 TBL.sh f1.9c5.4dm outfrq3s_npryz 9s .................................................FAIL! rc= 7 at Tue Nov 22 10:55:59 MST 2016
058 bl356 TBL.sh fsd1.9c5clbdm off1.9x2.5 9s ..................................................FAIL! rc= 7 at Tue Nov 22 11:18:54 MST 2016
061 bl368 TBL.sh f1.9c5mg2dm outfrq3s_macmic2_classnuc 9s .....................................FAIL! rc= 7 at Tue Nov 22 11:31:21 MST 2016
064 bl369 TBL.sh f1.9c5clbdm outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Tue Nov 22 11:46:55 MST 2016
070 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Tue Nov 22 11:59:49 MST 2016
073 bl426 TBL.sh f10c4wmxdm outfrq3s_ionos 9s .................................................FAIL! rc= 7 at Tue Nov 22 12:07:11 MST 2016
076 bl471 TBL.sh f10c5.5wmam4dm outfrq3s_modalstrat 9s ........................................FAIL! rc= 7 at Tue Nov 22 12:16:28 MST 2016
079 bl805 TBL.sh f4c5carmtt1dm outfrq3s_carma 9s ..............................................FAIL! rc= 7 at Tue Nov 22 12:35:55 MST 2016
hobart/pgi or jaguar/pgi:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s ...........................................................FAIL! rc= 7 at Tue Nov 22 09:56:51 MST 2016
010 bl112 TBL.sh e8c5dm ghgrmp_e8 9s ..........................................................FAIL! rc= 7 at Tue Nov 22 10:05:00 MST 2016
016 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Tue Nov 22 10:14:05 MST 2016
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Tue Nov 22 10:27:31 MST 2016
033 bl314 TBL.sh f10c4wscdm outfrq3s 9s .......................................................FAIL! rc= 7 at Tue Nov 22 10:49:15 MST 2016
036 bl315 TBL.sh f10c3dm outfrq3s+cam3 9s .....................................................FAIL! rc= 7 at Tue Nov 22 10:55:24 MST 2016
041 bl317 TBL.sh f10c5cdm outfrq3s 9s .........................................................FAIL! rc= 7 at Tue Nov 22 11:11:18 MST 2016
045 bl319 TBL.sh f10c5nonedm outfrq3s_bam 9s ..................................................FAIL! rc= 7 at Tue Nov 22 11:18:33 MST 2016
048 bl320 TBL.sh f10c5nonedm rad_diag 9s ......................................................FAIL! rc= 7 at Tue Nov 22 11:20:13 MST 2016
051 bl321 TBL.sh f10c5cdm atrain 9s ...........................................................FAIL! rc= 7 at Tue Nov 22 11:22:55 MST 2016
054 bl334 TBL.sh f10c5udm outfrq3s 9s .........................................................FAIL! rc= 7 at Tue Nov 22 11:31:17 MST 2016
057 bl354 TBL.sh f1.9c5.4dm outfrq3s_npryz 9s .................................................FAIL! rc= 7 at Tue Nov 22 11:51:04 MST 2016
060 bl420 TBL.sh f4c5mam7dm outfrq3s 9s .......................................................FAIL! rc= 7 at Tue Nov 22 12:02:08 MST 2016
063 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Tue Nov 22 12:12:11 MST 2016
** All failing baselines are due to changes in the external components
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: all
- what platforms/compilers: all
- nature of change (roundoff; larger than roundoff but same climate; new
climate): new climate
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository): https://svn-ccsm-models.cgd.ucar.edu/cam1/test_tags/cam5_4_94_n01
- platform/compilers: yellowstone/intel
- configure commandline: Using the F2000_DEV compset
- build-namelist command (or complete namelist):
- To get the CLM namelist changes which mimic Cecile's runs, users need to add to their create_newcase the option:
--user-mods-dir ../../../components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq1m_clm5
** NOTE ** This directory name will be changing in a future tag
- MSS location of output:
Currently: /glade/p/cesm0005/archive/f.e20.F2000_DEV.f09_f09.cam5_4_94_n01
Plan to put at: /CCSM/csm/f.e20.F2000_DEV.f09_f09.cam5_4_94_n01 (yellowstone is
currently down and have not been able to send these files to long-term archive)
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
http://webext.cgd.ucar.edu/FCLIMO/f.e20.F2000_DEV.f09_f09.cam5_4_94_n01/atm/f.e20.F2000_DEV.f09_f09.cam5_4_94_n01-f.e15.F2000_DEV.f09_f09.cam5_4_89.119/
===============================================================
===============================================================
Tag name: cam5_4_93
Originator(s): jjb278, brianpm, olson, eaton
Date: Wed Nov 2 08:55:33 MDT 2016
One-line Summary: Aquaplanet mods for cesm2.
Purpose of changes:
. Add option to fix the cloud droplet and crystal number concentrations in MG.
. Remove the mg1_5 development code which is superseded by mg2_0.
. Fix broken namelist variable aqua_planet_sst to set options for the
analytic SSTs.
. Mods to configure and buildnamelist to support SOM aquaplanet.
Bugs fixed (include bugzilla ID):
. Fix the aqua_planet_sst variable which has been broken longer than anyone
can remember.
Describe any changes made to build system:
. configure
- remove option '-microphys mg1.5'
- add -aquaplanet switch to turn on the aqua-planet mode in CAM. By
default this switch also sets the -ocn option to aquaplanet, but
this can be overridden from the command line.
- add option '-ocn som' to support aquaplanet with SOM using CAM
standalone scripts
Describe any changes made to the namelist:
. Add variables to allow fixing the cloud droplet and crystal numbers in MG
- micro_mg_nccons - set .true. to hold cloud droplet number constant
- micro_mg_nicons - set .true. to hold cloud ice number constant
- micro_mg_ncnst - droplet num concentration when nccons=.true. (default: 100.e6 m-3)
- micro_mg_ninst - ice num concentration when nicons=.true. (default: 0.1e6 m-3)
. modify output of docn namelist and stream files as appropriate for som.
List any changes to the defaults for the boundary datasets:
. Added the following initial condition files for aquaplanet runs:
aqua_0006-01-01_0.9x1.25_L26_c161020.nc: FV, 1-deg, cam4 phys
aqua_0006-01-01_0.9x1.25_L30_c161020.nc: FV, 1-deg, cam5 phys
aqua_0006-01-01_0.9x1.25_L32_c161020.nc: FV, 1-deg, cam5.5 phys
aqua_0006-01-01_1.9x2.5_L26_c161020.nc : FV, 2-deg, cam4 phys
aqua_0006-01-01_1.9x2.5_L30_c161020.nc : FV, 2-deg, cam5 phys
aqua_0006-01-01_1.9x2.5_L32_c161020.nc : FV, 2-deg, cam5.5 phys
aqua_0006-01-01_ne30_np4_L26_c161027.nc: SE, 1-deg, cam4 phys
aqua_0006-01-01_ne30_np4_L30_c161027.nc: SE, 1-deg, cam5 phys
aqua_0006-01-01_ne30_np4_L32_c161027.nc: SE, 1-deg, cam5.5 phys
aqua_0006-01-01_ne16_np4_L26_c161027.nc: SE, 2-deg, cam4 phys
aqua_0006-01-01_ne16_np4_L30_c161027.nc: SE, 2-deg, cam5 phys
aqua_0006-01-01_ne16_np4_L32_c161027.nc: SE, 2-deg, cam5.5 phys
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
components/cam/src/physics/cam/micro_mg1_5.F90
. remove old development code
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
components/cam/bld/build-namelist
. write aquaplanet namelist to file aquap_in
. modify output of docn namelist and stream files as appropriate for som.
components/cam/bld/config_files/definition.xml
components/cam/bld/configure
. remove mg1.5 as valid option
. add -aquaplanet switch.
. allow 'som' as a valid value for -ocn option. SOM is implemented using
the DOCN component.
components/cam/bld/namelist_files/namelist_defaults_cam.xml
. change attribute ocn="aquaplanet" to aquaplanet="1" in ncdata elements.
. add bndtvs and focndomain entries for som testing.
components/cam/bld/namelist_files/namelist_definition.xml
. entries for micro_mg_nccons, micro_mg_nicons,
micro_mg_ncnst, micro_mg_ninst
. remove mg1.5 as valid option
components/cam/cime_config/buildnml
. add code to copy the new aquap_in file to the run directory.
components/cam/src/physics/cam/micro_mg1_0.F90
. add 4 new variables that are namelist settable
- nccons - set .true. to hold cloud droplet number constant
- nicons - set .true. to hold cloud ice number constant
- ncnst - droplet num concentration when nccons=.true. (default: 100.e6 m-3)
- ninst - ice num concentration when nicons=.true. (default: 0.1e6 m-3)
components/cam/src/physics/cam/micro_mg2_0.F90
. Set nccons, nicons, ncnst, ninst via the initialization method from
namelist specified values.
components/cam/src/physics/cam/micro_mg_cam.F90
. add namelist variables (micro_mg_nccons, micro_mg_nicons,
micro_mg_ncnst, micro_mg_ninst) to allow constant number concentrations
and pass through the init method arg list.
. update mpi_bcast calls and add log output for namelist settings
. remove calls to mg1.5 routines
components/cam/utils/cam_aqua/ocn_comp.F90
. add namelist reader for aqua_planet_sst to set analytic SST option
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
060 bl731 TBL.sh h16c3aqdh aqua+aquaplanet_cam3 9s ............................................FAIL! rc= 7 at Tue Nov 1 21:17:44 MDT 2016
063 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ........................................FAIL! rc= 7 at Tue Nov 1 21:29:12 MDT 2016
072 bl751 TBL.sh h30c4aqdm outfrq3s+aquaplanet_cam4 9s ........................................FAIL! rc= 7 at Tue Nov 1 21:49:17 MDT 2016
yellowstone/intel/aux_cam: PASS
hobart/nag: all PASS except
067 bl734 TBL.sh h16c5naqdm outfrq3s_bam+aquaplanet_cam5 9s ...................................FAIL! rc= 7 at Mon Oct 31 14:49:43 MDT 2016
hobart/pgi: all PASS except
066 bl734 TBL.sh h16c5naqdm outfrq3s_bam+aquaplanet_cam5 9s ...................................FAIL! rc= 7 at Mon Oct 31 14:55:16 MDT 2016
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB. The aquaplanet tests that failed
baselines are using new IC files.
===============================================================
===============================================================
Tag name: cam5_4_92
Originator(s): cacraig
Date: Oct 27, 2016
One-line Summary: Update to current cime (to restore auto-resubmit functionality)
Purpose of changes:
- Update to current cime version - this will enable auto-resubmit to work again
- This version of cime requires a namelist parameter "cime_model" and the contents of
the docn_ocn_in file is now in ocn_in.
- Added FC6AQUAP compset
- Update the testlist format to version 2
- Remove "-file xx" from xmlchange commands in CAM regression testing scripts
- Added the ability for the CAM test_driver to rerun the CESM tests and only rerun
the parts which did not PASS
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M .
M SVN_EXTERNAL_DIRECTORIES
- update to current cime tag
- this fixes the issues with auto-restart
M components/cam/bld/build-namelist
M components/cam/bld/namelist_files/namelist_definition.xml
- cime_model is now a required namelist variable, hardwire it to cesm
- docn_ocn_in is no longer a separate file. Incorporate all its
values in ocn_in
M components/cam/cime_config/config_compsets.xml
- add in aquaplanet compset at the request of Brian Medeiros and Jim Benedict
M components/cam/cime_config/testdefs/testlist_cam.xml
- update to the version 2 of the testlist format
M components/cam/test/system/TCB_ccsm.sh
M components/cam/test/system/TER_ccsm.sh
- remove "-file xx" from xmlchange as it no longer is an allowed field (xmlchange
internally determines which file it needs to modify)
M components/cam/test/system/test_driver.sh
- add two new flags, "--rerun-cesm" and "--test-id"
* rerun-cesm will rerun the failed tests
* test-id is used to specify the actual testname
- if CAM_TAG is supplied, the CESM jobs will be collected in that directory in
the user's /glade/scratch directory
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB
yellowstone/intel/aux_cam: all BFB
ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s_clm5 (Overall: NLFAIL), details:
FAIL ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s_clm5 NLCOMP
ERP_Ln9.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq9s_clm5 (Overall: NLFAIL), details:
FAIL ERP_Ln9.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq9s_clm5 NLCOMP
SMS_Lm13.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq1m_clm5 (Overall: NLFAIL), details:
FAIL SMS_Lm13.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq1m_clm5 NLCOMP
hobart/nag: all BFB
hobart/pgi or jaguar/pgi: all BFB
===============================================================
===============================================================
Tag name: cam5_4_91
Originator(s): fvitt, mmills, tilmes, dkin
Date: 21 Oct 2016
One-line Summary: WACCM chemistry and other msic updates -- "Phase 3" science changes
Purpose of changes:
- Use vectorized implicit solver for the waccm_tsmlt_mam4 chemistry
- Include chemistry preprocessor capable of generating vectorized solvers
- Update chemistry in waccm_tsmlt_mam4 mechanism -- includes:
. VBS SOA scheme
. JPL15 rates
- Add chemical forcings associated with energetic particle ionization
to WACCM6 build-namelist use cases
- Unify tropopause definition for chemistry and aerosol processes for consistency
- Use CFC 11* climate forcings in SC-WACCM
- SAD calculation change for chemistry
- Update JEUV heating efficiency for WACCM and WACCMX
- Add EPP settings to WACCM6 build-namelist use cases
- Fix a bug in mo_strato_rates
- Removed obsolete chemistry packages which were for WACCM with 3-mode aerosol scheme
Bugs fixed (include bugzilla ID):
- mo_strato_rates
. gamma terms need to be added rather than multipled -- dkin
- epp_ionization
. a correction in the error checking logic
Describe any changes made to build system:
Describe any changes made to the namelist:
The following namelist options are added to keep CAM bit-for-bit unchanged.
These may be removed in the future.
prescribed_strataero_use_chemtrop (logical):
Indicates whether to use the unified chemistry tropopause method to set prescribed
stratospheric aerosols below the tropopause to zero. This has a maximum altitude
level corresponding to 300 hPa for latitudes poleward of 50 degrees.
Default: set by build-namelist
chem_use_chemtrop (logical):
Indicates whether to use the unified chemistry tropopause method to set the
tropopause used in gas phase and aerosol chemical processes. This has a maximum
altitude level corresponding to 300 hPa for latitudes poleward of 50 degrees.
Default: set by build-namelist
The following modification to the existing namelist option is for very long rate
summation strings in the namelist.
rxn_rate_sums (char*256(200)):
Give the user the ability to specify rate families (or groupings) diagnostics based
on reaction tag names. These group names can be added to history fincl variables.
To indicate the summation continues with next string use "&" character at the end
of the string.
Example:
rate_sums =
'OX_P = NO_HO2 + CH3O2_NO + 2*jo2_b ... ',
'OX_L = NO2_O_M + HO2_O3 + CLO_O ...',
'RO2_RO2_sum = CH3O2_CH3O2a + CH3O2_CH3O2b + C2H5O2_CH3O2 + C2H5O2_C2H5O2 + CH3CO3_CH3O2 &',
'+ CH3CO3_CH3CO3 + CH3H7O2_CH3O2 + RO2_CH3O2 + MACRO2_CH3O2 ...',
fincl1 = 'OX_P','OX_L', 'RO2_RO2_sum', ...
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: cacraig, eaton
List all subroutines eliminated:
D components/cam/src/chemistry/mozart/mo_exp_sol.F90
- now each mechanism has it's own version of exp_sol which may
or may not be vectorized
D components/cam/src/chemistry/pp_waccm_ma_mam3
D components/cam/src/chemistry/pp_waccm_sc_mam3
- removed the waccm 3-mode mechanism -- 4-mode aerosol scheme is preferred
D components/cam/bld/namelist_files/use_cases/waccm_2000_cam5.xml
D components/cam/bld/namelist_files/use_cases/waccm_sc_1955-2005_cam5.xml
D components/cam/bld/namelist_files/use_cases/waccm_sc_2000_cam5.xml
D components/cam/bld/namelist_files/use_cases/waccm_1955-2005_cam5.xml
D components/cam/bld/namelist_files/use_cases/waccm_1850_cam5.xml
- removed these obsolete use cases
D components/cam/test/system/config_files/f1.9c5wmam3dh
D components/cam/test/system/config_files/f10c5wmam3dm
D components/cam/test/system/config_files/f1.9c5wscmam3dh
D components/cam/test/system/config_files/f1.9c5wscmam3dm
D components/cam/test/system/config_files/f1.9c5wmam3h
D components/cam/test/system/config_files/f1.9c5wmclbdh
D components/cam/test/system/config_files/f4c5wmclbdh
D components/cam/test/system/config_files/f10c5wscmam3dm
- removed these obsolete test configurations
List all subroutines added and what they do:
A components/cam/src/chemistry/mozart/chem_prod_loss_diags.F90
- a utility module to be shared by scalar and vectorized solvers
to output chemistry production and loss diagnostics
A components/cam/bld/namelist_files/use_cases/waccm_ma_amip_cam55.xml
A components/cam/bld/namelist_files/use_cases/waccm_sc_2000_cam55.xml
A components/cam/bld/namelist_files/use_cases/waccm_ma_1850_cam55.xml
A components/cam/bld/namelist_files/use_cases/waccm_sc_amip_cam55.xml
A components/cam/bld/namelist_files/use_cases/waccm_ma_2000_cam55.xml
- some new waccm mid-atmos and simplified-chem use cases
A components/cam/src/chemistry/pp_super_fast_llnl/mo_exp_sol.F90
A components/cam/src/chemistry/pp_trop_mozart_mam3/mo_exp_sol.F90
A components/cam/src/chemistry/pp_none/mo_exp_sol.F90
A components/cam/src/chemistry/pp_trop_mam3/mo_exp_sol.F90
A components/cam/src/chemistry/pp_trop_mam4/mo_exp_sol.F90
A components/cam/src/chemistry/pp_trop_strat_soa/mo_exp_sol.F90
A components/cam/src/chemistry/pp_trop_bam/mo_exp_sol.F90
A components/cam/src/chemistry/pp_trop_mam7/mo_exp_sol.F90
A components/cam/src/chemistry/pp_waccm_ma/mo_exp_sol.F90
A components/cam/src/chemistry/pp_waccm_ma_sulfur/mo_exp_sol.F90
A components/cam/src/chemistry/pp_trop_ghg/mo_exp_sol.F90
A components/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_exp_sol.F90
A components/cam/src/chemistry/pp_waccm_tsmlt/mo_exp_sol.F90
A components/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_exp_sol.F90
A components/cam/src/chemistry/pp_trop_strat_mam3/mo_exp_sol.F90
A components/cam/src/chemistry/pp_waccm_sc/mo_exp_sol.F90
A components/cam/src/chemistry/pp_trop_mozart/mo_exp_sol.F90
A components/cam/src/chemistry/pp_trop_mozart_soa/mo_exp_sol.F90
A components/cam/src/chemistry/pp_trop_strat_mam7/mo_exp_sol.F90
A components/cam/src/chemistry/pp_terminator/mo_exp_sol.F90
A components/cam/src/chemistry/pp_waccm_ma_mam4/mo_exp_sol.F90
A components/cam/src/chemistry/pp_waccm_mad_mam4/mo_exp_sol.F90
A components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_exp_sol.F90
A components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_exp_sol.F90
- each chem package has it's own version of explicit solver and may differ
depending on if the solvers are scalar or vector versions
A components/cam/src/chemistry/pp_waccm_sc_mam4/mo_adjrxt.F90
A components/cam/src/chemistry/pp_waccm_sc_mam4/mo_rxt_rates_conv.F90
A components/cam/src/chemistry/pp_waccm_sc_mam4/mo_imp_sol.F90
A components/cam/src/chemistry/pp_waccm_sc_mam4/chem_mods.F90
A components/cam/src/chemistry/pp_waccm_sc_mam4/mo_prod_loss.F90
A components/cam/src/chemistry/pp_waccm_sc_mam4/mo_lin_matrix.F90
A components/cam/src/chemistry/pp_waccm_sc_mam4/mo_exp_sol.F90
A components/cam/src/chemistry/pp_waccm_sc_mam4/m_rxt_id.F90
A components/cam/src/chemistry/pp_waccm_sc_mam4/mo_phtadj.F90
A components/cam/src/chemistry/pp_waccm_sc_mam4/mo_nln_matrix.F90
A components/cam/src/chemistry/pp_waccm_sc_mam4/mo_lu_factor.F90
A components/cam/src/chemistry/pp_waccm_sc_mam4/mo_sim_dat.F90
A components/cam/src/chemistry/pp_waccm_sc_mam4/mo_indprd.F90
A components/cam/src/chemistry/pp_waccm_sc_mam4/m_spc_id.F90
A components/cam/src/chemistry/pp_waccm_sc_mam4/chem_mech.doc
A components/cam/src/chemistry/pp_waccm_sc_mam4/chem_mech.in
A components/cam/src/chemistry/pp_waccm_sc_mam4/mo_lu_solve.F90
A components/cam/src/chemistry/pp_waccm_sc_mam4/mo_setrxt.F90
A components/cam/src/chemistry/pp_waccm_sc_mam4
- waccm_sc_mam4 replaces pre-existing waccm_sc_mam3 chem package
A components/cam/test/system/config_files/f1.9c5.5wmam4h
A components/cam/test/system/config_files/f10c5.5wmam4dm
A components/cam/test/system/config_files/f1.9c5.5wmam4dh
A components/cam/test/system/config_files/f1.9c5.5wscmam4dh
A components/cam/test/system/config_files/f1.9c5.5wscmam4dm
A components/cam/test/system/config_files/f10c5.5wscmam4dm
- some test config files for testing 4-mode aerosol scheme in waccm
List all existing files that have been modified, and describe the changes:
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
- define defaults for use_chemtrop namelist options
- remove obsolete waccm*mam3 defaults and add defaulls for waccm*mam4
M components/cam/bld/namelist_files/namelist_definition.xml
- new namelist options added (see above)
M components/cam/bld/configure
- adjustment to the list of available chemistry packages
- removal of refrence to waccmx_ma chemistry which does not exit
M components/cam/bld/build-namelist
- set defaults for new *_chemtrop options (see above)
- removal of refrence to waccmx_ma chemistry which does not exit
- misc chemistry corrections
M components/cam/bld/config_files/definition.xml
- adjustments to the list of chemsitry packages
M components/cam/src/physics/cam/modal_aer_opt.F90
- addded aerosol optics diagnostics
M components/cam/src/chemistry/modal_aero/aero_model.F90
- provide effective radius for use in chemistry
M components/cam/src/chemistry/bulk_aero/aero_model.F90
- interface change
M components/cam/src/chemistry/mozart/epp_ionization.F90
- bug fix in the error checking logic
M components/cam/src/chemistry/mozart/mo_waccm_hrates.F90
M components/cam/src/chemistry/mozart/mo_jeuv.F90
- update JEUV heating efficiency
M components/cam/src/chemistry/mozart/rate_diags.F90
- allow very long summation strings to be continued in the next
string in the summation list in the namelist
M components/cam/src/chemistry/mozart/chemistry.F90
- mods relevant to the unified chemistry tropopause level
M components/cam/src/chemistry/mozart/exbdrift.F90
- removed outfld calls for ion drift velocities which is done in iondrag
M components/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
- mods relevant to the unified chemistry tropopause level
- SAD diagnostics for troposphere and stratopshere
- effective radius diagnostics
M components/cam/src/chemistry/mozart/cfc11star.F90
- added ability to use prescribed CFC tracers to compute CFCSTAR
M components/cam/src/chemistry/mozart/mo_neu_wetdep.F90
- allow wet removal of 'NDEP' and 'NHDEP'
M components/cam/src/chemistry/mozart/mo_usrrxt.F90
- JPL15 updates
- effective radius changes
M components/cam/src/chemistry/mozart/mo_strato_rates.F90
- Doug's bug fix
M components/cam/src/chemistry/utils/prescribed_strataero.F90
- convert radius units from centimeters to meters
- accept SAD units of cm2/cm3 in addition ot um2/cm3
- option to use chemtrop method for tropopause level below which to zero out aerosols
M components/cam/src/chemistry/utils/tracer_data.F90
- improvement on error message
M components/cam/src/chemistry/pp_super_fast_llnl/chem_mods.F90
M components/cam/src/chemistry/pp_super_fast_llnl/mo_prod_loss.F90
M components/cam/src/chemistry/pp_super_fast_llnl/mo_lin_matrix.F90
M components/cam/src/chemistry/pp_super_fast_llnl/m_rxt_id.F90
M components/cam/src/chemistry/pp_super_fast_llnl/mo_phtadj.F90
M components/cam/src/chemistry/pp_super_fast_llnl/mo_nln_matrix.F90
M components/cam/src/chemistry/pp_super_fast_llnl/mo_lu_factor.F90
M components/cam/src/chemistry/pp_super_fast_llnl/mo_sim_dat.F90
M components/cam/src/chemistry/pp_super_fast_llnl/mo_indprd.F90
M components/cam/src/chemistry/pp_super_fast_llnl/m_spc_id.F90
M components/cam/src/chemistry/pp_super_fast_llnl/mo_lu_solve.F90
M components/cam/src/chemistry/pp_super_fast_llnl/mo_setrxt.F90
M components/cam/src/chemistry/pp_super_fast_llnl/mo_adjrxt.F90
M components/cam/src/chemistry/pp_super_fast_llnl/mo_imp_sol.F90
M components/cam/src/chemistry/pp_trop_mozart_mam3/mo_imp_sol.F90
M components/cam/src/chemistry/pp_trop_mozart_mam3/chem_mods.F90
M components/cam/src/chemistry/pp_trop_mozart_mam3/mo_prod_loss.F90
M components/cam/src/chemistry/pp_trop_mozart_mam3/mo_lin_matrix.F90
M components/cam/src/chemistry/pp_trop_mozart_mam3/m_rxt_id.F90
M components/cam/src/chemistry/pp_trop_mozart_mam3/mo_phtadj.F90
M components/cam/src/chemistry/pp_trop_mozart_mam3/mo_nln_matrix.F90
M components/cam/src/chemistry/pp_trop_mozart_mam3/mo_lu_factor.F90
M components/cam/src/chemistry/pp_trop_mozart_mam3/mo_sim_dat.F90
M components/cam/src/chemistry/pp_trop_mozart_mam3/mo_indprd.F90
M components/cam/src/chemistry/pp_trop_mozart_mam3/m_spc_id.F90
M components/cam/src/chemistry/pp_trop_mozart_mam3/mo_lu_solve.F90
M components/cam/src/chemistry/pp_trop_mozart_mam3/mo_setrxt.F90
M components/cam/src/chemistry/pp_trop_mozart_mam3/mo_adjrxt.F90
M components/cam/src/chemistry/pp_none/mo_adjrxt.F90
M components/cam/src/chemistry/pp_none/mo_imp_sol.F90
M components/cam/src/chemistry/pp_none/chem_mods.F90
M components/cam/src/chemistry/pp_none/mo_prod_loss.F90
M components/cam/src/chemistry/pp_none/mo_lin_matrix.F90
M components/cam/src/chemistry/pp_none/mo_phtadj.F90
M components/cam/src/chemistry/pp_none/mo_nln_matrix.F90
M components/cam/src/chemistry/pp_none/mo_lu_factor.F90
M components/cam/src/chemistry/pp_none/mo_indprd.F90
M components/cam/src/chemistry/pp_none/m_spc_id.F90
M components/cam/src/chemistry/pp_none/mo_lu_solve.F90
M components/cam/src/chemistry/pp_trop_mam3/mo_adjrxt.F90
M components/cam/src/chemistry/pp_trop_mam3/mo_imp_sol.F90
M components/cam/src/chemistry/pp_trop_mam3/chem_mods.F90
M components/cam/src/chemistry/pp_trop_mam3/mo_prod_loss.F90
M components/cam/src/chemistry/pp_trop_mam3/mo_lin_matrix.F90
M components/cam/src/chemistry/pp_trop_mam3/m_rxt_id.F90
M components/cam/src/chemistry/pp_trop_mam3/mo_phtadj.F90
M components/cam/src/chemistry/pp_trop_mam3/mo_nln_matrix.F90
M components/cam/src/chemistry/pp_trop_mam3/mo_lu_factor.F90
M components/cam/src/chemistry/pp_trop_mam3/mo_sim_dat.F90
M components/cam/src/chemistry/pp_trop_mam3/mo_indprd.F90
M components/cam/src/chemistry/pp_trop_mam3/m_spc_id.F90
M components/cam/src/chemistry/pp_trop_mam3/mo_lu_solve.F90
M components/cam/src/chemistry/pp_trop_mam3/mo_setrxt.F90
M components/cam/src/chemistry/pp_trop_mam4/m_rxt_id.F90
M components/cam/src/chemistry/pp_trop_mam4/mo_phtadj.F90
M components/cam/src/chemistry/pp_trop_mam4/mo_nln_matrix.F90
M components/cam/src/chemistry/pp_trop_mam4/mo_lu_factor.F90
M components/cam/src/chemistry/pp_trop_mam4/mo_sim_dat.F90
M components/cam/src/chemistry/pp_trop_mam4/mo_indprd.F90
M components/cam/src/chemistry/pp_trop_mam4/m_spc_id.F90
M components/cam/src/chemistry/pp_trop_mam4/mo_lu_solve.F90
M components/cam/src/chemistry/pp_trop_mam4/mo_setrxt.F90
M components/cam/src/chemistry/pp_trop_mam4/mo_adjrxt.F90
M components/cam/src/chemistry/pp_trop_mam4/mo_imp_sol.F90
M components/cam/src/chemistry/pp_trop_mam4/chem_mods.F90
M components/cam/src/chemistry/pp_trop_mam4/mo_prod_loss.F90
M components/cam/src/chemistry/pp_trop_mam4/mo_lin_matrix.F90
M components/cam/src/chemistry/pp_trop_strat_soa/m_rxt_id.F90
M components/cam/src/chemistry/pp_trop_strat_soa/mo_phtadj.F90
M components/cam/src/chemistry/pp_trop_strat_soa/mo_nln_matrix.F90
M components/cam/src/chemistry/pp_trop_strat_soa/mo_lu_factor.F90
M components/cam/src/chemistry/pp_trop_strat_soa/mo_sim_dat.F90
M components/cam/src/chemistry/pp_trop_strat_soa/mo_indprd.F90
M components/cam/src/chemistry/pp_trop_strat_soa/m_spc_id.F90
M components/cam/src/chemistry/pp_trop_strat_soa/mo_lu_solve.F90
M components/cam/src/chemistry/pp_trop_strat_soa/mo_setrxt.F90
M components/cam/src/chemistry/pp_trop_strat_soa/mo_adjrxt.F90
M components/cam/src/chemistry/pp_trop_strat_soa/mo_imp_sol.F90
M components/cam/src/chemistry/pp_trop_strat_soa/chem_mods.F90
M components/cam/src/chemistry/pp_trop_strat_soa/mo_prod_loss.F90
M components/cam/src/chemistry/pp_trop_strat_soa/mo_lin_matrix.F90
M components/cam/src/chemistry/pp_trop_bam/mo_adjrxt.F90
M components/cam/src/chemistry/pp_trop_bam/mo_imp_sol.F90
M components/cam/src/chemistry/pp_trop_bam/chem_mods.F90
M components/cam/src/chemistry/pp_trop_bam/mo_prod_loss.F90
M components/cam/src/chemistry/pp_trop_bam/mo_lin_matrix.F90
M components/cam/src/chemistry/pp_trop_bam/m_rxt_id.F90
M components/cam/src/chemistry/pp_trop_bam/mo_phtadj.F90
M components/cam/src/chemistry/pp_trop_bam/mo_nln_matrix.F90
M components/cam/src/chemistry/pp_trop_bam/mo_lu_factor.F90
M components/cam/src/chemistry/pp_trop_bam/mo_indprd.F90
M components/cam/src/chemistry/pp_trop_bam/mo_sim_dat.F90
M components/cam/src/chemistry/pp_trop_bam/m_spc_id.F90
M components/cam/src/chemistry/pp_trop_bam/mo_lu_solve.F90
M components/cam/src/chemistry/pp_trop_bam/mo_setrxt.F90
M components/cam/src/chemistry/pp_trop_mam7/m_spc_id.F90
M components/cam/src/chemistry/pp_trop_mam7/mo_lu_solve.F90
M components/cam/src/chemistry/pp_trop_mam7/mo_setrxt.F90
M components/cam/src/chemistry/pp_trop_mam7/mo_adjrxt.F90
M components/cam/src/chemistry/pp_trop_mam7/mo_imp_sol.F90
M components/cam/src/chemistry/pp_trop_mam7/chem_mods.F90
M components/cam/src/chemistry/pp_trop_mam7/mo_prod_loss.F90
M components/cam/src/chemistry/pp_trop_mam7/mo_lin_matrix.F90
M components/cam/src/chemistry/pp_trop_mam7/m_rxt_id.F90
M components/cam/src/chemistry/pp_trop_mam7/mo_phtadj.F90
M components/cam/src/chemistry/pp_trop_mam7/mo_nln_matrix.F90
M components/cam/src/chemistry/pp_trop_mam7/mo_lu_factor.F90
M components/cam/src/chemistry/pp_trop_mam7/mo_sim_dat.F90
M components/cam/src/chemistry/pp_trop_mam7/mo_indprd.F90
M components/cam/src/chemistry/pp_waccm_ma/mo_imp_sol.F90
M components/cam/src/chemistry/pp_waccm_ma/chem_mods.F90
M components/cam/src/chemistry/pp_waccm_ma/mo_prod_loss.F90
M components/cam/src/chemistry/pp_waccm_ma/mo_lin_matrix.F90
M components/cam/src/chemistry/pp_waccm_ma/m_rxt_id.F90
M components/cam/src/chemistry/pp_waccm_ma/mo_phtadj.F90
M components/cam/src/chemistry/pp_waccm_ma/mo_nln_matrix.F90
M components/cam/src/chemistry/pp_waccm_ma/mo_lu_factor.F90
M components/cam/src/chemistry/pp_waccm_ma/mo_sim_dat.F90
M components/cam/src/chemistry/pp_waccm_ma/mo_indprd.F90
M components/cam/src/chemistry/pp_waccm_ma/m_spc_id.F90
M components/cam/src/chemistry/pp_waccm_ma/mo_lu_solve.F90
M components/cam/src/chemistry/pp_waccm_ma/mo_setrxt.F90
M components/cam/src/chemistry/pp_waccm_ma/mo_adjrxt.F90
M components/cam/src/chemistry/pp_waccm_ma_sulfur/mo_lu_solve.F90
M components/cam/src/chemistry/pp_waccm_ma_sulfur/mo_setrxt.F90
M components/cam/src/chemistry/pp_waccm_ma_sulfur/mo_adjrxt.F90
M components/cam/src/chemistry/pp_waccm_ma_sulfur/mo_imp_sol.F90
M components/cam/src/chemistry/pp_waccm_ma_sulfur/chem_mods.F90
M components/cam/src/chemistry/pp_waccm_ma_sulfur/mo_prod_loss.F90
M components/cam/src/chemistry/pp_waccm_ma_sulfur/mo_lin_matrix.F90
M components/cam/src/chemistry/pp_waccm_ma_sulfur/m_rxt_id.F90
M components/cam/src/chemistry/pp_waccm_ma_sulfur/mo_phtadj.F90
M components/cam/src/chemistry/pp_waccm_ma_sulfur/mo_nln_matrix.F90
M components/cam/src/chemistry/pp_waccm_ma_sulfur/mo_lu_factor.F90
M components/cam/src/chemistry/pp_waccm_ma_sulfur/mo_sim_dat.F90
M components/cam/src/chemistry/pp_waccm_ma_sulfur/mo_indprd.F90
M components/cam/src/chemistry/pp_waccm_ma_sulfur/m_spc_id.F90
M components/cam/src/chemistry/pp_trop_ghg/m_rxt_id.F90
M components/cam/src/chemistry/pp_trop_ghg/mo_phtadj.F90
M components/cam/src/chemistry/pp_trop_ghg/mo_nln_matrix.F90
M components/cam/src/chemistry/pp_trop_ghg/mo_lu_factor.F90
M components/cam/src/chemistry/pp_trop_ghg/mo_sim_dat.F90
M components/cam/src/chemistry/pp_trop_ghg/mo_indprd.F90
M components/cam/src/chemistry/pp_trop_ghg/m_spc_id.F90
M components/cam/src/chemistry/pp_trop_ghg/mo_lu_solve.F90
M components/cam/src/chemistry/pp_trop_ghg/mo_adjrxt.F90
M components/cam/src/chemistry/pp_trop_ghg/mo_imp_sol.F90
M components/cam/src/chemistry/pp_trop_ghg/chem_mods.F90
M components/cam/src/chemistry/pp_trop_ghg/mo_prod_loss.F90
M components/cam/src/chemistry/pp_trop_ghg/mo_lin_matrix.F90
M components/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_setrxt.F90
M components/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_adjrxt.F90
M components/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_imp_sol.F90
M components/cam/src/chemistry/pp_super_fast_llnl_mam3/chem_mods.F90
M components/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_prod_loss.F90
M components/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_lin_matrix.F90
M components/cam/src/chemistry/pp_super_fast_llnl_mam3/m_rxt_id.F90
M components/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_phtadj.F90
M components/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_nln_matrix.F90
M components/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_lu_factor.F90
M components/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_sim_dat.F90
M components/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_indprd.F90
M components/cam/src/chemistry/pp_super_fast_llnl_mam3/m_spc_id.F90
M components/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_lu_solve.F90
M components/cam/src/chemistry/pp_waccm_tsmlt/mo_lu_solve.F90
M components/cam/src/chemistry/pp_waccm_tsmlt/mo_setrxt.F90
M components/cam/src/chemistry/pp_waccm_tsmlt/mo_adjrxt.F90
M components/cam/src/chemistry/pp_waccm_tsmlt/mo_imp_sol.F90
M components/cam/src/chemistry/pp_waccm_tsmlt/chem_mods.F90
M components/cam/src/chemistry/pp_waccm_tsmlt/mo_prod_loss.F90
M components/cam/src/chemistry/pp_waccm_tsmlt/mo_lin_matrix.F90
M components/cam/src/chemistry/pp_waccm_tsmlt/m_rxt_id.F90
M components/cam/src/chemistry/pp_waccm_tsmlt/mo_phtadj.F90
M components/cam/src/chemistry/pp_waccm_tsmlt/mo_nln_matrix.F90
M components/cam/src/chemistry/pp_waccm_tsmlt/mo_lu_factor.F90
M components/cam/src/chemistry/pp_waccm_tsmlt/mo_sim_dat.F90
M components/cam/src/chemistry/pp_waccm_tsmlt/mo_indprd.F90
M components/cam/src/chemistry/pp_waccm_tsmlt/m_spc_id.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_sim_dat.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_indprd.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_sulfur/m_spc_id.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_lu_solve.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_setrxt.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_adjrxt.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_imp_sol.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_sulfur/chem_mods.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_prod_loss.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_lin_matrix.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_sulfur/m_rxt_id.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_phtadj.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_nln_matrix.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_lu_factor.F90
M components/cam/src/chemistry/pp_trop_strat_mam3/mo_sim_dat.F90
M components/cam/src/chemistry/pp_trop_strat_mam3/mo_indprd.F90
M components/cam/src/chemistry/pp_trop_strat_mam3/m_spc_id.F90
M components/cam/src/chemistry/pp_trop_strat_mam3/mo_lu_solve.F90
M components/cam/src/chemistry/pp_trop_strat_mam3/mo_setrxt.F90
M components/cam/src/chemistry/pp_trop_strat_mam3/mo_adjrxt.F90
M components/cam/src/chemistry/pp_trop_strat_mam3/mo_imp_sol.F90
M components/cam/src/chemistry/pp_trop_strat_mam3/chem_mods.F90
M components/cam/src/chemistry/pp_trop_strat_mam3/mo_prod_loss.F90
M components/cam/src/chemistry/pp_trop_strat_mam3/mo_lin_matrix.F90
M components/cam/src/chemistry/pp_trop_strat_mam3/m_rxt_id.F90
M components/cam/src/chemistry/pp_trop_strat_mam3/mo_phtadj.F90
M components/cam/src/chemistry/pp_trop_strat_mam3/mo_nln_matrix.F90
M components/cam/src/chemistry/pp_trop_strat_mam3/mo_lu_factor.F90
M components/cam/src/chemistry/pp_waccm_sc/mo_imp_sol.F90
M components/cam/src/chemistry/pp_waccm_sc/chem_mods.F90
M components/cam/src/chemistry/pp_waccm_sc/mo_prod_loss.F90
M components/cam/src/chemistry/pp_waccm_sc/mo_lin_matrix.F90
M components/cam/src/chemistry/pp_waccm_sc/m_rxt_id.F90
M components/cam/src/chemistry/pp_waccm_sc/mo_phtadj.F90
M components/cam/src/chemistry/pp_waccm_sc/mo_nln_matrix.F90
M components/cam/src/chemistry/pp_waccm_sc/mo_lu_factor.F90
M components/cam/src/chemistry/pp_waccm_sc/mo_sim_dat.F90
M components/cam/src/chemistry/pp_waccm_sc/mo_indprd.F90
M components/cam/src/chemistry/pp_waccm_sc/m_spc_id.F90
M components/cam/src/chemistry/pp_waccm_sc/mo_lu_solve.F90
M components/cam/src/chemistry/pp_waccm_sc/mo_adjrxt.F90
M components/cam/src/chemistry/pp_trop_mozart/mo_lu_solve.F90
M components/cam/src/chemistry/pp_trop_mozart/mo_setrxt.F90
M components/cam/src/chemistry/pp_trop_mozart/mo_adjrxt.F90
M components/cam/src/chemistry/pp_trop_mozart/mo_imp_sol.F90
M components/cam/src/chemistry/pp_trop_mozart/chem_mods.F90
M components/cam/src/chemistry/pp_trop_mozart/mo_prod_loss.F90
M components/cam/src/chemistry/pp_trop_mozart/mo_lin_matrix.F90
M components/cam/src/chemistry/pp_trop_mozart/m_rxt_id.F90
M components/cam/src/chemistry/pp_trop_mozart/mo_phtadj.F90
M components/cam/src/chemistry/pp_trop_mozart/mo_nln_matrix.F90
M components/cam/src/chemistry/pp_trop_mozart/mo_lu_factor.F90
M components/cam/src/chemistry/pp_trop_mozart/mo_indprd.F90
M components/cam/src/chemistry/pp_trop_mozart/mo_sim_dat.F90
M components/cam/src/chemistry/pp_trop_mozart/m_spc_id.F90
M components/cam/src/chemistry/pp_trop_mozart_soa/m_rxt_id.F90
M components/cam/src/chemistry/pp_trop_mozart_soa/mo_phtadj.F90
M components/cam/src/chemistry/pp_trop_mozart_soa/mo_nln_matrix.F90
M components/cam/src/chemistry/pp_trop_mozart_soa/mo_lu_factor.F90
M components/cam/src/chemistry/pp_trop_mozart_soa/mo_indprd.F90
M components/cam/src/chemistry/pp_trop_mozart_soa/mo_sim_dat.F90
M components/cam/src/chemistry/pp_trop_mozart_soa/m_spc_id.F90
M components/cam/src/chemistry/pp_trop_mozart_soa/mo_lu_solve.F90
M components/cam/src/chemistry/pp_trop_mozart_soa/mo_setrxt.F90
M components/cam/src/chemistry/pp_trop_mozart_soa/mo_adjrxt.F90
M components/cam/src/chemistry/pp_trop_mozart_soa/mo_imp_sol.F90
M components/cam/src/chemistry/pp_trop_mozart_soa/chem_mods.F90
M components/cam/src/chemistry/pp_trop_mozart_soa/mo_prod_loss.F90
M components/cam/src/chemistry/pp_trop_mozart_soa/mo_lin_matrix.F90
M components/cam/src/chemistry/pp_trop_strat_mam7/mo_indprd.F90
M components/cam/src/chemistry/pp_trop_strat_mam7/mo_sim_dat.F90
M components/cam/src/chemistry/pp_trop_strat_mam7/m_spc_id.F90
M components/cam/src/chemistry/pp_trop_strat_mam7/mo_lu_solve.F90
M components/cam/src/chemistry/pp_trop_strat_mam7/mo_setrxt.F90
M components/cam/src/chemistry/pp_trop_strat_mam7/mo_adjrxt.F90
M components/cam/src/chemistry/pp_trop_strat_mam7/mo_imp_sol.F90
M components/cam/src/chemistry/pp_trop_strat_mam7/chem_mods.F90
M components/cam/src/chemistry/pp_trop_strat_mam7/mo_prod_loss.F90
M components/cam/src/chemistry/pp_trop_strat_mam7/mo_lin_matrix.F90
M components/cam/src/chemistry/pp_trop_strat_mam7/m_rxt_id.F90
M components/cam/src/chemistry/pp_trop_strat_mam7/mo_phtadj.F90
M components/cam/src/chemistry/pp_trop_strat_mam7/mo_nln_matrix.F90
M components/cam/src/chemistry/pp_trop_strat_mam7/mo_lu_factor.F90
M components/cam/src/chemistry/pp_terminator/mo_imp_sol.F90
M components/cam/src/chemistry/pp_terminator/chem_mods.F90
M components/cam/src/chemistry/pp_terminator/mo_prod_loss.F90
M components/cam/src/chemistry/pp_terminator/mo_lin_matrix.F90
M components/cam/src/chemistry/pp_terminator/m_rxt_id.F90
M components/cam/src/chemistry/pp_terminator/mo_phtadj.F90
M components/cam/src/chemistry/pp_terminator/mo_nln_matrix.F90
M components/cam/src/chemistry/pp_terminator/mo_lu_factor.F90
M components/cam/src/chemistry/pp_terminator/mo_sim_dat.F90
M components/cam/src/chemistry/pp_terminator/mo_indprd.F90
M components/cam/src/chemistry/pp_terminator/m_spc_id.F90
M components/cam/src/chemistry/pp_terminator/mo_lu_solve.F90
M components/cam/src/chemistry/pp_terminator/mo_adjrxt.F90
M components/cam/src/chemistry/pp_waccm_ma_mam4/mo_lin_matrix.F90
M components/cam/src/chemistry/pp_waccm_ma_mam4/m_rxt_id.F90
M components/cam/src/chemistry/pp_waccm_ma_mam4/mo_phtadj.F90
M components/cam/src/chemistry/pp_waccm_ma_mam4/mo_nln_matrix.F90
M components/cam/src/chemistry/pp_waccm_ma_mam4/mo_lu_factor.F90
M components/cam/src/chemistry/pp_waccm_ma_mam4/mo_sim_dat.F90
M components/cam/src/chemistry/pp_waccm_ma_mam4/mo_indprd.F90
M components/cam/src/chemistry/pp_waccm_ma_mam4/m_spc_id.F90
M components/cam/src/chemistry/pp_waccm_ma_mam4/mo_lu_solve.F90
M components/cam/src/chemistry/pp_waccm_ma_mam4/mo_setrxt.F90
M components/cam/src/chemistry/pp_waccm_ma_mam4/mo_adjrxt.F90
M components/cam/src/chemistry/pp_waccm_ma_mam4/mo_imp_sol.F90
M components/cam/src/chemistry/pp_waccm_ma_mam4/chem_mods.F90
M components/cam/src/chemistry/pp_waccm_ma_mam4/mo_prod_loss.F90
M components/cam/src/chemistry/pp_waccm_mad_mam4/mo_nln_matrix.F90
M components/cam/src/chemistry/pp_waccm_mad_mam4/mo_lu_factor.F90
M components/cam/src/chemistry/pp_waccm_mad_mam4/mo_sim_dat.F90
M components/cam/src/chemistry/pp_waccm_mad_mam4/mo_indprd.F90
M components/cam/src/chemistry/pp_waccm_mad_mam4/m_spc_id.F90
M components/cam/src/chemistry/pp_waccm_mad_mam4/mo_lu_solve.F90
M components/cam/src/chemistry/pp_waccm_mad_mam4/mo_setrxt.F90
M components/cam/src/chemistry/pp_waccm_mad_mam4/mo_adjrxt.F90
M components/cam/src/chemistry/pp_waccm_mad_mam4/mo_imp_sol.F90
M components/cam/src/chemistry/pp_waccm_mad_mam4/chem_mods.F90
M components/cam/src/chemistry/pp_waccm_mad_mam4/mo_prod_loss.F90
M components/cam/src/chemistry/pp_waccm_mad_mam4/mo_lin_matrix.F90
M components/cam/src/chemistry/pp_waccm_mad_mam4/m_rxt_id.F90
M components/cam/src/chemistry/pp_waccm_mad_mam4/mo_phtadj.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/m_spc_id.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_lu_solve.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_setrxt.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_adjrxt.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_imp_sol.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/chem_mods.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_prod_loss.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_lin_matrix.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/m_rxt_id.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_phtadj.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_nln_matrix.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_lu_factor.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_sim_dat.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_indprd.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_sim_dat.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_indprd.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/m_spc_id.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/chem_mech.doc
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/chem_mech.in
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_lu_solve.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_setrxt.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_adjrxt.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_rxt_rates_conv.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_imp_sol.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/chem_mods.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_prod_loss.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_lin_matrix.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_phtadj.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/m_rxt_id.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_nln_matrix.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_lu_factor.F90
- all of the chemistry mechanism needed to be re-preprocessed due to changes in
the interface between chemdr and the chemical solvers
M components/cam/test/system/config_files/f1.9c5.4wtclbh
M components/cam/test/system/config_files/f1.9c5.4wtclbdh
M components/cam/test/system/config_files/f1.9c5.5wtmam4h
M components/cam/test/system/config_files/fsd1.9c5.4wtclbh
- use a reduced pcol settings (max number of columns in each physics chunk) to
mitigate issues with memory footprint in the vectorized implicit solver used
in the waccm_tsmlt_mam4 chemistry package
M components/cam/test/system/input_tests_master
M components/cam/bld/namelist_files/use_cases/sd_2007-2011_htap_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/1850_waccm_tsmlt_cam5.xml
M components/cam/bld/namelist_files/use_cases/2007-2011_htap_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/1950-2010_ccmi_refc1_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/2004-2100_ccmi_refc2_rcp45_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/waccm_tsmlt_1850_cam55.xml
M components/cam/bld/namelist_files/use_cases/waccmx_2000_cam4.xml
M components/cam/bld/namelist_files/use_cases/volc_waccm_ma_cam5.xml
M components/cam/bld/namelist_files/use_cases/sd_waccm5_geos5.xml
M components/cam/bld/namelist_files/use_cases/volc_waccm_tsmlt_nomegan_cam5.xml
M components/cam/bld/namelist_files/use_cases/1950-2010_ccmi_refc1_waccm_ma.xml
M components/cam/bld/namelist_files/use_cases/sd_1975-2010_ccmi_refc1_waccm_tsmlt.xml
M components/cam/bld/namelist_files/use_cases/1950-2010_ccmi_refc1_waccm_tsmlt.xml
M components/cam/bld/namelist_files/use_cases/waccm_tsmlt_2000_cam55.xml
M components/cam/bld/namelist_files/use_cases/volc_waccm_tsmlt_megan_cam5.xml
M components/cam/bld/namelist_files/use_cases/sd_1975-2010_ccmi_refc1_waccm_ma.xml
M components/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_waccm_tsmlt.xml
M components/cam/bld/namelist_files/use_cases/2000_ccmi_refc1_waccm_tsmlt.xml
M components/cam/bld/namelist_files/use_cases/waccmxi_2000_cam4.xml
M components/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_rcp85_waccm_ma.xml
M components/cam/bld/namelist_files/use_cases/1979-2010_sd_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/sd_1975-2010_ccmi_refc1_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/waccm_tsmlt_amip_cam55.xml
M components/cam/bld/namelist_files/use_cases/rcp6.0_waccm_tsmlt_nomegan_cam5.xml
M components/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_waccm_ma.xml
M components/cam/bld/namelist_files/use_cases/waccm_1974-2005_cam4.xml
M components/cam/bld/namelist_files/use_cases/waccm_1850_cam4.xml
M components/cam/bld/namelist_files/use_cases/waccmx_1996_cam4.xml
M components/cam/bld/namelist_files/use_cases/waccm_ma_2000_cam55.xml
M components/cam/bld/namelist_files/use_cases/waccm_2000_cam4.xml
M components/cam/bld/namelist_files/use_cases/waccm_1850-2005_cam4.xml
M components/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp45.xml
M components/cam/bld/namelist_files/master_aer_drydep_list.xml
M components/cam/bld/namelist_files/master_aer_wetdep_list.xml
M components/cam/bld/namelist_files/master_gas_drydep_list.xml
M components/cam/bld/namelist_files/master_gas_wetdep_list.xml
- mods needed for the VBS scheme and updated TSMLT chemistry
M components/cam/cime_config/config_pes.xml
- removed PER_NODE settting for yellowstone 2-degree FV that did not make sense
- added default PE layout for waccm
M components/cam/cime_config/config_compsets.xml
M components/cam/cime_config/testdefs/testlist_cam.xml
M components/cam/cime_config/config_component.xml
M components/cam/SVN_EXTERNAL_DIRECTORIES
- use chem preprocessor capable of providing vectorized solvers
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
td.183021.status:022 bl373 TBL.sh f1.9c4mozdh outfrq3s+2000_cam4_trop_chem 9s ..................................FAIL! rc= 7 at Fri Oct 21 02:24:34 MDT 2016
td.183021.status:027 bl485 TBL.sh f1.9c5.5wtmam4h outfrq3s_megan+waccm_tsmlt_2000_cam55 9s .....................FAIL! rc= 7 at Fri Oct 21 02:51:37 MDT 2016
td.183021.status:030 bl390 TBL.sh f1.9c4wmxh outfrq3s+waccmx_2000_cam4 9s ......................................FAIL! rc= 7 at Fri Oct 21 03:01:04 MDT 2016
td.183021.status:039 bl429 TBL.sh f1.9c4wmxh outfrq3s+waccmxi_2000_cam4 9s .....................................FAIL! rc= 7 at Fri Oct 21 03:23:15 MDT 2016
td.183021.status:041 bl430 TBL.sh f1.9c5mzmam3dh outfrq3s_NEUwetdep 9s .........................................FAIL! rc= 7 at Fri Oct 21 03:32:27 MDT 2016
td.183021.status:044 bl440 TBL.sh f1.9c4soadh outfrq3s+soa_chem_megan_emis 9s ..................................FAIL! rc= 7 at Fri Oct 21 03:50:41 MDT 2016
td.183021.status:047 bl470 TBL.sh fsd1.9c5.4wtclbh outfrq3s_sd+sd_waccm5_geos5 9s ..............................FAIL! rc= 7 at Fri Oct 21 04:15:26 MDT 2016
td.183021.status:050 bl482 TBL.sh f1.9c5.5wscmam4dh outfrq3s+waccm_sc_amip_cam55 9s ............................FAIL! rc= 7 at Fri Oct 21 04:54:22 MDT 2016
These are expected baseline failures due to:
. change in tropopause definition
. chemistry is updated in waccm_tsmlt_mam4 mechanism
. change in how SAD in chemistry is calculated
. adjustment of the EUV heating factor in WACCM(X)
yellowstone/intel/aux_cam:
ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s_clm5 (Overall: NLFAIL), details:
FAIL NLCOMP
ERP_Ln9.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq9s_clm5 (Overall: NLFAIL), details:
FAIL NLCOMP
SMS_Lm13.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq1m_clm5 (Overall: NLFAIL), details:
FAIL NLCOMP
Baseline tests passed. The namelist compare tests failed due to new namelist options added.
hobart/nag:
073 bl426 TBL.sh f10c4wmxdm outfrq3s_ionos 9s .................................................FAIL! rc= 7 at Thu Oct 20 20:05:16 MDT 2016
076 bl471 TBL.sh f10c5.5wmam4dm outfrq3s_modalstrat 9s ........................................FAIL! rc= 7 at Thu Oct 20 20:13:12 MDT 2016
These are expected baseline failures due to:
. chemistry is updated in waccm_tsmlt_mam4 mechanism
. change in how SAD in chemistry is calculated
. adjustment of the EUV heating factor in WACCM(X)
hobart/pgi: All pass
CAM tag used for the baseline comparison tests if different than previous
tag: NA
===============================================================
===============================================================
Tag name: cam5_4_90
Originator(s): fvitt, bardeenc
Date: 14 Oct 2016
One-line Summary: Added options to improve water vapor and ice clouds in the tropical UTLS and polar stratosphere
Purpose of changes:
- Provide namelist options to enable changes that have been shown to improve
stratospheric water vapor and surface area density in WACCM
- Improve water vapor and ice clouds in UTLS and polar stratosphere
- Provide a more consistent treatment of subgrid supersaturation for growth
and nucleation of ice clouds
- Provide some additional diagnostic output related to processes in the UTLS
and stratosphere
Note:
This leaves the default CAM & WACCM behavior unchanged (bit-for-bit)
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/bld/namelist_files/namelist_definition.xml
- new namelist options added:
nucleate_ice_subgrid_strat
nucleate_ice_use_troplev
cldfrc2m_rhminis
cldfrc2m_rhmaxis
M components/cam/bld/build-namelist
- set defaults for:
micro_mg_adjust_cpt
nucleate_ice_subgrid_strat
nucleate_ice_use_troplev
cldfrc2m_rhminis
cldfrc2m_rhmaxis
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
- define defaults for:
micro_mg_adjust_cpt
nucleate_ice_subgrid_strat
nucleate_ice_use_troplev
cldfrc2m_rhminis
cldfrc2m_rhmaxis
M components/cam/src/physics/cam/cldfrc2m.F90
M components/cam/src/physics/cam/cldwat2m_macro.F90
M components/cam/src/physics/cam/clubb_intr.F90
Add namelist variables (cldfrc2m_rhminis, cldfrc2m_rhmaxis) to allow rhmini and
rhmaxi to be configured differently in the stratosphere and troposphere. By default,
they will be configured the same.
M components/cam/src/physics/cam/clubb_intr.F90
Add the vertical diffusivity from CLUBB (KVH_CLUBB) as an output field. Also fixed
compiler issue with output of RTPHLP_CLUBB (from Francis).
M components/cam/src/physics/cam/hetfrz_classnuc_cam.F90
Initialize some variables to 0._r8 as they were not being explicitly initialized before.
M components/cam/src/physics/cam/micro_mg_cam.F90
Add the ability to adjust the cold point temperature used by the microphysics based
upon interpolating the temperature profile between model levels. This is to account
for the model not being able to resolve the cold point which should determine entry
value for stratospheric water vapor. This feature is disabled by default.
M components/cam/src/physics/cam/nucleate_ice.F90
M components/cam/src/physics/cam/nucleate_ice_cam.F90
Via namelist variables, allow the subgrid nucleation (nucleate_ice_subgrid) to be set
to the same value as used for subgrid growth (1/QSATFAC) rather than be set to a
fixed value. Also allow the subgrid value to be set differently in the troposphere and
the stratosphere (nucleate_ice_subgrid_strat). Add a namelist variable
(nucleate_ice_troplev) to control whether the chemical tropopause or hard coded
pressure levels are used to define where stratospheric ice is nucleated. Using the
chemical tropopause definition also adds the coarse and accumulation mode
number to the total sulfate concentration used for nucleation. Add more diagnostic
output (NISUBGRID, NITROP_PD). By default, fixed pressure levels are used, the
troposphere and stratosphere are configured the same, and the subgrid value for
nucleation is set to a fixed value of 1.2
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all pass
yellowstone/intel/aux_cam:
ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s_clm5 (Overall: NLFAIL), details:
PASS CREATE_NEWCASE
PASS XML
FAIL NLCOMP
PASS SHAREDLIB_BUILD
PASS MODEL_BUILD
PASS RUN
PASS COMPARE_base_rest
PASS GENERATE
PASS COMPARE_baseline
PASS TPUTCOMP
PASS MEMLEAK
ERP_Ln9.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq9s_clm5 (Overall: NLFAIL), details:
PASS CREATE_NEWCASE
PASS XML
FAIL NLCOMP
PASS SHAREDLIB_BUILD
PASS MODEL_BUILD
PASS RUN
PASS COMPARE_base_rest
PASS GENERATE
PASS COMPARE_baseline
PASS TPUTCOMP
PASS MEMLEAK
SMS_Lm13.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq1m_clm5 (Overall: NLFAIL), details:
PASS CREATE_NEWCASE
PASS XML
FAIL NLCOMP
PASS SHAREDLIB_BUILD
PASS MODEL_BUILD
PASS RUN
PASS GENERATE
PASS COMPARE_baseline
PASS MEMCOMP
PASS TPUTCOMP
PASS MEMLEAK
hobart/nag: all pass
hobart/pgi: all pass
===============================================================
===============================================================
Tag name: cam5_4_89
Originator(s): cacraig, hannay, bogensch, mmills
Date: Oct 10, 2016
One-line Summary: Phase1b of the CAM changes Cecile is currently using for CESM runs
Purpose of changes:
** WARNING ** The cime external in this tag has an error when running the short_term archiver.
Manual resubmits are required. The cime external will be updated in a future
tag.
- F2000_DEV replaces F2000_DONOTUSE compset as this compset is ready for general use. It should
be pointed out that this compset will be a "floating" compset and will follow the current
CAM development settings and will be updated as needed.
- The "chemical tropopause" is defined using the method tropopause_findChemTrop. Between
50N and 50S, this is the lapse rate tropopause backed up by the climatology. For high
latitudes (>50), a fixed pressure level (300 hPa) is used because the lapse rate tropopause
sometimes finds the tropopause at high altitudes. The chemical tropopause is used as the
boundary between tropospheric and stratospheric chemical reaction rates. When prognostic
stratospheric aerosols are used, the chemical tropopause sets the level for differences in
volatility between stratospheric and tropospheric modal aerosols.
- A cold point tropopause definition is added as an option for defining the tropopause, using
the method TROP_ALG_CPP. History outputs for cold point tropopause are added: TROPF_P,
TROPF_T, TROPF_Z, TROPF_DZ, TROPF_PD, and TROPF_FD.
- A function tropopause_interpolateP is added to find interpolated pressure at the specified
tropopause altitude.
- Modification to allow the highest possible level for CLUBB's water vapor tendencies to be
applied is the tropopause top, to prevent a drying out of the upper troposphere and stratosphere,
as seen in WACCM simulations. Includes an option to apply this fixer to heat tendencies. In
addition, a fix is included to prevent small energy and water conservation violations
- To get the CLM namelist changes which mimic Cecile's runs, users need to add to their create_newcase the option:
--user-mods-dir ../../../components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq1m_clm5
** NOTE ** This directory name will be changing in a future tag
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: cacraig
List all subroutines eliminated:
List all subroutines added and what they do:
A components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq1m_clm5
A components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq1m_clm5/shell_commands
A components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq1m_clm5/user_nl_cam
A components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq1m_clm5/user_nl_clm
- add settings for outputting monthly
- add spunup CLM initial condition file
A components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq9s_clm5
A components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq9s_clm5/shell_commands
A components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq9s_clm5/user_nl_cam
A components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq9s_clm5/user_nl_clm
- add spunup CLM initial condition file
List all existing files that have been modified, and describe the changes:
M components/cam/bld/namelist_files/use_cases/2000_cam6.xml
- added ext_frc_specifier files to match what Cecile is currently using
M components/cam/cime_config/config_compsets.xml
M components/cam/cime_config/testdefs/testlist_cam.xml
- F2000_DONOTUSE now becomes F2000_DEV as Cecile has confirmed that this
compset now matches what she is using for development
- add test for 13 month simulation for F2000_DEV
- use spunup CLM finidat file for CLM5 runs
M components/cam/cime_config/config_pes.xml
- Turn off hyperthreading for all F compsets by default (seeing improved
perfomance with hyperthreading turned off)
- 1degree FV F compsets now use 720 tasks
M components/cam/src/physics/cam/clubb_intr.F90
- Modification to prevent a drying out of the upper troposphere and stratosphere
A fix is included to prevent small energy and water conservation violations
M components/cam/src/physics/cam/tropopause.F90
- changes to tropopause detailed extensively above.
M components/cam/test/system/test_driver.sh
- add a warning if don't supply a CAM_TAG for generating the basline
- echo the ./create_test options to the log file also
- increase the wall clock time to 3 hours for all CESM jobs until update
to the new testdefs formatted files (when update to beta02 externals)
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB except:
015 bl357 TBL.sh f1.9c5.5h outfrq3s+1850_cam55 9s .............................................FAIL! rc= 7 at Fri Oct 7 19:10:07 MDT 2016
018 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Fri Oct 7 19:22:03 MDT 2016
027 bl485 TBL.sh f1.9c5.5wtmam4h outfrq3s_megan+waccm_tsmlt_2000_cam55 9s .....................FAIL! rc= 7 at Fri Oct 7 20:00:45 MDT 2016
047 bl470 TBL.sh fsd1.9c5.4wtclbh outfrq3s_sd+sd_waccm5_geos5 9s ..............................FAIL! rc= 7 at Fri Oct 7 21:18:53 MDT 2016
yellowstone/intel/aux_cam:
ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s (Overall: DIFF), details:
FAIL NLCOMP
FAIL COMPARE_baseline
ERP_Ln9.f09_f09.F2000_DEV.yellowstone_intel.cam-outfrq9s (Overall: DIFF), details:
FAIL NLCOMP
FAIL COMPARE_baseline
-- Namelist fails due to changes to match Cecile's configuration
Baseline comparison fails due to answer changes in CLUBB
hobart/nag: All BFB except:
058 bl356 TBL.sh fsd1.9c5clbdm off1.9x2.5 9s ..................................................FAIL! rc= 7 at Mon Oct 10 14:01:33 MDT 2016
064 bl369 TBL.sh f1.9c5clbdm outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Mon Oct 10 14:02:18 MDT 2016
hobart/pgi or jaguar/pgi: All BFB
-- All tests above fail due to answer changes in CLUBB
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: all tests which use CLUBB
- what platforms/compilers: all
- nature of change (roundoff; larger than roundoff but same climate; new
climate): impact of phase1a and phase1b is new climate
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
https://svn-ccsm-models.cgd.ucar.edu/cam1/branch_tags/cam5_4_75_phase1b_tags/cam5_4_75_phase1b_n05_cam5_4_88
Cecile also added changes for CLM namelist settings which were not in this branch tag. These CLM changes
are in cam5_4_89. A 13 month run was compared against Cecile's run at 13 months and the files were identical
(except at the 7-8th digit as her run was storing single precision and the trunk run was storing double).
- platform/compilers: yellowstone intel
- configure commandline: Using the F2000_Dev compset
- build-namelist command (or complete namelist): the use case automatically supplied with F2000_Dev
- MSS location of output: not stored on HPSS
URL for AMWG diagnostics output used to validate new climate:
http://webext.cgd.ucar.edu/FCLIMO/f.e15.F2000_DEV.f09_f09.cam5_4_75_phase1b_n05_cam5_4_88_land_ic/atm/f.e15.F2000_DEV.f09_f09.cam5_4_75_phase1b_n05_cam5_4_88_land_ic-f.e15.F2000.f09_f09.pd.114_cam5_4_75/
===============================================================
===============================================================
Tag name: cam5_4_88
Originator(s): cacraig, juliob, hannay, bardeenc
Date: Oct 5, 2016
One-line Summary: Phase1a of the CAM changes Cecile is currently using for CESM runs
Purpose of changes:
- Replacing analytical kinetic energy change with diagnosed kinetic energy
change across diffusion scheme for beljaars runs.
- Fixes a bug in MG2 in calculating the fall velocity for small ice particles and moves
the threshold for using this fall velocity from 8um to 18um. This allows the fall
velocities for the small particles to blend more smoothly with the fall velocity
used for larger particles.
- The wrong versions of the nucleate_ice files were included in cam5_4_63. This corrected
version affects how ice nucleation is calculated when cloud fraction is less than 1 and
how the use_incloud_nuc namelist variable is applied.
Bugs fixed (include bugzilla ID):
- MG2 fall velocity for small particles corrected
- correct versions of nucleate_ice routines
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: cacraig, fvitt
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
M components/cam/src/physics/cam/beljaars_drag_cam.F90
- allow do_beljaars to be used outside this module
M components/cam/src/physics/cam/diffusion_solver.F90
- Replacing analytical kinetic energy change with diagnosed kinetic energy
change across diffusion scheme
M components/cam/src/physics/cam/micro_mg2_0.F90
- Fixes a bug in calculating the fall velocity for small ice particles and moves
the threshold for using this fall velocity from 8um to 18um. This allows the fall
velocities for the small particles to blend more smoothly with the fall velocity
used for larger particles.
M components/cam/src/physics/cam/nucleate_ice.F90
M components/cam/src/physics/cam/nucleate_ice_cam.F90
- The wrong versions of these files were included in cam5_4_63. This corrected
version affects how ice nucleation is calculated when cloud fraction is less than 1 and
how the use_incloud_nuc namelist variable is applied.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
011 bl353 TBL.sh f1.9c5.4dh outfrq3s 9s .......................................................FAIL! rc= 7 at Tue Oct 4 20:35:19 MDT 2016
015 bl357 TBL.sh f1.9c5.5h outfrq3s+1850_cam55 9s .............................................FAIL! rc= 7 at Tue Oct 4 20:40:28 MDT 2016
018 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Tue Oct 4 20:52:21 MDT 2016
027 bl485 TBL.sh f1.9c5.5wtmam4h outfrq3s_megan+waccm_tsmlt_2000_cam55 9s .....................FAIL! rc= 7 at Tue Oct 4 21:30:08 MDT 2016
047 bl470 TBL.sh fsd1.9c5.4wtclbh outfrq3s_sd+sd_waccm5_geos5 9s ..............................FAIL! rc= 7 at Tue Oct 4 22:46:41 MDT 2016
yellowstone/intel/aux_cam:
ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s (Overall: DIFF), details:
FAIL COMPARE_baseline
ERP_Ln9.f09_f09.F2000_DONOTUSE.yellowstone_intel.cam-outfrq9s (Overall: DIFF), details:
FAIL COMPARE_baseline
hobart/nag:
055 bl354 TBL.sh f1.9c5.4dm outfrq3s_npryz 9s .................................................FAIL! rc= 7 at Wed Oct 5 10:34:38 MDT 2016
058 bl356 TBL.sh fsd1.9c5clbdm off1.9x2.5 9s ..................................................FAIL! rc= 7 at Wed Oct 5 10:35:06 MDT 2016
061 bl368 TBL.sh f1.9c5mg2dm outfrq3s_macmic2_classnuc 9s .....................................FAIL! rc= 7 at Wed Oct 5 10:35:20 MDT 2016
064 bl369 TBL.sh f1.9c5clbdm outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Wed Oct 5 10:35:39 MDT 2016
hobart/pgi or jaguar/pgi:
057 bl354 TBL.sh f1.9c5.4dm outfrq3s_npryz 9s .................................................FAIL! rc= 7 at Wed Oct 5 10:43:25 MDT 2016
Summarize any changes to answers, i.e.,
- what code configurations: all jobs which use mg2, do_beljaars=.true., certain conditions in nucleate_ice (see above)
- what platforms/compilers: all
These changes have been tested offline by Cecile Hannay and the validation will be combined with
an upcoming CAM tag and reported in that tag.
===============================================================
===============================================================
Tag name: cam5_4_87
Originator(s): aconley,fvitt,goldy
Date: 2016-09-30
One-line Summary: Add terminator chemistry package
Purpose of changes:
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/bld/config_files/definition.xml
- Added terminator as a valid -chem option
M components/cam/bld/configure
- Added terminator as a valid -chem option. Set proper number of tracers
A + components/cam/bld/namelist_files/use_cases/dctest_baro_moist.xml
- Beginning of use case for dycore test use case (uses terminator chem package)
M components/cam/cime_config/config_component.xml
- Added a dycore test compset CAM options (DCTBM)
M components/cam/cime_config/config_compsets.xml
- Added dycore test compset (FDCBAROMOIST)
M components/cam/cime_config/testdefs/testlist_cam.xml
- Created aux_cam_short testlist (same as aux_cam for now).
M components/cam/doc/ChangeLog
- Improve disk drive manufacturer bottom line but using more storage
M components/cam/src/chemistry/mozart/chemistry.F90
- Make fh2o a required argument
M components/cam/src/chemistry/pp_none/chemistry.F90
- Make fh2o an optional argument
A + components/cam/src/chemistry/pp_terminator
A + components/cam/src/chemistry/pp_terminator/chem_mech.doc
A + components/cam/src/chemistry/pp_terminator/chem_mech.in
A + components/cam/src/chemistry/pp_terminator/chem_mods.F90
A + components/cam/src/chemistry/pp_terminator/chemistry.F90
A + components/cam/src/chemistry/pp_terminator/m_rxt_id.F90
A + components/cam/src/chemistry/pp_terminator/m_spc_id.F90
A + components/cam/src/chemistry/pp_terminator/mo_adjrxt.F90
A + components/cam/src/chemistry/pp_terminator/mo_imp_sol.F90
A + components/cam/src/chemistry/pp_terminator/mo_indprd.F90
A + components/cam/src/chemistry/pp_terminator/mo_lin_matrix.F90
A + components/cam/src/chemistry/pp_terminator/mo_lu_factor.F90
A + components/cam/src/chemistry/pp_terminator/mo_lu_solve.F90
A + components/cam/src/chemistry/pp_terminator/mo_nln_matrix.F90
A + components/cam/src/chemistry/pp_terminator/mo_phtadj.F90
A + components/cam/src/chemistry/pp_terminator/mo_prod_loss.F90
A + components/cam/src/chemistry/pp_terminator/mo_rxt_rates_conv.F90
A + components/cam/src/chemistry/pp_terminator/mo_setrxt.F90
A + components/cam/src/chemistry/pp_terminator/mo_sim_dat.F90
- Added new terminator chemistry package
M components/cam/src/dynamics/eul/dyn_comp.F90
- Coordinates passed to cnst_init_default must be in radians
M components/cam/src/dynamics/fv/dyn_comp.F90
- Coordinates passed to cnst_init_default must be in radians
M components/cam/src/dynamics/se/dyn_comp.F90
- Coordinates passed to cnst_init_default must be in radians
M components/cam/src/dynamics/sld/dyn_comp.F90
- Coordinates passed to cnst_init_default must be in radians
M components/cam/src/physics/cam/clubb_intr.F90
- Only initialize constituents where mask is .true.
M components/cam/src/physics/cam/co2_cycle.F90
- Only initialize constituents where mask is .true.
M components/cam/src/physics/cam/const_init.F90
- Pass correct latitude array for each block in cnst_init_default
M components/cam/src/physics/cam/micro_mg_cam.F90
- Only initialize constituents where mask is .true.
M components/cam/src/physics/cam/physpkg.F90
- Show that fh2o is an optional argument (use key=value form)
M components/cam/src/physics/cam/rk_stratiform.F90
- Only initialize constituents where mask is .true.
M components/cam/src/physics/cam/tracers_suite.F90
- Only initialize constituents where mask is .true.
M components/cam/src/physics/cam/unicon_cam.F90
- Only initialize constituents where mask is .true.
M components/cam/src/physics/carma/cam/carma_intr.F90
- Only initialize constituents where mask is .true.
M components/cam/src/physics/carma/models/test_growth/carma_model_mod.F90
M components/cam/src/physics/carma/models/test_passive/carma_model_mod.F90
M components/cam/src/physics/carma/models/test_radiative/carma_model_mod.F90
M components/cam/src/physics/carma/models/test_swelling/carma_model_mod.F90
M components/cam/src/physics/carma/models/test_tracers/carma_model_mod.F90
M components/cam/src/physics/carma/models/test_tracers2/carma_model_mod.F90
- Only initialize constituents where mask is .true.
M components/cam/src/physics/simple/physpkg.F90
- Add chemistry to simple physics (chem_register, chem_init, chem_timestep_tend)
A + components/cam/test/system/config_files/f10adhterm
- New test config using adiabatic physics with terminator chemistry
M components/cam/test/system/input_tests_master
- Add new terminator chemistry test (339)
A + components/cam/test/system/nl_files/terminator
- Namelist file for terminator chemistry test
M components/cam/test/system/test_driver.sh
- Made processing of CAM regression tests and/or CESM auxiliary tests optional
- Allow alternate CESM test or test suite
- Cleaned up old comments and created usage statement with documentation
M components/cam/test/system/tests_pretag_hobart_nag
- Add new terminator chemistry test (339)
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All pass
yellowstone/intel/aux_cam: All pass
hobart/nag: Baseline fail for new test (no baseline).
052 bl339 TBL.sh f10adhterm terminator 9s .....................................................FAIL! rc= 7 at Fri Sep 30 13:49:06 MDT 2016
hobart/pgi:
CAM tag used for the baseline comparison tests if different than previous
tag: NA
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate): NA
If bitwise differences were observed, how did you show they were no worse
than roundoff? NA
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
NA
===============================================================
===============================================================
Tag name: cam5_4_86
Originator(s): cacraig, hannay, mmills, fvitt
Date: September 29, 2016
One-line Summary: Update solar files and use cases
Purpose of changes:
- Update solar files used for CAM/WACCM 1850, HIST and 2000 cases
- Add use cases which capture F and FC5 compsets (cam4 and cam5)
at this current time. This allows the namelist defaults to follow the
cam6 configuration without impacting older versions.
- Added F1850_DONOTUSE compset - this compset is VERY PRELIMINARY and
really should not be used
- Started work on the FHIST_DONOTUSE compset, but it is broken and will
be worked on for future tags.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
A components/cam/bld/namelist_files/use_cases/1850_cam6.xml
A components/cam/bld/namelist_files/use_cases/hist_cam6.xml
- With input from Cecile, we are setting up preliminary use cases for FHIST and F1850 compsets
Note that hist_cam6.xml does not run to completion yet.
List all existing files that have been modified, and describe the changes:
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
M components/cam/bld/namelist_files/use_cases/2000_cam6.xml
- update solar_data_file (CAM and WACCM) and solar_parms_file (WACCM only) to files
provided by Mike Mills
M components/cam/bld/namelist_files/use_cases/waccm_tsmlt_1850_cam55.xml
M components/cam/bld/namelist_files/use_cases/waccm_tsmlt_2000_cam55.xml
M components/cam/bld/namelist_files/use_cases/waccm_tsmlt_amip_cam55.xml
- update solar_data_file, solar_parms_file, epp_all_filepath and epp_all_varname
to settings provided by Mike Mills
M components/cam/cime_config/config_component.xml
- add hooks for hist_cam6 use case
- update the 1850_CAM6 compsets to use 1850_cam6 use case
M components/cam/cime_config/config_compsets.xml
- add F1850_DONOTUSE and FHIST_DONOTUSE compsets
M components/cam/cime_config/testdefs/testlist_cam.xml
- add aux_cam and prealpha testing for F1850_DONOTUSE
M components/cam/doc/ChangeLog
- added additional information for the cam5_4_84 tag as the bug fix for clubb_do_liqsupersat had
a larger impact than was first thought
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
011 bl353 TBL.sh f1.9c5.4dh outfrq3s 9s .......................................................FAIL! rc= 7 at Thu Sep 29 16:34:36 MDT 2016
027 bl485 TBL.sh f1.9c5.5wtmam4h outfrq3s_megan+waccm_tsmlt_2000_cam55 9s .....................FAIL! rc= 7 at Thu Sep 29 16:37:22 MDT 2016
yellowstone/intel/aux_cam:
ERP_Ln9.f09_f09.F1850_DONOTUSE.yellowstone_intel.cam-outfrq9s (Overall: DIFF), details:
FAIL NLCOMP
FAIL COMPARE_baseline
ERP_Ln9.f09_f09.F2000_DONOTUSE.yellowstone_intel.cam-outfrq9s (Overall: DIFF), details:
FAIL NLCOMP
FAIL COMPARE_baseline
** Namelist tests fail due to new solar forcing file
** F2000 baseline test fails due to new solar forcing file
** F1850 baseline test fails due to it being a new test
hobart/nag:
017 bl221 TBL.sh f10c5spcamsm outfrq3s 9s .....................................................FAIL! rc= 7 at Thu Sep 29 16:21:18 MDT 2016
052 bl354 TBL.sh f1.9c5.4dm outfrq3s_npryz 9s .................................................FAIL! rc= 7 at Thu Sep 29 16:33:18 MDT 2016
hobart/pgi or jaguar/pgi:
057 bl354 TBL.sh f1.9c5.4dm outfrq3s_npryz 9s .................................................FAIL! rc= 7 at Thu Sep 29 16:31:46 MDT 2016
All of the above tests are expected to change due to the change in the SolarForcing file (which was provided by mmills)
===============================================================
===============================================================
Tag name: cam5_4_85
Originator(s): fvitt, erik, cacraig
Date: 21 Sep 2016
One-line Summary: Update build-namelist use cases with updated MEGAN emissions factors and update CLM
Purpose of changes:
Update build-namelist use cases used in CLM50 compsets with MEGAN active
to include emission factors for 79 PFTs
Update CLM external to include MEGAN model for 79 PFTs used in clm45 and clm50
for CIME testing
Change the aux_cam CIME smoke test to a restart test (ERP)
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
When using clm45 and clm50 the megan_factors_file needs to be set to
$INPUTDATA/atm/cam/chem/trop_mozart/emis/megan21_emis_factors_79pft_c20150518.nc
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
A components/cam/test/system/nl_files/outfrq3s_megan
- overrides megan_factors_file with an appropriate input file for clm4 used in cam regression tests
List all existing files that have been modified, and describe the changes:
M SVN_EXTERNAL_DIRECTORIES
- update CLM external to esmciport_n06_clm4_5_12_r194 which has updates to
MEGAN for clm45 and clm50
M components/cam/bld/namelist_files/use_cases/waccm_tsmlt_2000_cam55.xml
M components/cam/bld/namelist_files/use_cases/volc_waccm_tsmlt_megan_cam5.xml
M components/cam/bld/namelist_files/use_cases/waccm_tsmlt_amip_cam55.xml
M components/cam/bld/namelist_files/use_cases/rcp6.0_waccm_tsmlt_nomegan_cam5.xml
M components/cam/bld/namelist_files/use_cases/waccm_tsmlt_1850_cam55.xml
M components/cam/bld/namelist_files/use_cases/sd_cam5_trop_strat_mam3.xml
- update megan_factors_file in these use cases to include 79 PFTs
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
- update the default megan_factors_file to include 79 PFTs
M components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq9s/shell_commands
M components/cam/cime_config/testdefs/testlist_cam.xml
- mods for aux_cam test -- workaround for the glc model issue when the test is less than 1 day
M components/cam/test/system/input_tests_master
- use outfrq3s_megan in waccm5.5 test
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
027 bl485 TBL.sh f1.9c5.5wtmam4h outfrq3s_megan+waccm_tsmlt_2000_cam55 9s .....................FAIL! rc= 7 at Wed Sep 21 09:00:31 MDT 2016
- this is actually bit-for-bit -- change "outfrq3s" to "outfrq3s_megan" caused this baseline to fail
yellowstone/intel/aux_cam:
PASS ERP_Ln9.f09_f09.F2000_DONOTUSE.yellowstone_intel.cam-outfrq9s CREATE_NEWCASE
PASS ERP_Ln9.f09_f09.F2000_DONOTUSE.yellowstone_intel.cam-outfrq9s XML
FAIL ERP_Ln9.f09_f09.F2000_DONOTUSE.yellowstone_intel.cam-outfrq9s NLCOMP
PASS ERP_Ln9.f09_f09.F2000_DONOTUSE.yellowstone_intel.cam-outfrq9s SHAREDLIB_BUILD time=579
PASS ERP_Ln9.f09_f09.F2000_DONOTUSE.yellowstone_intel.cam-outfrq9s MODEL_BUILD time=674
PASS ERP_Ln9.f09_f09.F2000_DONOTUSE.yellowstone_intel.cam-outfrq9s RUN time=431
PASS ERP_Ln9.f09_f09.F2000_DONOTUSE.yellowstone_intel.cam-outfrq9s COMPARE_base_rest
PASS ERP_Ln9.f09_f09.F2000_DONOTUSE.yellowstone_intel.cam-outfrq9s GENERATE
FAIL ERP_Ln9.f09_f09.F2000_DONOTUSE.yellowstone_intel.cam-outfrq9s COMPARE_baseline
PASS ERP_Ln9.f09_f09.F2000_DONOTUSE.yellowstone_intel.cam-outfrq9s MEMLEAK insuffiencient data for memleak test
- this baseline is expected to fail -- clm50 has changed answers
hobart/nag: all pass
jaguar/pgi: all pass
===============================================================
===============================================================
Tag name: cam5_4_84
Originator(s): cacraig, bogensh, hannay, fvitt
Date: Sept 19, 2016
One-line Summary: Add namelist parameter for CLUBB C14 and fix clubb_do_liqsupersat bug
Purpose of changes:
- Add C14 tuning parameter for CLUBB to the namelist. Change its default value from
1. to 1.83
- Add missing mpi_bcast for clubb_do_liqsupersat
- After additional use, it was discovered that the clubb_do_liqsupersat fix has a significant impact
Bugs fixed (include bugzilla ID):
- Fix bug which was causing CESM ERP tests using clubb to not be BFB
clubb_do_liqsupersat needed to be mpi_bcast to all MPI tasks
Describe any changes made to build system:
Describe any changes made to the namelist:
- Add namelist clubb_c14 (constant for u'^2 and v'^2 terms)
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/bld/build-namelist
M components/cam/bld/namelist_files/namelist_definition.xml
- Add clubb_c14 tuning namelist parameter
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
- Change default value for clubb_c14 from 1. to 1.83
M components/cam/src/physics/cam/clubb_intr.F90
- Add clubb_c14 tuning namelist parameter
- Fix bug where clubb_do_liqsupersat was not being broadcast to all mpi tasks
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB except:
015 bl357 TBL.sh f1.9c5.5h outfrq3s+1850_cam55 9s .............................................FAIL! rc= 7 at Fri Sep 16 17:00:48 MDT 2016
018 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Fri Sep 16 17:12:46 MDT 2016
027 bl485 TBL.sh f1.9c5.5wtmam4h outfrq3s+waccm_tsmlt_2000_cam55 9s ...........................FAIL! rc= 7 at Fri Sep 16 17:52:47 MDT 2016
047 bl470 TBL.sh fsd1.9c5.4wtclbh outfrq3s_sd+sd_waccm5_geos5 9s ..............................FAIL! rc= 7 at Fri Sep 16 19:10:06 MDT 2016
yellowstone/intel/aux_cam:
FAIL SMS_Ln9.f09_f09.F2000_DONOTUSE.yellowstone_intel.cam-outfrq9s NLCOMP
- added new namelist parameter
FAIL SMS_Ln9.f09_f09.F2000_DONOTUSE.yellowstone_intel.cam-outfrq9s COMPARE_baseline
- expect the answers to change with the tuned parameter
hobart/nag: all BFB except:
055 bl356 TBL.sh fsd1.9c5clbdm off1.9x2.5 9s ..................................................FAIL! rc= 7 at Fri Sep 16 15:13:46 MDT 2016
061 bl369 TBL.sh f1.9c5clbdm outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Fri Sep 16 15:40:51 MDT 2016
hobart/pgi or jaguar/pgi: all BFB (no CLUBB tests included in this testlist)
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: all runs which use CLUBB (including ones which use CAM5.5)
- what platforms/compilers: all
- nature of change (roundoff; larger than roundoff but same climate; new
climate): new climate
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
Case directories:
/glade/p/cesmdata/cseg/runs/cesm1_5/f.e15.F2000.f09_f09.pd.114_cam5_4_75
and:
/glade/p/cesmdata/cseg/runs/cesm1_5/f.e15.F2000.f09_f09.pd.114_cam5_4_75_C14
Output data:
/glade/p/cesm0005/archive/f.e15.F2000.f09_f09.pd.114_cam5_4_75
and
/glade/p/cesm0005/archive/f.e15.F2000.f09_f09.pd.114_cam5_4_75_C14
- MSS location of output
MSS location of control simulations used to validate new climate:
- Not stored on HPSS
URL for AMWG diagnostics output used to validate new climate:
http://webext.cgd.ucar.edu/FCLIMO/f.e15.F2000.f09_f09.pd.114_cam5_4_75/atm/f.e15.F2000.f09_f09.pd.114_cam5_4_75-f.e15.F2000.f09_f09.pd.114_cam5_4_75_C14/
URL for impact of the clubb_do_liqsupersat bug fix
http://webext.cgd.ucar.edu/FCLIMO/f.e15.F2000.f09_f09.pd.114_cam5_4_75_clubb_fix_tropopause_tend/atm/f.e15.F2000.f09_f09.pd.114_cam5_4_75_clubb_fix_tropopause_tend-f.e15.F2000.f09_f09.pd.114_cam5_4_75_clubb_fix_tropopause_tend_nompi/
===============================================================
===============================================================
Tag name: cam5_4_83
Originator(s): eaton, mvr, hannay, cacraig
Date: Sept 15, 2016
One-line Summary: radiation bugfix
Purpose of changes:
- Mods were needed due to some refactoring in cloud_rad_props.F90. The fix provided by Mat
went into subroutine interpolate_ice_optics_lw and consequently needed to be applied to both
snow and ice cloud optics and not just to the snow optics as in the original mod.
- Changes were merged in from: https://svn-ccsm-models.cgd.ucar.edu/cam1/branch_tags/rrtmg_tags/rrtmg01_cam5_4_79
These changes were identical to the ones that Cecile used in her testing
- added abs() to time_period_freq when the nhtfrq is negative (when specifying hours)
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/src/control/cam_history.F90
- added abs() to nhtfrq when it is negative
M components/cam/src/physics/cam/micro_mg_cam.F90
- replace '.lt.' by '.le.' at line 2248
M components/cam/src/physics/rrtmg/aer_src/rrtmg_lw_rtrnmc.f90
- fix the case where the diffusivity angle was going negative under very
moist conditions
- the code has been changed so the interpolation for the lw ice optics has
been factored out and is now used for for the ice and snow cloud optics.
So the mod to limit the ice optical depths is now applied to both ice and
snow clouds.
M components/cam/src/physics/rrtmg/cloud_rad_props.F90
- Limit the snow/ice optical depths fix the case where there are large masses
of snow at high altitudes
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
011 bl353 TBL.sh f1.9c5.4dh outfrq3s 9s .......................................................FAIL! rc= 7 at Tue Sep 13 17:52:12 MDT 2016
015 bl357 TBL.sh f1.9c5.5h outfrq3s+1850_cam55 9s .............................................FAIL! rc= 7 at Tue Sep 13 17:57:15 MDT 2016
018 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Tue Sep 13 18:09:18 MDT 2016
024 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...............................................FAIL! rc= 7 at Tue Sep 13 18:27:05 MDT 2016
027 bl485 TBL.sh f1.9c5.5wtmam4h outfrq3s+waccm_tsmlt_2000_cam55 9s ...........................FAIL! rc= 7 at Tue Sep 13 18:47:22 MDT 2016
036 bl421 TBL.sh f1.9c5mam4dh outfrq3s 9s .....................................................FAIL! rc= 7 at Tue Sep 13 19:09:05 MDT 2016
041 bl430 TBL.sh f1.9c5mzmam3dh outfrq3s_NEUwetdep 9s .........................................FAIL! rc= 7 at Tue Sep 13 19:25:45 MDT 2016
047 bl470 TBL.sh fsd1.9c5.4wtclbh outfrq3s_sd+sd_waccm5_geos5 9s ..............................FAIL! rc= 7 at Tue Sep 13 20:03:40 MDT 2016
050 bl482 TBL.sh f1.9c5wscmam3dh outfrq3s_2005+waccm_sc_1955-2005_cam5 9s .....................FAIL! rc= 7 at Tue Sep 13 20:38:48 MDT 2016
063 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ........................................FAIL! rc= 7 at Tue Sep 13 21:03:18 MDT 2016
068 bl740 TBL.sh h30c5h outfrq3s_lb0 9s .......................................................FAIL! rc= 7 at Tue Sep 13 21:07:52 MDT 2016
yellowstone/intel/aux_cam:
FAIL SMS_Ln9.f09_f09.F2000_DONOTUSE.yellowstone_intel.cam-outfrq9s COMPARE_baseline
hobart/nag:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s ...........................................................FAIL! rc= 7 at Tue Sep 13 15:27:37 MD
T 2016
010 bl112 TBL.sh e8c5dm ghgrmp_e8 9s ..........................................................FAIL! rc= 7 at Tue Sep 13 15:31:46 MD
T 2016
014 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Tue Sep 13 15:35:19 MD
T 2016
021 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Tue Sep 13 15:47:52 MD
T 2016
034 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .........................................FAIL! rc= 7 at Tue Sep 13 15:59:51 MD
T 2016
040 bl324 TBL.sh f4c4rrtmgdm outfrq3s 9s ......................................................FAIL! rc= 7 at Tue Sep 13 16:08:47 MD
T 2016
043 bl325 TBL.sh f10c5t5mdmsc outfrq3s_subcol 9s ..............................................FAIL! rc= 7 at Tue Sep 13 16:13:04 MD
T 2016
049 bl338 TBL.sh f10c5dm rad_diag_mam 9s ......................................................FAIL! rc= 7 at Tue Sep 13 16:16:40 MD
T 2016
052 bl354 TBL.sh f1.9c5.4dm outfrq3s_npryz 9s .................................................FAIL! rc= 7 at Tue Sep 13 16:29:12 MD
T 2016
055 bl356 TBL.sh fsd1.9c5clbdm off1.9x2.5 9s ..................................................FAIL! rc= 7 at Tue Sep 13 16:51:48 MD
T 2016
058 bl368 TBL.sh f1.9c5mg2dm outfrq3s_macmic2_classnuc 9s .....................................FAIL! rc= 7 at Tue Sep 13 17:03:56 MD
T 2016
061 bl369 TBL.sh f1.9c5clbdm outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Tue Sep 13 17:19:11 MD
T 2016
064 bl734 TBL.sh h16c5naqdm outfrq3s_bam+aquaplanet_cam5 9s ...................................FAIL! rc= 7 at Tue Sep 13 17:26:15 MDT 2016
067 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Tue Sep 13 17:31:31 MDT 2016
073 bl471 TBL.sh f10c5wmam3dm outfrq3s_modalstrat 9s ..........................................FAIL! rc= 7 at Tue Sep 13 17:46:16 MDT 2016
076 bl805 TBL.sh f4c5carmtt1dm outfrq3s_carma 9s ..............................................FAIL! rc= 7 at Tue Sep 13 18:01:57 MDT 2016
079 bl011 TBL.sh f4c5portdm outfrq24h_port 2d .................................................FAIL! rc= 7 at Tue Sep 13 18:07:20 MDT 2016
hobart/pgi or jaguar/pgi:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s ...........................................................FAIL! rc= 7 at Tue Sep 13 15:31:34 MDT 2016
010 bl112 TBL.sh e8c5dm ghgrmp_e8 9s ..........................................................FAIL! rc= 7 at Tue Sep 13 15:39:34 MDT 2016
012 bl113 TBL.sh e8c5paqdm aqpgro+aquaplanet_cam5 3s ..........................................FAIL! rc= 7 at Tue Sep 13 15:45:51 MDT 2016
016 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Tue Sep 13 15:48:35 MDT 2016
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Tue Sep 13 16:01:56 MDT 2016
038 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .........................................FAIL! rc= 7 at Tue Sep 13 16:35:51 MDT 2016
041 bl317 TBL.sh f10c5cdm outfrq3s 9s .........................................................FAIL! rc= 7 at Tue Sep 13 16:44:56 MDT 2016
045 bl319 TBL.sh f10c5nonedm outfrq3s_bam 9s ..................................................FAIL! rc= 7 at Tue Sep 13 16:51:58 MDT 2016
048 bl320 TBL.sh f10c5nonedm rad_diag 9s ......................................................FAIL! rc= 7 at Tue Sep 13 16:53:39 MDT 2016
051 bl321 TBL.sh f10c5cdm atrain 9s ...........................................................FAIL! rc= 7 at Tue Sep 13 16:56:58 MDT 2016
054 bl334 TBL.sh f10c5udm outfrq3s 9s .........................................................FAIL! rc= 7 at Tue Sep 13 17:05:15 MDT 2016
057 bl354 TBL.sh f1.9c5.4dm outfrq3s_npryz 9s .................................................FAIL! rc= 7 at Tue Sep 13 17:25:06 MDT 2016
060 bl420 TBL.sh f4c5mam7dm outfrq3s 9s .......................................................FAIL! rc= 7 at Tue Sep 13 17:36:01 MDT 2016
063 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Tue Sep 13 17:46:07 MDT 2016
066 bl734 TBL.sh h16c5naqdm outfrq3s_bam+aquaplanet_cam5 9s ...................................FAIL! rc= 7 at Tue Sep 13 17:58:34 MDT 2016
068 bl011 TBL.sh f4c5portdm outfrq24h_port 2d .................................................FAIL! rc= 7 at Tue Sep 13 18:05:53 MDT 2016
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: all which use rrtmg radiation
- what platforms/compilers: all
- nature of change (roundoff; larger than roundoff but same climate; new
climate): larger than roundoff, but not climate changing
URL for AMWG diagnostics output:
http://webext.cgd.ucar.edu/FCLIMO/f.e15.F2000.f09_f09.pd.100_rrtmg_cam5_4_79/atm/f.e15.F2000.f09_f09.pd.100_rrtmg_cam5_4_79-f.e15.F2000.f09_f09.pd.100_cam5_4_79/
===============================================================
===============================================================
Tag name: cam5_4_82
Originator(s): cacraig, aliceb, mvertens, fvitt, mmills
Date: September 13, 2016
One-line Summary: Update to cime5.0.11 and other misc changes
Purpose of changes:
- Update to cime5.0.11 to eliminate problem with --generate flag
on create_test
- Received from aliceb, bug fix for time_period_freq
- Received from mvertens, bug fix for config_compsets
- Changed CAM54 to CAM60 in WACCM FW5 compset (per fvitt and mmills)
Bugs fixed (include bugzilla ID):
- bug fix for time_period_freq in cam_history.F90
- bug fix in config_compsets when specifying grid=gland
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M .
M SVN_EXTERNAL_DIRECTORIES
- update to cime5.0.11 as cime5.0.8 had answer changes when using
the --generate flag in create_test
M components/cam/cime_config/config_component.xml
- removed CAM54
M components/cam/cime_config/config_compsets.xml
- changed the CAM54 reference in FW5 to CAM60
- grid=gland is now specified properly
M components/cam/src/control/cam_history.F90
- when calculating time_period_freq, convert nhtfrq to always be in seconds
(when it is negative it is the number of hours)
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB
yellowstone/intel/aux_cam: (compared with cam5_4_80 as cam5_4_81 had differences due to a bug in cime)
FAIL SMS_Ln9.f09_f09.F2000_DONOTUSE.yellowstone_intel.cam-outfrq9s COMPARE_baseline
- As discovered in cam5_4_81, the two files which had differences were the clm2.h0 and cpl.hi files.
In both cases, the only two fields which had differences (and they were roundoff level)
were VOLR and VOLRMCH (and the coupler versions of these fields). Bill Sacks said that
these two fields can exhibit roundoff level differences and CLM has learned to ignore them.
hobart/nag: all BFB
hobart/pgi or jaguar/pgi: all BFB
CAM tag used for the baseline comparison tests if different than previous
tag: cam5_4_80 - cam5_4_81 was BFB with cam5_4_80 but had answer differences in the aux_cam test due to
a bug in cime. For convenience, used cam5_4_80 for all the baseline comparisons
===============================================================
===============================================================
Tag name: cam5_4_81
Originator(s): mvertens, santos, jedwards, cacraig
Date: August 19, 2016
One-line Summary: Update to CESM2 beta01 externals (cime5)
Purpose of changes:
- Updates necessary to work with the new version of CIME.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M .
M SVN_EXTERNAL_DIRECTORIES
- Update the externals to match CESM beta01
M components/cam/bld/configure
- The CIME/CESM scripts now look for a file called "CESM_cppdefs"
instead of "CCSM_cppdefs", so configure has to output to that file.
M components/cam/cime_config/buildlib
M components/cam/cime_config/buildnml
- These CIME scripts are translated from Perl to Python.
- Queries to XML files now use CIME Python modules rather than xmlquery,
which substantially improves buildnml timing in CIME.
M components/cam/cime_config/config_component.xml
M components/cam/cime_config/config_compsets.xml
- Compatibility fixes for new CIME scripts.
- Compset long names always use two version numbers for CAM (i.e. CAM40,
CAM50, and CAM55).
- WACCM+CLUBB compsets are now specified by using CAM55%WC*, so the
%WCCB modifier is removed.
M components/cam/test/system/TSM_ccsm.sh
- The CIMEROOT environment variable must be specified when running the
CIME run script.
M components/cam/test/system/test_driver.sh
- added CAM_TAG which is used in the aux_cam create_testcase to generate
the automatic baseline files for the CESM test(s)
- The Python module must be loaded on yellowstone so that Python 2.7 is
available for the CIME scripts.
M components/cam/cime_config/testdefs/testmods_dirs/cam/chemproc/shell_commands
M components/cam/cime_config/testdefs/testmods_dirs/cam/cosp/shell_commands
M components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq1d/shell_commands
M components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq3d/shell_commands
M components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq9s/shell_commands
M components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq9s_sat_hist/shell_commands
M components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq9s_wse/shell_commands
- cleaned up shell commands
M components/cam/cime_config/config_pes.xml
- change WACCM pe layouts to be for any version, not just cam4 or cam5
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB
yellowstone/intel/aux_cam:
FAIL SMS_Ln9.f09_f09.F2000_DONOTUSE.yellowstone_intel.cam-outfrq9s COMPARE_baseline
- Upon further investigation, the two files which had differences were the clm2.h0 and cpl.hi files.
In both cases, the only two fields which had differences (and they were roundoff level)
were VOLR and VOLRMCH (and the coupler versions of these fields). Bill Sacks said that
these two fields can exhibit roundoff level differences and CLM has learned to ignore them.
hobart/nag: all BFB
hobart/pgi or jaguar/pgi: all BFB
===============================================================
===============================================================
Tag name: cam5_4_80
Originator(s): goldy
Date: 2016-08-17
One-line Summary: Create separate package for simple physics options
Purpose of changes: Create a new physics package for adiabatic and
Held-Suarez simulations in order to enable incorporation of other
idealized or simplified physics parameterizations.
Bugs fixed (include bugzilla ID): NA
Describe any changes made to build system: NA
Describe any changes made to the namelist: NA
List any changes to the defaults for the boundary datasets:
- New default datasets for Eulerian Held-Suarez simulations
HS1994.128x256.L30_c062216.nc
HS1994.128x256.L60_c061516.nc
HS1994.64x128.L30_c061616.nc
Describe any substantial timing or memory changes: NA
Code reviewed by: Eaton, Craig, Vitt
List all subroutines eliminated:
- phys_run1_adiabatic_or_ideal -- replaced with simple/physpkg
- tphysidl -- replace with proper Held-Suarez parameterization
List all subroutines added and what they do:
A + components/cam/bld/namelist_files/use_cases/held_suarez_1994.xml
- Use cases for Held-Suarez runs for Simpler Models project
A + components/cam/src/physics/simple
- New directory to hold various simplified physics options
A + components/cam/src/physics/simple/physpkg.F90
- Separate physpkg interfaces for simplified physics options
A components/cam/src/physics/simple/held_suarez.F90
- Portable layer Held-Suarez tendency calculation
A components/cam/src/physics/simple/held_suarez_cam.F90
- CAM interface layer for Held-Suarez parameterization
M components/cam/src/control/cam_control_mod.F90
- cam_ctrl_set_physics_type: Set physics type after namelist read
List all existing files that have been modified, and describe the changes:
M components/cam/bld/config_files/definition.xml
- Changed name of ideal physics package to held_suarez
M components/cam/bld/configure
- Changed name of ideal physics package to held_suarez. Added simple_phys
variable to help keep track of new simple physics options.
- Removed trailing spaces.
M components/cam/bld/namelist_files/namelist_definition.xml
- Fix some history documentation (the max number of history files is larger)
M components/cam/cime_config/config_component.xml
- Change IDEAL to HS94 for the -phys held_suarez configuration option
M components/cam/cime_config/config_compsets.xml
- Change the FIDEAL compset to FHS94
M components/cam/cime_config/config_pes.xml
- Took out redundant EOS entry at request of jedwards.
M components/cam/doc/ChangeLog
- Had Watson fill out cam5_4_80 entry
M components/cam/src/control/cam_comp.F90
- cam_init no longer takes adiabatic or ideal_phys inputs
M components/cam/src/control/cam_control_mod.F90
- cam_ctrl_init no longer takes adiabatic or ideal_phys inputs
- new subroutine, cam_ctrl_set_physics_type (see above)
M components/cam/src/control/cam_history.F90
- Remove SPMD and use real MPI calls
M components/cam/src/control/camsrfexch.F90
- Make wet pbuf fields optional
M components/cam/src/cpl/atm_comp_mct.F90
- No longer look for adiabatic or ideal_phys in infodata
M components/cam/src/dynamics/eul/tfilt_massfix.F90
- Skip energy fixer for Held-Suarez simulations
M components/cam/src/physics/cam/cam_diagnostics.F90
- Separated code into dry and moist sections. Moist sections are only
called for moist physics runs
M components/cam/src/physics/cam/microp_driver.F90
- Include name of microp_scheme if unrecognized
M components/cam/src/physics/cam/phys_control.F90
- Call cam_ctrl_set_physics_type to set physics type after namelist read
- Replace MPI shorthand with real MPI
M components/cam/src/physics/cam/physics_buffer.F90.in
- Improved endrun message for out of range error in pbuf_get_field
M components/cam/src/physics/cam/physpkg.F90
- Removed simple physics options (adiabatic and Held-Suarez)
- Replace MPI shorthand with real MPI
M components/cam/test/system/config_files/e48idh
M components/cam/test/system/config_files/e8idm
M components/cam/test/system/config_files/f10idm
M components/cam/test/system/config_files/f4idm
M components/cam/test/system/config_files/s32idh
M components/cam/test/system/config_files/s48idh
M components/cam/test/system/config_files/s64idh
- Replaced '-phys ideal' with '-phys held_suarez'
M components/cam/test/system/nl_files/idphys
- Added a second history file with U,V,T,TTEND_TOT,DTCORE
M components/cam/test/system/test_driver.sh
- Add an option (-e) to email test summary to USER (or at least try)
- Create test summary and add to cam_test_summaries (and maybe also email)
- Added code to run CAM AUX tests but need CIME 5 to get working.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All Pass
yellowstone/intel/aux_cam: All Pass
hobart/nag: Timestep relaxation filter removed from Held-Suarez parameterization
047 bl335 TBL.sh f10idm idphys 9s .............................................................FAIL! rc= 7 at Wed Aug 17 17:21:46 MDT 2016
hobart/pgi: Timestep relaxation filter removed from Held-Suarez parameterization
019 bl115 TBL.sh e8idm idphys 9s ..............................................................FAIL! rc= 7 at Wed Aug 17 16:55:29 MDT 2016
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate): Held-Suarez tests are larget than roundoff due to removal of
model timestep-based relaxation (parameterization matches original
1994 paper). Also, QRS diagnostic is purely the temperature
tendency from the heating rate (previous version also included
dycore energy fix).
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers: Long runs were conducted per 1994 paper. Evaluation
is on Simpler Models page (https://www2.cesm.ucar.edu/models/simpler-models/dynamical-core-test/adiabatic).
MSS location of control simulations used to validate new climate: NA
URL for AMWG diagnostics output used to validate new climate: NA
===============================================================
===============================================================
Tag name: cam5_4_79
Originator(s): cacraig, hannay
Date: Aug 10, 2016
One-line Summary: Add F2000_DONOTUSE compset
Purpose of changes:
- In preparation for the release, the compset F2000_DONOTUSE has been created.
This compset will be modified as directed by Cecile to mimic her setups.
NOTE--THIS COMPSET IS NOT TO BE USED BY OTHER DEVELOPERS AT THIS TIME
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
A components/cam/bld/namelist_files/use_cases/2000_cam6.xml
- Preliminary use_case for F2000 CAM6 setup
** THIS USE_CASE MAY BE MODIFIED UP UNTIL THE CESM2 RELEASE
List all existing files that have been modified, and describe the changes:
M .
M SVN_EXTERNAL_DIRECTORIES
- added cism, ww3, and mosart libraries, so that F2000 runs may
be made with CAM checkout
M components/cam/cime_config/config_component.xml
- add use_case for current 2000_cam55
M components/cam/cime_config/config_compsets.xml
- added F2000_DONOTUSE compset information
M components/cam/cime_config/testdefs/testlist_cam.xml
- add yellowstone, intel F2000_DONOTUSE regression test
M components/cam/doc/ChangeLog_template
- added new yellowstone aux_cam test reporting
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
yellowstone/intel/aux_cam: new test - no baseline to compare against
To run the new aux_cam tests (inside cime/scripts):
./create_test -xml_mach yellowstone -xml_compiler intel -xml_category aux_cam -testid cam5_4_79 -generate cam5_4_79
The summary can be seen in the file TestStatus
-- ran the F2000_DONOTUSE test by hand and compared F2000_DONOTUSE
results with cam5_4_70
and it was BFB
hobart/nag:
hobart/pgi or jaguar/pgi:
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_4_78
Originator(s): goldy
Date: 2016-08-01
One-line Summary: New Hobart environment + small fixes
Purpose of changes:
- Modify test_driver for new Hobart environment (NAG 6.1)
- Small fix to makefile
- Adjust SPCAM regression test
Bugs fixed (include bugzilla ID): NA
Describe any changes made to build system: NAG 6.1 plus update mismatch flag
Describe any changes made to the namelist: NA
List any changes to the defaults for the boundary datasets: NA
Describe any substantial timing or memory changes: NA
Code reviewed by: goldy, cacraig
List all subroutines eliminated: NA
List all subroutines added and what they do: NA
List all existing files that have been modified, and describe the changes:
M components/cam/bld/Makefile.in
- Added mpi_alltoallw to the NAG mismatch flags (-wmismatch)
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
- Corrected the default value of eul_hdif_coef for T85 and T42 cases
M components/cam/doc/ChangeLog
- Dusted off the old template filling auto-bot
M components/cam/src/dynamics/fv/pfixer.F90
- Corrected incorrect usage of mpi_allgatherv (overlapping send and dest)
M components/cam/test/system/config_files/f10c5spcamsm
- Modified spcam_nx to be 32
M components/cam/test/system/test_driver.sh
- Updated NAG compiler to 6.1
- Do not separately load an MPI library (only compiler module load).
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS
hobart/nag: Roundoff-level changes due to compiler update
005 bl111 TBL.sh e8c5t5mdm ttrac 9s ...........................................................FAIL! rc= 7 at Mon Aug 1 13:12:57 MDT 2016
010 bl112 TBL.sh e8c5dm ghgrmp_e8 9s ..........................................................FAIL! rc= 7 at Mon Aug 1 13:18:10 MDT 2016
014 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Mon Aug 1 13:22:43 MDT 2016
017 bl221 TBL.sh f10c5spcamsm outfrq3s 9s .....................................................FAIL! rc= 7 at Mon Aug 1 13:30:09 MDT 2016
021 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Mon Aug 1 13:36:31 MDT 2016
034 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .........................................FAIL! rc= 7 at Mon Aug 1 13:50:01 MDT 2016
040 bl324 TBL.sh f4c4rrtmgdm outfrq3s 9s ......................................................FAIL! rc= 7 at Mon Aug 1 13:59:39 MDT 2016
043 bl325 TBL.sh f10c5t5mdmsc outfrq3s_subcol 9s ..............................................FAIL! rc= 7 at Mon Aug 1 14:04:11 MDT 2016
049 bl338 TBL.sh f10c5dm rad_diag_mam 9s ......................................................FAIL! rc= 7 at Mon Aug 1 14:08:44 MDT 2016
052 bl354 TBL.sh f1.9c5.4dm outfrq3s_npryz 9s .................................................FAIL! rc= 7 at Mon Aug 1 14:21:07 MDT 2016
055 bl356 TBL.sh fsd1.9c5clbdm off1.9x2.5 9s ..................................................FAIL! rc= 7 at Mon Aug 1 14:43:05 MDT 2016
058 bl368 TBL.sh f1.9c5mg2dm outfrq3s_macmic2_classnuc 9s .....................................FAIL! rc= 7 at Mon Aug 1 14:54:56 MDT 2016
061 bl369 TBL.sh f1.9c5clbdm outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Mon Aug 1 15:09:49 MDT 2016
064 bl734 TBL.sh h16c5naqdm outfrq3s_bam+aquaplanet_cam5 9s ...................................FAIL! rc= 7 at Mon Aug 1 15:16:31 MDT 2016
067 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Mon Aug 1 15:21:34 MDT 2016
070 bl426 TBL.sh f10c4wmxdm outfrq3s_ionos 9s .................................................FAIL! rc= 7 at Mon Aug 1 15:28:29 MDT 2016
073 bl471 TBL.sh f10c5wmam3dm outfrq3s_modalstrat 9s ..........................................FAIL! rc= 7 at Mon Aug 1 15:36:22 MDT 2016
076 bl805 TBL.sh f4c5carmtt1dm outfrq3s_carma 9s ..............................................FAIL! rc= 7 at Mon Aug 1 15:52:15 MDT 2016
079 bl011 TBL.sh f4c5portdm outfrq24h_port 2d .................................................FAIL! rc= 7 at Mon Aug 1 15:57:46 MDT 2016
hobart/pgi: ALL PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate): roundoff level diffs in some NAG baseline tests
If bitwise differences were observed, how did you show they were no worse
than roundoff?
- Hand examine cprnc output
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository): NA
MSS location of control simulations used to validate new climate: NA
URL for AMWG diagnostics output used to validate new climate: NA
===============================================================
===============================================================
Tag name: cam5_4_77
Originator(s): goldy
Date: 2016-07-09
One-line Summary: Fix default constant initialization bug
Purpose of changes: Fix default constant initialization bug in SE
Bugs fixed (include bugzilla ID): NA
Describe any changes made to build system: NA
Describe any changes made to the namelist: NA
List any changes to the defaults for the boundary datasets: NA
Describe any substantial timing or memory changes: NA
Code reviewed by: goldy
List all subroutines eliminated: NA
List all subroutines added and what they do: NA
List all existing files that have been modified, and describe the changes:
M components/cam/doc/ChangeLog
- Adding yet another wall of shame
M components/cam/src/dynamics/se/dyn_comp.F90
- Set redundant tracer points to zero before boundary exchange
A f10c5spcamsm
- Config file for testing SPCAM (should have been added to cam5_4_76)
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS
hobart/nag: All PASS
hobart/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag: NA
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate): NA
If bitwise differences were observed, how did you show they were no worse
than roundoff? NA
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository): NA
===============================================================
===============================================================
Tag name: cam5_4_76
Originator(s): cacraig, mark@atmos.colostate.edu, minghuai.wang@nju.edu.cn
Date: July 8, 2016
One-line Summary: super-parameterized CAM (SPCAM)
Purpose of changes:
- Super-parameterized CAM (SPCAM) implements a 2D cloud resolving model (the
System for Atmospheric Modeling SAM, version 6.8.2) in CAM. When it is turned on,
it replaces CAM's parameterization for moist convection and large-scale condensation
with this alternate model.
- At this time, only the sam1mom (One moment SAM microphysics) version of SPCAM is allowed.
The m2005 version (two moment microphysics) is awaiting scientific verification before will be enabled.
- The SPCAM package allows CLUBB to be used or not. It is important to note that there is
a SPCAM-specific version of CLUBB within the CRM package and it is not the same CLUBB being
used by CAM.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
- introduced two new options to -phys: sam1mom and m2005 (though m2005 is not currently enabled)
- New configure options:
spcam_clubb_sgs: turns on the SPCAM version of CLUBB
spcam_nx: SPCAM's x-grid. defaults to 4
spcam_ny: SPCAM's y-grid
spcam_dx: SPCAM's horizontal grid spacing
spcam_dt SPCAM's timestep
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton and CAM softwared engineering team
List all subroutines eliminated:
List all subroutines added and what they do:
A components/cam/src/physics/cam/spcam_drivers.F90
- driver interface routines between CAM and SPCAM
A components/cam/src/physics/spcam/crm_physics.F90
- wrapper routines around the CRM package
A components/cam/src/physics/spcam/crmclouds_camaerosols.F90
- subroutines to use cloud fields from the CRM model within CAM
A components/cam/src/physics/spcam/crmdims.F90
- module to setup crm dimensions
A components/cam/src/physics/spcam
A components/cam/src/physics/spcam/crm
A components/cam/src/physics/spcam/crm/ADV_MPDATA
A components/cam/src/physics/spcam/crm/ADV_MPDATA/crmx_advect_scalar.F90
A components/cam/src/physics/spcam/crm/ADV_MPDATA/crmx_advect_scalar2D.F90
A components/cam/src/physics/spcam/crm/ADV_MPDATA/crmx_advect_scalar3D.F90
A components/cam/src/physics/spcam/crm/ADV_MPDATA/crmx_advection.F90
A components/cam/src/physics/spcam/crm/CLUBB
A components/cam/src/physics/spcam/crm/CLUBB/crmx_Skw_module.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_T_in_K_module.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_advance_helper_module.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_advance_windm_edsclrm_module.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_advance_wp2_wp3_module.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_advance_xm_wpxp_module.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_advance_xp2_xpyp_module.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_anl_erf.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_array_index.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_calendar.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_clip_explicit.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_clip_semi_implicit.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_clubb_core.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_clubb_precision.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_constants_clubb.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_corr_matrix_module.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_csr_matrix_class_3array.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_diagnose_correlation_module.f90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_diffusion.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_endian.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_error_code.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_extrapolation.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_file_functions.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_fill_holes.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_gmres_cache.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_gmres_wrap.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_grid_class.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_hydrostatic_module.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_hyper_diffusion_4th_ord.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_input_names.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_input_reader.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_interpolation.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_lapack_wrap.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_matrix_operations.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_mean_adv.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_mixing_length.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_model_flags.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_mono_flux_limiter.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_mt95.f90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_numerical_check.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_output_grads.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_output_netcdf.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_parameter_indices.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_parameters_microphys.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_parameters_model.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_parameters_radiation.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_parameters_tunable.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_pdf_closure_module.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_pdf_parameter_module.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_pos_definite_module.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_saturation.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_sigma_sqd_w_module.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_sponge_layer_damping.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_stat_file_module.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_stats_LH_sfc.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_stats_LH_zt.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_stats_rad_zm.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_stats_rad_zt.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_stats_sfc.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_stats_subs.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_stats_type.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_stats_variables.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_stats_zm.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_stats_zt.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_surface_varnce_module.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_variables_diagnostic_module.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_variables_prognostic_module.F90
A components/cam/src/physics/spcam/crm/CLUBB/crmx_variables_radiation_module.F90
A components/cam/src/physics/spcam/crm/CLUBB/recl.inc
A components/cam/src/physics/spcam/crm/MICRO_M2005
A components/cam/src/physics/spcam/crm/MICRO_M2005/README.MICRO_M2005
A components/cam/src/physics/spcam/crm/MICRO_M2005/crmx_drop_activation.F90
A components/cam/src/physics/spcam/crm/MICRO_M2005/crmx_microphysics.F90
A components/cam/src/physics/spcam/crm/MICRO_M2005/crmx_module_mp_graupel.F90
A components/cam/src/physics/spcam/crm/MICRO_SAM1MOM
A components/cam/src/physics/spcam/crm/MICRO_SAM1MOM/crmx_cloud.F90
A components/cam/src/physics/spcam/crm/MICRO_SAM1MOM/crmx_micro_params.F90
A components/cam/src/physics/spcam/crm/MICRO_SAM1MOM/crmx_microphysics.F90
A components/cam/src/physics/spcam/crm/MICRO_SAM1MOM/crmx_precip_init.F90
A components/cam/src/physics/spcam/crm/MICRO_SAM1MOM/crmx_precip_proc.F90
A components/cam/src/physics/spcam/crm/MICRO_SAM1MOM/crmx_precip_proc_clubb.F90
A components/cam/src/physics/spcam/crm/Readme_codes_merging_sam6.8.2_sam6.10.4.txt
A components/cam/src/physics/spcam/crm/SGS_CLUBBkvhkvm
A components/cam/src/physics/spcam/crm/SGS_CLUBBkvhkvm/crmx_clubb_sgs.F90
A components/cam/src/physics/spcam/crm/SGS_CLUBBkvhkvm/crmx_clubb_silhs_vars.F90
A components/cam/src/physics/spcam/crm/SGS_CLUBBkvhkvm/crmx_clubbvars.F90
A components/cam/src/physics/spcam/crm/SGS_CLUBBkvhkvm/crmx_diffuse_mom.F90
A components/cam/src/physics/spcam/crm/SGS_CLUBBkvhkvm/crmx_diffuse_mom2D.F90
A components/cam/src/physics/spcam/crm/SGS_CLUBBkvhkvm/crmx_diffuse_mom2D_xy.F90
A components/cam/src/physics/spcam/crm/SGS_CLUBBkvhkvm/crmx_diffuse_mom2D_z.F90
A components/cam/src/physics/spcam/crm/SGS_CLUBBkvhkvm/crmx_diffuse_mom3D.F90
A components/cam/src/physics/spcam/crm/SGS_CLUBBkvhkvm/crmx_diffuse_mom3D_xy.F90
A components/cam/src/physics/spcam/crm/SGS_CLUBBkvhkvm/crmx_diffuse_mom3D_z.F90
A components/cam/src/physics/spcam/crm/SGS_CLUBBkvhkvm/crmx_diffuse_scalar.F90
A components/cam/src/physics/spcam/crm/SGS_CLUBBkvhkvm/crmx_diffuse_scalar2D.F90
A components/cam/src/physics/spcam/crm/SGS_CLUBBkvhkvm/crmx_diffuse_scalar2D_xy.F90
A components/cam/src/physics/spcam/crm/SGS_CLUBBkvhkvm/crmx_diffuse_scalar2D_z.F90
A components/cam/src/physics/spcam/crm/SGS_CLUBBkvhkvm/crmx_diffuse_scalar3D.F90
A components/cam/src/physics/spcam/crm/SGS_CLUBBkvhkvm/crmx_diffuse_scalar3D_xy.F90
A components/cam/src/physics/spcam/crm/SGS_CLUBBkvhkvm/crmx_diffuse_scalar3D_z.F90
A components/cam/src/physics/spcam/crm/SGS_CLUBBkvhkvm/crmx_diffuse_scalar_xy.F90
A components/cam/src/physics/spcam/crm/SGS_CLUBBkvhkvm/crmx_diffuse_scalar_z.F90
A components/cam/src/physics/spcam/crm/SGS_CLUBBkvhkvm/crmx_fluxes_scalar_z.F90
A components/cam/src/physics/spcam/crm/SGS_CLUBBkvhkvm/crmx_sgs.F90
A components/cam/src/physics/spcam/crm/SGS_CLUBBkvhkvm/crmx_shear_prod2D.F90
A components/cam/src/physics/spcam/crm/SGS_CLUBBkvhkvm/crmx_shear_prod3D.F90
A components/cam/src/physics/spcam/crm/SGS_CLUBBkvhkvm/crmx_stat_clubb.F90
A components/cam/src/physics/spcam/crm/SGS_CLUBBkvhkvm/crmx_tke_full.F90
A components/cam/src/physics/spcam/crm/SGS_TKE
A components/cam/src/physics/spcam/crm/SGS_TKE/crmx_diffuse_mom.F90
A components/cam/src/physics/spcam/crm/SGS_TKE/crmx_diffuse_mom2D.F90
A components/cam/src/physics/spcam/crm/SGS_TKE/crmx_diffuse_mom3D.F90
A components/cam/src/physics/spcam/crm/SGS_TKE/crmx_diffuse_scalar.F90
A components/cam/src/physics/spcam/crm/SGS_TKE/crmx_diffuse_scalar2D.F90
A components/cam/src/physics/spcam/crm/SGS_TKE/crmx_diffuse_scalar3D.F90
A components/cam/src/physics/spcam/crm/SGS_TKE/crmx_sgs.F90
A components/cam/src/physics/spcam/crm/SGS_TKE/crmx_shear_prod2D.F90
A components/cam/src/physics/spcam/crm/SGS_TKE/crmx_shear_prod3D.F90
A components/cam/src/physics/spcam/crm/SGS_TKE/crmx_tke_full.F90
A components/cam/src/physics/spcam/crm/crmx_abcoefs.F90
A components/cam/src/physics/spcam/crm/crmx_adams.F90
A components/cam/src/physics/spcam/crm/crmx_advect2_mom_xy.F90
A components/cam/src/physics/spcam/crm/crmx_advect2_mom_z.F90
A components/cam/src/physics/spcam/crm/crmx_advect_all_scalars.F90
A components/cam/src/physics/spcam/crm/crmx_advect_mom.F90
A components/cam/src/physics/spcam/crm/crmx_atmosphere.F90
A components/cam/src/physics/spcam/crm/crmx_bound_duvdt.F90
A components/cam/src/physics/spcam/crm/crmx_bound_exchange.F90
A components/cam/src/physics/spcam/crm/crmx_boundaries.F90
A components/cam/src/physics/spcam/crm/crmx_buoyancy.F90
A components/cam/src/physics/spcam/crm/crmx_compress3D.F90
A components/cam/src/physics/spcam/crm/crmx_coriolis.F90
A components/cam/src/physics/spcam/crm/crmx_crm_module.F90
A components/cam/src/physics/spcam/crm/crmx_crmsurface.F90
A components/cam/src/physics/spcam/crm/crmx_crmtracers.F90
A components/cam/src/physics/spcam/crm/crmx_damping.F90
A components/cam/src/physics/spcam/crm/crmx_diagnose.F90
A components/cam/src/physics/spcam/crm/crmx_domain.F90
A components/cam/src/physics/spcam/crm/crmx_ecppvars.F90
A components/cam/src/physics/spcam/crm/crmx_forcing.F90
A components/cam/src/physics/spcam/crm/crmx_grid.F90
A components/cam/src/physics/spcam/crm/crmx_ice_fall.F90
A components/cam/src/physics/spcam/crm/crmx_init.F90
A components/cam/src/physics/spcam/crm/crmx_kurant.F90
A components/cam/src/physics/spcam/crm/crmx_module_ecpp_crm_driver.F90
A components/cam/src/physics/spcam/crm/crmx_module_ecpp_stats.F90
A components/cam/src/physics/spcam/crm/crmx_params.F90
A components/cam/src/physics/spcam/crm/crmx_periodic.F90
A components/cam/src/physics/spcam/crm/crmx_precip_fall.F90
A components/cam/src/physics/spcam/crm/crmx_press_grad.F90
A components/cam/src/physics/spcam/crm/crmx_press_rhs.F90
A components/cam/src/physics/spcam/crm/crmx_pressure.F90
A components/cam/src/physics/spcam/crm/crmx_random.F90
A components/cam/src/physics/spcam/crm/crmx_sat.F90
A components/cam/src/physics/spcam/crm/crmx_setparm.F90
A components/cam/src/physics/spcam/crm/crmx_setperturb.F90
A components/cam/src/physics/spcam/crm/crmx_stepout.F90
A components/cam/src/physics/spcam/crm/crmx_task_init.F90
A components/cam/src/physics/spcam/crm/crmx_task_util_NOMPI.F90
A components/cam/src/physics/spcam/crm/crmx_utils.F90
A components/cam/src/physics/spcam/crm/crmx_uvw.F90
A components/cam/src/physics/spcam/crm/crmx_vars.F90
A components/cam/src/physics/spcam/crm/crmx_zero.F90
A components/cam/src/physics/spcam/crm/fft.F
A components/cam/src/physics/spcam/crm/gammafff.c
- Routines for the CRM model - includes a CRM-specific version of CLUBB
A components/cam/src/physics/spcam/ecpp
A components/cam/src/physics/spcam/ecpp/ecpp_modal_aero_activate.F90
A components/cam/src/physics/spcam/ecpp/ecpp_modal_cloudchem.F90
A components/cam/src/physics/spcam/ecpp/ecpp_modal_wetscav.F90
A components/cam/src/physics/spcam/ecpp/module_data_ecpp1.F90
A components/cam/src/physics/spcam/ecpp/module_data_mosaic_asect.F90
A components/cam/src/physics/spcam/ecpp/module_data_radm2.F90
A components/cam/src/physics/spcam/ecpp/module_ecpp_ppdriver2.F90
A components/cam/src/physics/spcam/ecpp/module_ecpp_td2clm.F90
A components/cam/src/physics/spcam/ecpp/module_ecpp_util.F90
- Explicit-Cloud-Parametrized-Pollutant (ECPP) approach which is
used to treat cloud processing of aerosols with statistics of
cloud properties resolved by the cloud resolving model.
List all existing files that have been modified, and describe the changes:
M components/cam/bld/build-namelist
- sets up a few namelist settings based on spcam configure settings
M components/cam/bld/config_files/definition.xml
M components/cam/bld/configure
- introduced two new options to -phys: sam1mom and m2005 (though m2005 is not currently enabled)
- New configure options:
spcam_clubb_sgs: turns on the SPCAM version of CLUBB
spcam_nx: SPCAM's x-grid. defaults to 4
spcam_ny: SPCAM's y-grid
spcam_dx: SPCAM's horizontal grid spacing
spcam_dt SPCAM's timestep
- spcam specific settings are added internally to phys_pkg, microphys_pkg,
macrophys_pkg and pbl_pkg
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
M components/cam/bld/namelist_files/namelist_definition.xml
- specify namelist settings for SPCAM
M components/cam/cime_config/config_component.xml
M components/cam/cime_config/config_compsets.xml
M components/cam/cime_config/testdefs/testlist_cam.xml
- added FSPCAMS compset to use SPCAM sam1mom
M components/cam/src/chemistry/modal_aero/aero_model.F90
- introduced accessor functions to retrieve aerosol data
- aqueous chemistry is done in ECPP and is not done in this module
M components/cam/src/chemistry/modal_aero/modal_aero_convproc.F90
M components/cam/src/chemistry/modal_aero/modal_aero_data.F90
M components/cam/src/physics/cam/constituents.F90
- constituents species class is now handled like convtran2
M components/cam/src/chemistry/mozart/gas_wetdep_opts.F90
- introduce an 'OFF' option for gas_wetdep_method
M components/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
- comment cleanup
M components/cam/src/chemistry/mozart/mo_neu_wetdep.F90
- do_neu_wetdep is no longer public
M components/cam/src/chemistry/utils/modal_aero_calcsize.F90
M components/cam/src/chemistry/utils/modal_aero_wateruptake.F90
M components/cam/src/physics/cam/modal_aer_opt.F90
- introduce a new routine modal_aero_calcdry and remove
this logic from modal_aero_wateruptake_dr
M components/cam/src/physics/cam/cloud_diagnostics.F90
- introduce logic for one and two moment clouds and
set them according to the type of microphysics package
being used
M components/cam/src/physics/cam/cloud_fraction.F90
M components/cam/src/physics/cam/conv_water.F90
M components/cam/src/physics/cam/convect_deep.F90
M components/cam/src/physics/cam/convect_shallow.F90
M components/cam/src/physics/cam/diffusion_solver.F90
M components/cam/src/physics/cam/microp_driver.F90
M components/cam/src/physics/cam/phys_control.F90
M components/cam/src/physics/cam/physpkg.F90
M components/cam/src/physics/cam/pkg_cldoptics.F90
M components/cam/src/physics/cam/vertical_diffusion.F90
- add checks for the spcam schemes
M components/cam/src/physics/cam/ndrop.F90
- add additional logic if called from spcam
M components/cam/src/physics/cam/radiation_data.F90
- add logic to disable writing of radiation data
within the radiation routine.
M components/cam/src/physics/cam/unicon_cam.F90
M components/cam/src/physics/cam/zm_conv_intr.F90
- additional pbuf fields used by other methods
needed to be added here as well.
M components/cam/src/physics/rrtmg/radiation.F90
- add a few extra variables to the radiation output type
M components/cam/src/unit_drivers/offline_driver.F90
- introduce a new offline driver register method
M components/cam/src/unit_drivers/rad/unit_driver.F90
- introduce a new unit driver register method
M components/cam/src/unit_drivers/stub/unit_driver.F90
- stub out unit_driver_reg
M components/cam/test/system/input_tests_master
M components/cam/test/system/tests_pretag_hobart_nag
- add regression test for sam1mom
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB
hobart/nag: all BFB except for the new added test which has no baseline
017 bl221 TBL.sh f10c5spcamsm outfrq3s 9s .....................................................FAIL! rc= 7 at Thu Jul 7 17:55:11 MDT 2016
hobart/pgi or jaguar/pgi: all BFB
Also ran the new aux_spcam compsets on hobart (nag and pgi) and yellowstone (intel) and they ran to completion
===============================================================
===============================================================
Tag name: cam5_4_75
Originator(s): goldy
Date: 2016-06-18
One-line Summary: Consolidate default initialization of constituents into
one routine
Purpose of changes: Currently, each dycore replicates all the logic
necessary to initialize a constituent to a default value. In
addition to being a maintenance issue, this functionality
naturally belongs in the physics code. The new function,
cnst_init_default_col, also has a clearer interface. Latitude and
longitue are passed explicitly along with an optional mask which
defines valid indices in the constituent array.
Bugs fixed (include bugzilla ID): NA
Describe any changes made to build system: NA
Describe any changes made to the namelist: NA
List any changes to the defaults for the boundary datasets: NA
Describe any substantial timing or memory changes: NA
Code reviewed by: goldy, eaton (design review)
List all subroutines eliminated:
- convect_shallow_init_cnst -- unused code
- convect_shallow_implements_cnst -- unused code
List all subroutines added and what they do:
- cnst_init_default -- Allows a dycore to initialize a constituent to its
default value
List all existing files that have been modified, and describe the changes:
M components/cam/doc/ChangeLog
- Brilliant prose added in changelogic pentameter
M components/cam/src/chemistry/mozart/chemistry.F90
M components/cam/src/chemistry/pp_none/chemistry.F90
- Updated chem_init_cnst to use the new interface
M components/cam/src/dynamics/eul/dyn_comp.F90
M components/cam/src/dynamics/fv/dyn_comp.F90
M components/cam/src/dynamics/se/dyn_comp.F90
M components/cam/src/dynamics/sld/dyn_comp.F90
- Replaced the old default initialization logic with a call to cnst_init_default_col
M components/cam/src/dynamics/se/dyn_grid.F90
- Made pelat_deg and pelon_deg public to allow access to cnst_init_default_col
M components/cam/src/physics/cam/aoa_tracers.F90
- Updated aoa_tracers_init_cnst to use the new interface
M components/cam/src/physics/cam/carma_intr.F90
- Updated carma_init_cnst to use the new interface
M components/cam/src/physics/cam/clubb_intr.F90
- Updated clubb_init_cnst to use the new interface
M components/cam/src/physics/cam/co2_cycle.F90
- Updated co2_init_cnst to use the new interface
A + components/cam/src/physics/cam/const_init.F90
- New module to hold the cnst_init_default_col subroutine
M components/cam/src/physics/cam/convect_shallow.F90
- Removed unused routines, convect_shallow_init_cnst and convect_shallow_implements_cnst
M components/cam/src/physics/cam/micro_mg_cam.F90
- Updated micro_mg_cam_init_cnst to use the new interface
M components/cam/src/physics/cam/microp_driver.F90
- Updated microp_driver_init_cnst to use the new interface
M components/cam/src/physics/cam/rk_stratiform.F90
- Updated rk_stratiform_init_cnst to use the new interface
M components/cam/src/physics/cam/tracers.F90
- Updated tracers_init_cnst to use the new interface
M components/cam/src/physics/cam/tracers_suite.F90
- Updated init_cnst_tr, init_cnst_lw, init_cnst_md, init_cnst_hi, and, init_cnst_un to use the new interface
M components/cam/src/physics/cam/unicon_cam.F90
- Updated unicon_init_cnst to use the new interface
M components/cam/src/physics/carma/cam/carma_intr.F90
- Updated carma_init_cnst to use the new interface
M components/cam/src/physics/carma/models/bc_strat/carma_model_mod.F90
M components/cam/src/physics/carma/models/cirrus/carma_model_mod.F90
M components/cam/src/physics/carma/models/cirrus_dust/carma_model_mod.F90
M components/cam/src/physics/carma/models/dust/carma_model_mod.F90
M components/cam/src/physics/carma/models/meteor_impact/carma_model_mod.F90
M components/cam/src/physics/carma/models/meteor_smoke/carma_model_mod.F90
M components/cam/src/physics/carma/models/mixed_sulfate/carma_model_mod.F90
M components/cam/src/physics/carma/models/pmc/carma_model_mod.F90
M components/cam/src/physics/carma/models/pmc_sulfate/carma_model_mod.F90
M components/cam/src/physics/carma/models/sea_salt/carma_model_mod.F90
M components/cam/src/physics/carma/models/sulfate/carma_model_mod.F90
M components/cam/src/physics/carma/models/test_detrain/carma_model_mod.F90
M components/cam/src/physics/carma/models/test_growth/carma_model_mod.F90
M components/cam/src/physics/carma/models/test_passive/carma_model_mod.F90
M components/cam/src/physics/carma/models/test_radiative/carma_model_mod.F90
M components/cam/src/physics/carma/models/test_swelling/carma_model_mod.F90
M components/cam/src/physics/carma/models/test_tracers/carma_model_mod.F90
M components/cam/src/physics/carma/models/test_tracers2/carma_model_mod.F90
M components/cam/src/physics/carma/models/tholin/carma_model_mod.F90
- Updated CARMA_InitializeParticle to use the new interface
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All Pass
hobart/nag: All Pass
hobart/pgi: All Pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: None
If bitwise differences were observed, how did you show they were no worse
than roundoff? NA
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced: NA
MSS location of control simulations used to validate new climate: NA
URL for AMWG diagnostics output used to validate new climate: NA
===============================================================
===============================================================
Tag name: cam5_4_74
Originator(s): cacraig
Date: June 17, 2016
One-line Summary: Update svn externals to match cesm1_5_alpha07c+
Purpose of changes:
- Update svn externals to cesm1_5_alpha07c
- Update cice external to cice5_2160614
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M .
M SVN_EXTERNAL_DIRECTORIES
- update svn externals
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
004 bl132 TBL.sh e48c4dh ghgrmp_e8+1850_cam4 9s ...............................................FAIL! rc= 7 at Thu Jun 16 21:53:22 MDT 2016
008 bl331 TBL.sh f4c4dh co2rmp+1850_cam4 9s ...................................................FAIL! rc= 7 at Thu Jun 16 21:56:29 MDT 2016
011 bl353 TBL.sh f1.9c5.4dh outfrq3s 9s .......................................................FAIL! rc= 7 at Thu Jun 16 22:08:17 MDT 2016
015 bl357 TBL.sh f1.9c5.5h outfrq3s+1850_cam55 9s .............................................FAIL! rc= 7 at Thu Jun 16 22:14:25 MDT 2016
018 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Thu Jun 16 22:26:48 MDT 2016
022 bl373 TBL.sh f1.9c4mozdh outfrq3s+2000_cam4_trop_chem 9s ..................................FAIL! rc= 7 at Thu Jun 16 22:42:56 MDT 2016
024 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...............................................FAIL! rc= 7 at Thu Jun 16 22:45:45 MDT 2016
027 bl485 TBL.sh f1.9c5.5wtmam4h outfrq3s+waccm_tsmlt_2000_cam55 9s ...........................FAIL! rc= 7 at Thu Jun 16 23:06:55 MDT 2016
030 bl390 TBL.sh f1.9c4wmxh outfrq3s+waccmx_2000_cam4 9s ......................................FAIL! rc= 7 at Thu Jun 16 23:16:19 MDT 2016
033 bl397 TBL.sh f1.9c4cm outfrq3s+1850-2005_cam4 9s ..........................................FAIL! rc= 7 at Thu Jun 16 23:19:40 MDT 2016
036 bl421 TBL.sh f1.9c5mam4dh outfrq3s 9s .....................................................FAIL! rc= 7 at Thu Jun 16 23:30:59 MDT 2016
039 bl429 TBL.sh f1.9c4wmxh outfrq3s+waccmxi_2000_cam4 9s .....................................FAIL! rc= 7 at Thu Jun 16 23:39:28 MDT 2016
041 bl430 TBL.sh f1.9c5mzmam3dh outfrq3s_NEUwetdep 9s .........................................FAIL! rc= 7 at Thu Jun 16 23:48:55 MDT 2016
044 bl440 TBL.sh f1.9c4soadh outfrq3s+soa_chem_megan_emis 9s ..................................FAIL! rc= 7 at Fri Jun 17 00:08:25 MDT 2016
047 bl470 TBL.sh fsd1.9c5.4wtclbh outfrq3s_sd+sd_waccm5_geos5 9s ..............................FAIL! rc= 7 at Fri Jun 17 00:30:35 MDT 2016
050 bl482 TBL.sh f1.9c5wscmam3dh outfrq3s_2005+waccm_sc_1955-2005_cam5 9s .....................FAIL! rc= 7 at Fri Jun 17 01:06:56 MDT 2016
054 bl387 TBL.sh f1.9c4dm outfrq3s+1850-2005_cam4 9s ..........................................FAIL! rc= 7 at Fri Jun 17 01:14:41 MDT 2016
057 bl388 TBL.sh f1.9c4m outfrq24h+1850_cam4 2d ...............................................FAIL! rc= 7 at Fri Jun 17 01:16:15 MDT 2016
068 bl740 TBL.sh h30c5h outfrq3s_lb0 9s .......................................................FAIL! rc= 7 at Fri Jun 17 01:36:09 MDT 2016
hobart/nag:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s ...........................................................FAIL! rc= 7 at Thu Jun 16 13:32:41 MDT 2016
010 bl112 TBL.sh e8c5dm ghgrmp_e8 9s ..........................................................FAIL! rc= 7 at Thu Jun 16 13:36:06 MDT 2016
014 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Thu Jun 16 13:38:44 MDT 2016
018 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Thu Jun 16 13:42:59 MDT 2016
028 bl314 TBL.sh f10c4wscdm outfrq3s 9s .......................................................FAIL! rc= 7 at Thu Jun 16 13:51:09 MDT 2016
034 bl318 TBL.sh f10c4cdm sat_hist 9s .........................................................FAIL! rc= 7 at Thu Jun 16 13:58:42 MDT 2016
037 bl324 TBL.sh f4c4rrtmgdm outfrq3s 9s ......................................................FAIL! rc= 7 at Thu Jun 16 14:01:06 MDT 2016
040 bl325 TBL.sh f10c5t5mdmsc outfrq3s_subcol 9s ..............................................FAIL! rc= 7 at Thu Jun 16 14:04:43 MDT 2016
046 bl338 TBL.sh f10c5dm rad_diag_mam 9s ......................................................FAIL! rc= 7 at Thu Jun 16 14:07:30 MDT 2016
049 bl354 TBL.sh f1.9c5.4dm outfrq3s_npryz 9s .................................................FAIL! rc= 7 at Thu Jun 16 14:19:02 MDT 2016
052 bl356 TBL.sh fsd1.9c5clbdm off1.9x2.5 9s ..................................................FAIL! rc= 7 at Thu Jun 16 14:37:03 MDT 2016
055 bl368 TBL.sh f1.9c5mg2dm outfrq3s_macmic2_classnuc 9s .....................................FAIL! rc= 7 at Thu Jun 16 14:46:48 MDT 2016
058 bl369 TBL.sh f1.9c5clbdm outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Thu Jun 16 14:59:16 MDT 2016
064 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Thu Jun 16 15:09:23 MDT 2016
067 bl426 TBL.sh f10c4wmxdm outfrq3s_ionos 9s .................................................FAIL! rc= 7 at Thu Jun 16 15:15:08 MDT 2016
070 bl471 TBL.sh f10c5wmam3dm outfrq3s_modalstrat 9s ..........................................FAIL! rc= 7 at Thu Jun 16 15:21:47 MDT 2016
073 bl805 TBL.sh f4c5carmtt1dm outfrq3s_carma 9s ..............................................FAIL! rc= 7 at Thu Jun 16 15:35:23 MDT 2016
hobart/pgi or jaguar/pgi:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s ...........................................................FAIL! rc= 7 at Thu Jun 16 13:35:32 MDT 2016
010 bl112 TBL.sh e8c5dm ghgrmp_e8 9s ..........................................................FAIL! rc= 7 at Thu Jun 16 13:41:54 MDT 2016
016 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Thu Jun 16 13:49:14 MDT 2016
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Thu Jun 16 14:00:07 MDT 2016
033 bl314 TBL.sh f10c4wscdm outfrq3s 9s .......................................................FAIL! rc= 7 at Thu Jun 16 14:16:35 MDT 2016
036 bl315 TBL.sh f10c3dm outfrq3s+cam3 9s .....................................................FAIL! rc= 7 at Thu Jun 16 14:21:01 MDT 2016
041 bl317 TBL.sh f10c5cdm outfrq3s 9s .........................................................FAIL! rc= 7 at Thu Jun 16 14:33:20 MDT 2016
045 bl319 TBL.sh f10c5nonedm outfrq3s_bam 9s ..................................................FAIL! rc= 7 at Thu Jun 16 14:38:45 MDT 2016
048 bl320 TBL.sh f10c5nonedm rad_diag 9s ......................................................FAIL! rc= 7 at Thu Jun 16 14:39:58 MDT 2016
051 bl321 TBL.sh f10c5cdm atrain 9s ...........................................................FAIL! rc= 7 at Thu Jun 16 14:42:52 MDT 2016
054 bl334 TBL.sh f10c5udm outfrq3s 9s .........................................................FAIL! rc= 7 at Thu Jun 16 14:49:01 MDT 2016
057 bl354 TBL.sh f1.9c5.4dm outfrq3s_npryz 9s .................................................FAIL! rc= 7 at Thu Jun 16 15:06:29 MDT 2016
060 bl420 TBL.sh f4c5mam7dm outfrq3s 9s .......................................................FAIL! rc= 7 at Thu Jun 16 15:15:08 MDT 2016
063 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Thu Jun 16 15:23:01 MDT 2016
066 bl734 TBL.sh h16c5naqdm outfrq3s_bam+aquaplanet_cam5 9s ...................................FAIL! rc= 7 at Thu Jun 16 15:33:40 MDT 2016
068 bl011 TBL.sh f4c5portdm outfrq24h_port 2d .................................................FAIL! rc= 7 at Thu Jun 16 15:39:30 MDT 2016
Changes are due to answer changes in external libraries
===============================================================
===============================================================
Tag name: cam5_4_73
Originator(s): fvitt, tilmes
Date: 16 Jun 2016
One-line Summary: Modifications to VBS SOA model
Purpose of changes:
Provide capability for a simpler VBS SOA chemistry scheme (fewer species).
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
A components/cam/test/system/config_files/f0.9c5.5mam4vbsh
- configure options for VBS SOA chemsitry
A components/cam/bld/namelist_files/use_cases/2000_trop_strat_vbs_cam55.xml
A components/cam/bld/namelist_files/use_cases/1850_trop_strat_vbs_cam55.xml
- new use case files for VBS SOA chemistry
List all existing files that have been modified, and describe the changes:
M components/cam/bld/build-namelist
- VOC emis update
- removed add_default for "use_rdggamma_file" -- this namelist option does not exist
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
M components/cam/bld/namelist_files/master_gas_drydep_list.xml
M components/cam/bld/namelist_files/master_gas_wetdep_list.xml
- changes to accomadate more SOA* names
M components/cam/src/chemistry/modal_aero/modal_aero_gasaerexch.F90
M components/cam/src/chemistry/mozart/mo_usrrxt.F90
- GLYOXAL_aer user reaction rate added
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/m_rxt_id.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_nln_matrix.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_lu_factor.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_sim_dat.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_indprd.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/m_spc_id.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/chem_mech.doc
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/chem_mech.in
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_lu_solve.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_setrxt.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_adjrxt.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_rxt_rates_conv.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/chem_mods.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_prod_loss.F90
M components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_lin_matrix.F90
- SOA chemistry updates
M components/cam/src/chemistry/mozart/rate_diags.F90
- use larger sized temperary strings to process rate sums that have a
large number of terms
M components/cam/test/system/input_tests_master
- added a test for VBS SOA chemsitry
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All Pass
hobart/nag: All Pass
hobart/pgi: All Pass
===============================================================
===============================================================
Tag name: cam5_4_72
Originator(s): fvitt
Date: 15 Jun 2016
One-line Summary: CAM Makefile change
Purpose of changes:
Decrease the intel compiler optimization level for the uwshcu module
to be consistent with CIME's intel build
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
. components/cam/bld/Makefile.in
- remove uwshcu.F90 from the intel compiler O3 optimization group
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
024 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...............................................FAIL! rc= 7 at Wed Jun 15 16:52:05 MDT 2016
068 bl740 TBL.sh h30c5h outfrq3s_lb0 9s .......................................................FAIL! rc= 7 at Wed Jun 15 17:11:58 MDT 2016
- these baseline failures are expected due to change in intel compiler
optimization option for the uwshcu module used in cam5 physics
hobart/nag: All Pass
hobart/pgi: All Pass
===============================================================
===============================================================
Tag name: cam5_4_71
Originator(s): fvitt
Date: 14 Jun 2016
One-line Summary: update cime external
Purpose of changes:
Update to cime4.5.15, which includes extensions to the deposition tables in
driver_cpl/shr/seq_drydep_mod.F90 needed for upcoming changes to chemistry
VBS SOA aerosol package
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: none
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
. SVN_EXTERNAL_DIRECTORIES
- update to cime4.5.15
. components/cam/bld/configure
- include stub ESP component in stand-alone cam build
. components/cam/test/system/TCB_ccsm.sh
- set PE layout for ESP component
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
004 bl132 TBL.sh e48c4dh ghgrmp_e8+1850_cam4 9s ...............................................FAIL! rc= 7 at Tue Jun 14 17:36:48 MDT 2016
008 bl331 TBL.sh f4c4dh co2rmp+1850_cam4 9s ...................................................FAIL! rc= 7 at Tue Jun 14 17:36:51 MDT 2016
011 bl353 TBL.sh f1.9c5.4dh outfrq3s 9s .......................................................FAIL! rc= 7 at Tue Jun 14 17:37:00 MDT 2016
015 bl357 TBL.sh f1.9c5.5h outfrq3s+1850_cam55 9s .............................................FAIL! rc= 7 at Tue Jun 14 17:37:12 MDT 2016
018 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Tue Jun 14 17:37:23 MDT 2016
022 bl373 TBL.sh f1.9c4mozdh outfrq3s+2000_cam4_trop_chem 9s ..................................FAIL! rc= 7 at Tue Jun 14 17:37:46 MDT 2016
024 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...............................................FAIL! rc= 7 at Tue Jun 14 17:37:50 MDT 2016
027 bl485 TBL.sh f1.9c5.5wtmam4h outfrq3s+waccm_tsmlt_2000_cam55 9s ...........................FAIL! rc= 7 at Tue Jun 14 17:40:32 MDT 2016
030 bl390 TBL.sh f1.9c4wmxh outfrq3s+waccmx_2000_cam4 9s ......................................FAIL! rc= 7 at Tue Jun 14 17:42:12 MDT 2016
033 bl397 TBL.sh f1.9c4cm outfrq3s+1850-2005_cam4 9s ..........................................FAIL! rc= 7 at Tue Jun 14 17:42:21 MDT 2016
036 bl421 TBL.sh f1.9c5mam4dh outfrq3s 9s .....................................................FAIL! rc= 7 at Tue Jun 14 17:42:28 MDT 2016
039 bl429 TBL.sh f1.9c4wmxh outfrq3s+waccmxi_2000_cam4 9s .....................................FAIL! rc= 7 at Tue Jun 14 17:44:11 MDT 2016
041 bl430 TBL.sh f1.9c5mzmam3dh outfrq3s_NEUwetdep 9s .........................................FAIL! rc= 7 at Tue Jun 14 17:44:26 MDT 2016
044 bl440 TBL.sh f1.9c4soadh outfrq3s+soa_chem_megan_emis 9s ..................................FAIL! rc= 7 at Tue Jun 14 17:44:57 MDT 2016
047 bl470 TBL.sh fsd1.9c5.4wtclbh outfrq3s_sd+sd_waccm5_geos5 9s ..............................FAIL! rc= 7 at Tue Jun 14 17:47:58 MDT 2016
050 bl482 TBL.sh f1.9c5wscmam3dh outfrq3s_2005+waccm_sc_1955-2005_cam5 9s .....................FAIL! rc= 7 at Tue Jun 14 17:48:17 MDT 2016
054 bl387 TBL.sh f1.9c4dm outfrq3s+1850-2005_cam4 9s ..........................................FAIL! rc= 7 at Tue Jun 14 17:48:21 MDT 2016
057 bl388 TBL.sh f1.9c4m outfrq24h+1850_cam4 2d ...............................................FAIL! rc= 7 at Tue Jun 14 17:48:25 MDT 2016
060 bl731 TBL.sh h16c3aqdh aqua+aquaplanet_cam3 9s ............................................FAIL! rc= 7 at Tue Jun 14 17:48:28 MDT 2016
063 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ........................................FAIL! rc= 7 at Tue Jun 14 17:48:35 MDT 2016
068 bl740 TBL.sh h30c5h outfrq3s_lb0 9s .......................................................FAIL! rc= 7 at Tue Jun 14 17:51:22 MDT 2016
072 bl751 TBL.sh h30c4aqdm outfrq3s+aquaplanet_cam4 9s ........................................FAIL! rc= 7 at Tue Jun 14 17:54:19 MDT 2016
Baseline failures are expected due to a correction to shr_orb_cosz (in cime4.5.11)
076 eq993 TEQ_ccsm.sh ne30_ne30 FC5 h30c5h fcase 9s ...........................................FAIL! rc= 7 at Tue Jun 14 17:56:41 MDT 2016
This failure is due to CIME's change in the intel compiler optimization level of uwshcu.F90
hobart/nag:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s ...........................................................FAIL! rc= 7 at Tue Jun 14 11:47:18 MDT 2016
010 bl112 TBL.sh e8c5dm ghgrmp_e8 9s ..........................................................FAIL! rc= 7 at Tue Jun 14 11:50:31 MDT 2016
014 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Tue Jun 14 11:52:39 MDT 2016
018 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Tue Jun 14 11:57:19 MDT 2016
025 bl313 TBL.sh f10c3aqdm outfrq3s+aquaplanet_cam3 3s ........................................FAIL! rc= 7 at Tue Jun 14 12:02:02 MDT 2016
028 bl314 TBL.sh f10c4wscdm outfrq3s 9s .......................................................FAIL! rc= 7 at Tue Jun 14 12:05:26 MDT 2016
031 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .........................................FAIL! rc= 7 at Tue Jun 14 12:07:52 MDT 2016
034 bl318 TBL.sh f10c4cdm sat_hist 9s .........................................................FAIL! rc= 7 at Tue Jun 14 12:13:10 MDT 2016
037 bl324 TBL.sh f4c4rrtmgdm outfrq3s 9s ......................................................FAIL! rc= 7 at Tue Jun 14 12:15:37 MDT 2016
040 bl325 TBL.sh f10c5t5mdmsc outfrq3s_subcol 9s ..............................................FAIL! rc= 7 at Tue Jun 14 12:19:24 MDT 2016
046 bl338 TBL.sh f10c5dm rad_diag_mam 9s ......................................................FAIL! rc= 7 at Tue Jun 14 12:22:17 MDT 2016
049 bl354 TBL.sh f1.9c5.4dm outfrq3s_npryz 9s .................................................FAIL! rc= 7 at Tue Jun 14 12:33:57 MDT 2016
052 bl356 TBL.sh fsd1.9c5clbdm off1.9x2.5 9s ..................................................FAIL! rc= 7 at Tue Jun 14 12:52:16 MDT 2016
055 bl368 TBL.sh f1.9c5mg2dm outfrq3s_macmic2_classnuc 9s .....................................FAIL! rc= 7 at Tue Jun 14 13:02:16 MDT 2016
058 bl369 TBL.sh f1.9c5clbdm outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Tue Jun 14 13:14:44 MDT 2016
061 bl734 TBL.sh h16c5naqdm outfrq3s_bam+aquaplanet_cam5 9s ...................................FAIL! rc= 7 at Tue Jun 14 13:20:24 MDT 2016
064 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Tue Jun 14 13:24:44 MDT 2016
067 bl426 TBL.sh f10c4wmxdm outfrq3s_ionos 9s .................................................FAIL! rc= 7 at Tue Jun 14 13:30:29 MDT 2016
070 bl471 TBL.sh f10c5wmam3dm outfrq3s_modalstrat 9s ..........................................FAIL! rc= 7 at Tue Jun 14 13:37:10 MDT 2016
073 bl805 TBL.sh f4c5carmtt1dm outfrq3s_carma 9s ..............................................FAIL! rc= 7 at Tue Jun 14 13:51:04 MDT 2016
076 bl011 TBL.sh f4c5portdm outfrq24h_port 2d .................................................FAIL! rc= 7 at Tue Jun 14 13:55:41 MDT 2016
hobart/pgi:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s ...........................................................FAIL! rc= 7 at Tue Jun 14 11:50:31 MDT 2016
010 bl112 TBL.sh e8c5dm ghgrmp_e8 9s ..........................................................FAIL! rc= 7 at Tue Jun 14 11:57:42 MDT 2016
012 bl113 TBL.sh e8c5paqdm aqpgro+aquaplanet_cam5 3s ..........................................FAIL! rc= 7 at Tue Jun 14 12:02:39 MDT 2016
016 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Tue Jun 14 12:04:42 MDT 2016
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Tue Jun 14 12:15:45 MDT 2016
030 bl313 TBL.sh f10c3aqdm outfrq3s+aquaplanet_cam3 3s ........................................FAIL! rc= 7 at Tue Jun 14 12:26:31 MDT 2016
033 bl314 TBL.sh f10c4wscdm outfrq3s 9s .......................................................FAIL! rc= 7 at Tue Jun 14 12:32:43 MDT 2016
036 bl315 TBL.sh f10c3dm outfrq3s+cam3 9s .....................................................FAIL! rc= 7 at Tue Jun 14 12:37:30 MDT 2016
038 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .........................................FAIL! rc= 7 at Tue Jun 14 12:42:51 MDT 2016
041 bl317 TBL.sh f10c5cdm outfrq3s 9s .........................................................FAIL! rc= 7 at Tue Jun 14 12:50:14 MDT 2016
045 bl319 TBL.sh f10c5nonedm outfrq3s_bam 9s ..................................................FAIL! rc= 7 at Tue Jun 14 12:56:13 MDT 2016
048 bl320 TBL.sh f10c5nonedm rad_diag 9s ......................................................FAIL! rc= 7 at Tue Jun 14 12:57:49 MDT 2016
051 bl321 TBL.sh f10c5cdm atrain 9s ...........................................................FAIL! rc= 7 at Tue Jun 14 13:00:24 MDT 2016
054 bl334 TBL.sh f10c5udm outfrq3s 9s .........................................................FAIL! rc= 7 at Tue Jun 14 13:07:07 MDT 2016
057 bl354 TBL.sh f1.9c5.4dm outfrq3s_npryz 9s .................................................FAIL! rc= 7 at Tue Jun 14 13:25:06 MDT 2016
060 bl420 TBL.sh f4c5mam7dm outfrq3s 9s .......................................................FAIL! rc= 7 at Tue Jun 14 13:34:13 MDT 2016
063 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Tue Jun 14 13:42:19 MDT 2016
066 bl734 TBL.sh h16c5naqdm outfrq3s_bam+aquaplanet_cam5 9s ...................................FAIL! rc= 7 at Tue Jun 14 13:53:11 MDT 2016
068 bl011 TBL.sh f4c5portdm outfrq24h_port 2d .................................................FAIL! rc= 7 at Tue Jun 14 13:59:05 MDT 2016
Baseline failures are expected due to a correction to shr_orb_cosz (in cime4.5.11)
===============================================================
===============================================================
Tag name: cam5_4_70
Originator(s): eaton
Date: Tue Jun 14 07:44:27 MDT 2016
One-line Summary: fix for FV in serial mode
Purpose of changes:
. The previous commit broke serial mode for FV due to some missing
initializers in the internal FV state object.
Bugs fixed (include bugzilla ID):
. Fix FV in serial mode (actually non-mpi mode; the fix applies to a pure
OMP configuration as well as serial).
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
components/cam/src/dynamics/fv/dynamics_vars.F90
. initialize t_fvdycore_grid components (jfirst, jlast, ifirstxy, ilastxy,
jfirstxy, jlastxy, kfirst, klast, klastp) with values appropriate for
non-decomposed grids
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS
yellowstone/intel/aux_cam: Ran test by hand: F2000_DONOTUSE test is BFB with run made with cam5_4_68
hobart/nag: All PASS
hobart/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB. CAM doesn't have any regression
tests for FV serial which is why this was missed in the previous commit.
===============================================================
===============================================================
Tag name: cam5_4_69
Originator(s): eaton
Date: Fri Jun 10 07:40:19 MDT 2016
One-line Summary: mods towards unifying initial and restart code paths
Purpose of changes:
. Use a single filehandle to point to either the initial file or the
primary restart file. Move the call to open the initial or restart file
to right after the namelist read. The previous individual calls to open
either the initial or restart file are removed from the conditional on
whether it's an initial run or not.
. Initialize the grids immediately after the initial or restart file is
open. This requires a large refactoring in the dynamics interface code
for all 4 dycores. The initialization of the grid and the decomposition
are moved to the dyn_grid_init method. Move the call to dyn_grid_init
outside of the conditions on whether the run is initial or restart.
. Move the call to initialize the physics grid decomposition to be
immediately after the call to dyn_grid_init (outside the conditional on
whether the run is initial or restart).
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist:
. move cam_branch_file to group cam_initfiles_nl
. move ct_overlap, trac_decomp from dyn_fv_inparm to spmd_fv_inparm
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
components/cam/src/control/readinitial.F90
. functionality moved to dyn_grid_init methods
components/cam/src/control/startup_initialconds.F90
. functionality moved to dyn_init methods
components/cam/src/dynamics/eul/inidat.F90
. move all routines to dyn_comp.F90
components/cam/src/dynamcis/eul/inital.F90
. eliminate the cam_initial interface and call dyn_init directly from
cam_init.
components/cam/src/dynamics/eul/initcom.F90
. functionality moved to dyn_grid_init
components/cam/src/dynamics/fv/fv_control_mod.F90
. eliminate unused data and make use of the t_fvdycore_state object for
everything else
components/cam/src/dynamics/fv/inidat.F90
. move all routines to dyn_comp.F90
components/cam/src/dynamcis/fv/inital.F90
. eliminate the cam_initial interface and call dyn_init directly from
cam_init.
components/cam/src/dynamics/fv/initcom.F90
. functionality moved to dyn_grid_init
components/cam/src/dynamics/se/inidat.F90
. read_inidat moved to dyn_comp
components/cam/src/dynamcis/se/inital.F90
. eliminate the cam_initial interface and call dyn_init directly from
cam_init.
components/cam/src/dynamics/se/initcom.F90
. this was a stub module that's no longer needed
components/cam/src/dynamics/sld/inidat.F90
. move all contents to dyn_comp module
components/cam/src/dynamics/sld/inital.F90
. code moved to dyn_grid_init, dyn_init, cam_init
components/cam/src/dynamics/sld/initcom.F90
. code moved to dyn_grid_init
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
components/cam/bld/configure
. remove setting the CPP macro STAGGERED. It is no longer used.
components/cam/bld/namelist_files/namelist_definition.xml
. move cam_branch_file to group cam_initfiles_nl
. move ct_overlap, trac_decomp from dyn_fv_inparm to spmd_fv_inparm
components/cam/src/control/cam_comp.F90
. cam_init:
- move call to cam_initfiles_open from inside if(initial_run) conditional
to right after read_namelist.
- call dyn_grid_init after cam_initfiles_open
- add calls to phys_grid_init and chem_surfvals_init
components/cam/src/control/cam_initfiles.F90
. add cam_branch_file to cam_initfiles_nl namelist group.
. set name of restart pointer file here (moved from cam_restart.F90). Name
is public so cam_restart module can update contents as restart files are
written.
. set name of initial restart file for restart or branch runs (moved from
cam_restart).
. set caseid_prev here (moved from cam_restart). Since initial restart
files are opened here this module will do consistency check on caseid in
case of branch run. Also is responsible for providing this name to the
component interface as long as the separate surface restart file is
needed.
. add the get_restcase and get_restartdir methods. This module is now
responsible for the data that those accessor functions provide.
. remove call to read_initial from cam_initfiles_open. Functionality moved
to dycore specific dyn_grid_init
components/cam/src/control/cam_restart.F90
. remove cam_restart_nl namelist group and cam_restart_readnl routine.
. remove opening of restart file (moved to cam_initfiles).
. modify write_rest_pfile
- take the restart filename as an argument
- access the name of the pointer file from cam_initfiles
- remove adding the names of other restart files as comments. No other
component does this and it adds unnecessary complexity.
. move the get_restcase and get_restartdir methods to cam_initfiles
. cam_read_restart
- remove call to initcom
- remove call to ref_pres_init
- remove calls to phys_grid_init, chem_surfvals_init, spmdbuf
components/cam/src/control/runtime_opts.F90
. remove call to cam_restart_readnl
components/cam/src/cpl/atm_comp_esmf.F90
components/cam/src/cpl/atm_comp_mct.F90
. update use association of get_restcase and get_restartdir to use the
methods in cam_initfiles
components/cam/src/dynamics/eul/dyn_comp.F90
. dyn_init
- move subroutine trunc() to module dyn_grid
- remove calls to trunc, spmdinit_dyn, hycoef_init (moved to dyn_grid_init)
- add calls to initialize_prognostics, scanslt_alloc, spmdbuf, and
initial_conds. The initial_conds call will be replaces by a
read_inidat call which must be inside a conditional to only be called
on an initial run.
components/cam/src/dynamics/eul/dyn_grid.F90
. change default module access to private
. add public statement for methods
. turn the initgrid method into dyn_grid_init
- add check on grid size previously done in read_initial
- add call to trunc()
- add call to spmdinit_dyn
- add call to hycoef_init
- add call to ref_pres_init
- remove dyn_grid_get_pref
. move subroutine trunc here from dyn_comp
. remove unused get_block_ldof_d
components/cam/src/dynamics/eul/restart_dynamics.F90
. read_restart_dynamics
- remove calls to initialize_prognostics, scanslt_alloc which are now
done in dyn_init
components/cam/src/dynamics/fv/cd_core.F90
. pass div24del2flag, del2coef through arg list rather than access from
fv_control_mod.
components/cam/src/dynamics/fv/dyn_comp.F90
. dyn_readnl
- remove namelist variables fv_ct_overlap and fv_trac_decomp
. dyn_register
- add this method and put the pbuf add calls there. Will need to add
this method to the dyn_comp modules of all dycores.
. dyn_init
- move hycoef_init call to dyn_grid_init
- move setting of namelist values in t_fvdycore_state object into
dyn_readnl
- move spmdinit_dyn call to dyn_grid_init
- clean up after moving all the grid initialization to dyn_grid_init.
- remove unused arg fh and replace the dyn_state arg by local pointer
and get state using get_dyn_state
. dyn_run
- access ct_overlap, trac_decomp from grid object rather than
fv_control_mod.
- access div24del2flag, del2coef from dyn_state and pass through cd_core
interface.
components/cam/src/dynamics/fv/dyn_grid.F90
. change default module access to private
. add public statement for methods
. turn the initgrid method into dyn_grid_init
- add check on grid size previously done in read_initial
- add call to hycoef_init
- add call to ref_pres_init
- remove dyn_grid_get_pref
- add call to spmdinit_dyn (remove SPMD conditional)
. remove unused get_block_ldof_d
. dyn_grid_init
- move the call to grid_vars_init here (was dynpkg_init called from
dynamics_init which has been removed)
- move call to define_cam_grids here (from dyn_comp::dyn_init)
- pass fft_flt through pftinit arg list
. fix use associations of pmgrid (most values now live only in grid
object).
components/cam/src/dynamics/fv/dynamics_vars.F90
. remove the dynamics_init subroutine. It's functionality is replaced by
setting the components of the grid object as soon as possible, and by
making direct calls to dynpkg_init and spmd_vars_init.
. dynpkg_init -- rename to grid_vars_init
- move the setting of ng_c, ng_d, ng_s into spmd_dyn::spmdinit_dyn. The
only info needed to set these vars is jord which is read in
dyn_readnl.
. spmd_vars_init
- remove the dummy args previously used to set components of the grid
object, but which have already been set elsewhere. Turn those args
into local variables set from the grid object components so don't need
to modify much internal code.
. remove unused components from t_fvdycore_state: run_times, num_calls,
dotime, dodyn
. add new components to t_fvdycore_state that were previously in the
fv_control_mod module: fft_flt, div24del2flag, del2coef
. add new components to t_fvdycore_grid that were previously in
fv_control_mod: ct_overlap, trac_decomp
. spmd_vars_init
- access ct_overlap, trac_decomp from grid object instead of fv_control_mod
. grid_vars_init
- pass fft_flt through arg list so can be passed to pftinit
components/cam/src/dynamics/fv/gravity_waves_sources.F90
. replace use of XY decomp info in pmgrid by grid object
components/cam/src/dynamics/fv/metdata.F90
. add only qualifier to pmgrid use statement
components/cam/src/dynamics/fv/pft_module.F90
. change fft_flt to be module data rather than access via fv_control_mod
. pftinit
- add fft_flt as input arg for initializing module data
components/cam/src/dynamics/fv/restart_dynamics.F90
. replace use of XY decomp info in pmgrid by grid object
. remove tmass0
. read_restart_dynamics
- update interface to dyn_init call
components/cam/src/dynamics/fv/spmd_dyn.F90
. Narrow the scope of the SPMD ifdef so the readnl and init methods can
always be called. There is currently a bunch of code in dyn_init that
belongs in these spmd_dyn methods.
. spmd_readnl
- put the namelist data into the grid object. Anything in the grid
object should be removed from pmgrid: npr_y, npr_z, npes_yz, nprxy_x,
nprxy_y, npes_xy, twod_decomp
- add fv_ct_overlap, fv_trac_decomp. Store values in grid object.
. spmdinit_dyn
- remove calls to parsplit. these are duplicated in spmd_vars_init and
those calls are the ones setting the communicators in the grid object
- put decomp data directly into grid object and remove it from pmgrid:
beglat, endlat, beglev, endlev, endlevp, endlevp1
beglonxy, endlonxy, beglatxy, endlatxy
- remove code to compute nmostlat and smostlat which appears to be
unused.
- move code here to set imxy, jmxy, jmyz, kmyz. Was in dyn_init. Will
pass these arrays through spmd_vars_init interface.
components/cam/src/dynamics/fv/stepon.F90
. fix use associations of pmgrid
. stepon_init
- remove check for STAGGERED
components/cam/src/dynamics/se/dyn_comp.F90
. dyn_register
- add this routine. Move the pbuf_add_field calls for FRONTGF and
FRONTGA here.
. dyn_init1
- remove this routine. move the code in this routine to dyn_grid_init,
except for the init of the dyn_in and dyn_out objects. This needs to
be moved to the dyn_init2 routine because dyn_in/dyn_out are not args
to dyn_grid_init. Also move create_native_mapping_files to dyn_init2
since otherwise a circular dependency occurs between native_mapping and
dyn_grid.
. public data timelevel, dom_mt, hvcoord moved to dyn_grid
. dyn_init2
- change name to dyn_init
- call to read_inidat here and eliminate the call to initial_conds. Only
make this call for an initial run for now. Eventually should have the
same code reading either the initial or restart file.
. read_inidat
- moved here from inidat module
components/cam/src/dynamics/se/dyn_grid.F90
. dyn_grid_init
- add check for size of grid in initial file
components/cam/src/dynamics/se/gravity_waves_sources.F90
. access dom_mt, hvcoord from dyn_grid, not dyn_comp
components/cam/src/dynamics/se/restart_dynamics.F90
. read_restart_dynamics
- remove call to dyn_init1
- access elem via dyn_grid since dyn_in isn't initialized until the call
to dyn_init (renamed from dyn_init2)
components/cam/src/dynamics/se/stepon.F90
. access timelevel from dyn_grid, not dyn_comp
components/cam/src/dynamics/sld/dyn_comp.F90
. move subroutine trunc() to module dyn_grid
. dyn_init
- add dyn_in and dyn_out as intent(out) dummy args
- remove calls to trunc, spmdinit_dyn, hycoef_init, and define_cam_grids
(moved to dyn_grid_init)
components/cam/src/dynamics/sld/dyn_grid.F90
. add private attribute
. add public statement for methods
. remove unused get_block_ldof_d
. remove dyn_grid_get_pref
. add dyn_grid_init
- add check on grid size previously done in read_initial
- add call to trunc() and the trunc subroutine (moved from dyn_init)
- add call to spmdinit_dyn (moved from dyn_init)
- add call to hycoef_init (moved from dyn_init)
- add call to ref_pres_init
- add call to define_cam_grids (moved from dyn_init)
components/cam/src/dynamics/sld/restart_dynamics.F90
. read_restart_dynamics
- update dyn_init actual args
- remove calls to initialize_prognostics and slt_alloc
components/cam/src/dynamics/sld/scanslt.F90
. remove initializer for nlonex. add it to the declaration statement in
the module data.
components/cam/src/physics/cam/physpkg.F90
. call dyn_register from phys_register
components/cam/src/physics/cam/ref_pres.F90
. refactor ref_pres_init to take the reference pressures as arguments
rather than via use of a dyn_grid method. That avoids creating a
circular dependency since we want to call ref_pres_init from dyn_grid.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS
hobart/nag: All PASS
hobart/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_4_68
Originator(s): fvitt
Date: 8 Jun 2016
One-line Summary: Energetic particle inputs and WACCM updates
Purpose of changes:
- dynamically update geomagnetic coordinates on yearly basis
- read IGRF coefficient data from netCDF file
- energetic particle inputs on geomagnetic coordinates
- add D-region chemistry
- rename middle atmosphere "waccm_mozart" chemistry mechanisms as "waccm_ma"
- update aurora module to be consistent with TIEGCM
- chemistry changes associated the excited states of O+ in the waccm middle
atmosphere chemistry mechanism -- used by WACCMX and WACCM4
- in iondrag set electron density as the sum of the densities O+, O2+, and NO+
- ion drift velocities history fields 'UE', 'VE' renamed to 'UI', 'VI'
- add chemistry associated with O+ excited states to WACCM middle atmosphere
chemistry
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
D components/cam/src/chemistry/pp_waccmx_mozart
- removed -- waccmx uses pp_waccm_ma
D components/cam/src/chemistry/pp_waccm_mozart_mam3
D components/cam/src/chemistry/pp_waccm_mozart_mam4
D components/cam/src/chemistry/pp_waccm_mozart
D components/cam/src/chemistry/pp_waccm_mozart_sulfur
- renamed "waccm_mozart" as "waccm_ma"
D components/cam/src/chemistry/mozart/mo_solarproton.F90
D components/cam/src/chemistry/mozart/spedata.F90
- removed -- SPE functionality replaced by epp_ionization module
List all subroutines added and what they do:
A components/cam/bld/namelist_files/use_cases/sd_waccm5_mad_geos5.xml
- new use case for waccm d-region
A components/cam/src/chemistry/pp_waccm_ma_sulfur
- renamed "waccm_mozart" as "waccm_ma"
- correction to "OCS + OH" reaction
A components/cam/src/chemistry/pp_waccm_ma
A components/cam/src/chemistry/pp_waccm_ma_mam3
A components/cam/src/chemistry/pp_waccm_ma_mam4
- renamed "waccm_mozart" as "waccm_ma"
A components/cam/src/chemistry/pp_waccm_mad_mam4
- new mechanism for D-region ion chemistry
A components/cam/src/chemistry/mozart/epp_ionization.F90
- new module to read in EPP data sets (on geo-mag grid)
A components/cam/test/system/nl_files/outfrq24h_epp
A components/cam/test/system/nl_files/outfrq3s_epp
A components/cam/test/system/config_files/fsd1.9c5.4wmdionh
A components/cam/test/system/config_files/f10c5wmdiond
- new tests for waccm d-region chemistry
List all existing files that have been modified, and describe the changes:
M components/cam/src/chemistry/bulk_aero/aero_model.F90
M components/cam/src/chemistry/mozart/mo_setext.F90
- changes for the new EPP inputs
- EPP forcings for D-region chemistry
M components/cam/src/chemistry/mozart/charge_neutrality.F90
- refactored and expanded for D-region ions
M components/cam/src/chemistry/mozart/mo_photo.F90
- D-region ionizations added
M components/cam/src/chemistry/mozart/mo_aurora.F90
- made consistent with TIEGCM
- removed dead SPE and HEE code
- code clean up
M components/cam/src/chemistry/mozart/gcr_ionization.F90
- split out HOx production code
M components/cam/src/chemistry/utils/mo_msis_ubc.F90
- fixed restart bug in MSIS when run start time is not mid-night UT
- added diagnostics
M components/cam/src/chemistry/mozart/upper_bc.F90
- fixed restart bug in MSIS when run start time is not mid-night UT
M components/cam/src/chemistry/mozart/wei96.F90
M components/cam/src/chemistry/mozart/mo_waccm_hrates.F90
M components/cam/src/chemistry/mozart/mo_chm_diags.F90
M components/cam/src/chemistry/mozart/spehox.F90
M components/cam/src/chemistry/mozart/efield.F90
- code clean up
M components/cam/src/chemistry/mozart/mo_apex.F90
- variable geo-magnetic year implimented
M components/cam/src/chemistry/mozart/mo_usrrxt.F90
- reactions rates added for D-region chemistry
- reactions rates added for quenching of excited states of O+
M components/cam/src/chemistry/mozart/mo_chemini.F90
- removed references to spedata
M components/cam/src/chemistry/mozart/chemistry.F90
- removed references to spedata
- use new epp_ionization module
M components/cam/src/chemistry/mozart/exbdrift.F90
- pbuf name change for ion drift velocities
- ion drift velocities now 3-dimensional
M components/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
- enforce charge neutrality at the end of chemistry time step
M components/cam/src/chemistry/modal_aero/sox_cldaero_mod.F90
M components/cam/src/chemistry/modal_aero/aero_model.F90
- fixed bug in diagnostics array dimensions
M components/cam/src/chemistry/utils/tracer_data.F90
- improvement in error messages
M components/cam/src/chemistry/utils/mo_solar_parms.F90
- removed unused code for hemispheric power
- added log messages
M components/cam/src/chemistry/utils/apex.F90
- read IGRF data from netcdf file
- added ability to re-initialize for change in geo-magnetic date
- code clean up
M components/cam/src/physics/cam/physpkg.F90
M components/cam/src/physics/waccmx/ionosphere.F90
M components/cam/src/physics/waccm/iondrag.F90
M components/cam/test/system/input_tests_master
M components/cam/test/system/tests_waccm_hybrid
M components/cam/test/system/tests_waccm_mpi
M components/cam/test/system/config_files/f1.9c4wmm
M components/cam/test/system/config_files/f10c4wmxdm
M components/cam/test/system/config_files/f1.9c5wmam3h
M components/cam/test/system/config_files/h5c4wmdm
M components/cam/test/system/config_files/f4c5wmclbdh
M components/cam/test/system/config_files/f1.9c4wmxdh
M components/cam/test/system/config_files/f1.9c4wmdh
M components/cam/test/system/config_files/fsd1.9c4wcarmsuldm
M components/cam/test/system/config_files/f1.9c5wmclbdh
M components/cam/test/system/config_files/fsd1.9c4wmh
M components/cam/test/system/config_files/f1.9c4carmsuldh
M components/cam/test/system/config_files/f1.9c4wcarmbc_m
M components/cam/test/system/config_files/fsd1.9c4wmdh
M components/cam/test/system/config_files/f4c4wmdh
M components/cam/test/system/config_files/f10c5wmam3dm
M components/cam/test/system/config_files/f4c4wmh
M components/cam/test/system/config_files/f1.9c5wmam3dh
M components/cam/test/system/config_files/f10c4wmdm
M components/cam/test/system/config_files/f1.9c4wmxh
M components/cam/test/system/config_files/f1.9c4wmh
M components/cam/test/system/TPF.sh
M components/cam/bld/namelist_files/use_cases/waccm_1850_cam5.xml
M components/cam/bld/namelist_files/use_cases/volc_waccm_tsmlt_megan_cam5.xml
M components/cam/bld/namelist_files/use_cases/waccmx_1996_cam4.xml
M components/cam/bld/namelist_files/use_cases/sd_waccm_sulfur.xml
M components/cam/bld/namelist_files/use_cases/waccm_tsmlt_1850_cam55.xml
M components/cam/bld/namelist_files/use_cases/waccmx_2000_cam4.xml
M components/cam/bld/namelist_files/use_cases/2000_ccmi_refc1_waccm_tsmlt.xml
M components/cam/bld/namelist_files/use_cases/volc_waccm_ma_cam5.xml
M components/cam/bld/namelist_files/use_cases/waccmxi_2000_cam4.xml
M components/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_rcp85_waccm_ma.xml
M components/cam/bld/namelist_files/use_cases/waccm_refb1.xml
M components/cam/bld/namelist_files/use_cases/rcp6.0_waccm_tsmlt_nomegan_cam5.xml
M components/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_waccm_ma.xml
M components/cam/bld/namelist_files/use_cases/1950-2010_ccmi_refc1_waccm_tsmlt.xml
M components/cam/bld/namelist_files/use_cases/1850_waccm_tsmlt_cam5.xml
M components/cam/bld/namelist_files/use_cases/waccm_1850-2005_cam4.xml
M components/cam/bld/namelist_files/use_cases/waccm_2000_cam4.xml
M components/cam/bld/namelist_files/use_cases/waccm_2000_cam5.xml
M components/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp45.xml
M components/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp85.xml
M components/cam/bld/namelist_files/use_cases/2007-2011_htap_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/waccm_tsmlt_2000_cam55.xml
M components/cam/bld/namelist_files/use_cases/sd_1975-2010_ccmi_refc1_waccm_ma.xml
M components/cam/bld/namelist_files/use_cases/sd_waccm_geos5.xml
M components/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_waccm_tsmlt.xml
M components/cam/bld/namelist_files/use_cases/waccm_sc_1955-2005_cam5.xml
M components/cam/bld/namelist_files/use_cases/sd_waccm5_geos5.xml
M components/cam/bld/namelist_files/use_cases/volc_waccm_tsmlt_nomegan_cam5.xml
M components/cam/bld/namelist_files/use_cases/waccm_tsmlt_amip_cam55.xml
M components/cam/bld/namelist_files/use_cases/1950-2010_ccmi_refc1_waccm_ma.xml
M components/cam/bld/namelist_files/use_cases/sd_1975-2010_ccmi_refc1_waccm_tsmlt.xml
M components/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp26.xml
M components/cam/bld/namelist_files/use_cases/waccm_1955-2005_cam4.xml
M components/cam/bld/namelist_files/use_cases/waccm_1955-2005_cam5.xml
M components/cam/bld/namelist_files/use_cases/waccm_1974-2005_cam4.xml
M components/cam/bld/namelist_files/use_cases/waccm_carma_bc_2013_cam4.xml
M components/cam/bld/namelist_files/use_cases/waccm_1850_cam4.xml
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
- added default IGRF input file
- changes for renaming "waccm_mozart" as "waccm_ma"
M components/cam/bld/namelist_files/namelist_definition.xml
- removed SPE namelist options
- added EPP, IGRF, and fixed geo-mag year namelist options
M components/cam/bld/build-namelist
- changes for renaming "waccm_mozart" as "waccm_ma"
- sanity checks for the EPP settings
M components/cam/bld/configure
M components/cam/bld/config_files/definition.xml
M components/cam/bld/config_files/defaults_waccmx.xml
M components/cam/cime_config/config_compsets.xml
M components/cam/cime_config/testdefs/testlist_cam.xml
M components/cam/cime_config/config_component.xml
- changes for renaming "waccm_mozart" as "waccm_ma"
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
027 bl485 TBL.sh f1.9c5.5wtmam4h outfrq3s+waccm_tsmlt_2000_cam55 9s ...........................FAIL! rc= 7 at Wed Jun 8 10:43:38 MDT 2016
030 bl390 TBL.sh f1.9c4wmxh outfrq3s+waccmx_2000_cam4 9s ......................................FAIL! rc= 7 at Wed Jun 8 10:52:24 MDT 2016
039 bl429 TBL.sh f1.9c4wmxh outfrq3s+waccmxi_2000_cam4 9s .....................................FAIL! rc= 7 at Wed Jun 8 11:13:58 MDT 2016
047 bl470 TBL.sh fsd1.9c5.4wtclbh outfrq3s_sd+sd_waccm5_geos5 9s ..............................FAIL! rc= 7 at Wed Jun 8 12:00:30 MDT 2016
expected failures due to changes in geomag date and aurora updates
yellowstone/intel/aux_cam: Ran test by hand: F2000_DONOTUSE test is BFB with run made with cam5_4_67
hobart/nag:
067 bl426 TBL.sh f10c4wmxdm outfrq3s_ionos 9s .................................................FAIL! rc= 7 at Tue Jun 7 13:38:35 MDT 2016
070 bl471 TBL.sh f10c5wmam3dm outfrq3s_modalstrat 9s ..........................................FAIL! rc= 7 at Tue Jun 7 13:45:21 MDT 2016
expected failures due to changes in geomag date and aurora updates
hobart/pgi: all pass
===============================================================
===============================================================
Tag name: cam5_4_67
Originator(s): bdobbins, eaton
Date: Tue Jun 7 08:34:23 MDT 2016
One-line Summary: fix for ridge scheme; optimization in heterogeneous
freezing code
Purpose of changes:
. Add the constituent tendencies and the energy change calculation to the
ridge scheme. Cecile verified that the impact on the simulation is
small.
. Optimization from ASAP group (bdobbins) to hetfrz_classnuc.F90. Improves
CAM runtime w/ MAM4 chem by about 4%.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: about 4% speedup for
cam6.
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
components/cam/src/physics/cam/gw_drag.F90
. gw_rdg_calc
- uncomment the code that adds in the constituent tendencies
- uncomment the code that calculates the energy change used by the energy
conservation checker.
components/cam/src/physics/cam/hetfrz_classnuc.F90
. move an expensive calculation that was being duplicated for every
gridcell at every timestep into an init method.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS
yellowstone/intel/aux_cam: Ran test by hand: F2000_DONOTUSE test is ANSWER CHANGING with run made with cam5_4_66
- this is expected due to the ridge changes as this test has use_gw_rdg_beta=.true.
hobart/nag: All PASS
hobart/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB except when ridge scheme is turned
on. It is currently off by default, but will be on for cam6 once we get
the required datasets.
===============================================================
===============================================================
Tag name: cam5_4_66
Originator(s): andrew, vitt, cacraig
Date: May 27, 2016
One-line Summary: Bug fixes for use_cases (including 1850_cam55), cosp simulator and MG_SADICE and MG_SADSNOW diagnostics
Purpose of changes:
- Use case fixes for 1850 CAM5.5 and a number of WACCM use cases
- Correct estimation of diagnostic precipitation flux calculations to exactly
track precipitation in MG1 and MG2. Affects LS_FLXPRC and LS_FLXSNW used for
COSP simulator. Adds LS_FLXICE and LS_FLXLIQ for MG2 to be consistent with its methods.
- unpack sadice and sadsnow for MG_SADICE and MG_SADSNOW diagnostics
Bugs fixed (include bugzilla ID):
- bugzilla 2315: perscribed_strataero_cycle_yr being set to incorrect year
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: cacraig
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/src/physics/cam/micro_mg1_0.F90
M components/cam/src/physics/cam/micro_mg2_0.F90
- Changes to support LS_FLXPRC and LS_FLXSNW in COSP when using MG2
M components/cam/src/physics/cam/micro_mg_cam.F90
- Changes to support LS_FLXPRC and LS_FLXSNW in COSP when using MG2
- unpack sadice and sadsnow for MG_SADICE and MG_SADSNOW diagnostics
-- has no effect on climate
M components/cam/bld/namelist_files/use_cases/waccm_1850_cam5.xml
- correction to SAD input file
M components/cam/bld/namelist_files/use_cases/1850_cam55.xml
- correction to cycle year for prescribed stratospheric aerosols
M components/cam/bld/namelist_files/use_cases/waccm_tsmlt_1850_cam55.xml
M components/cam/bld/namelist_files/use_cases/waccm_tsmlt_amip_cam55.xml
M components/cam/bld/namelist_files/use_cases/volc_waccm_tsmlt_megan_cam5.xml
M components/cam/bld/namelist_files/use_cases/1850_waccm_tsmlt_cam5.xml
M components/cam/bld/namelist_files/use_cases/waccm_tsmlt_2000_cam55.xml
- correction to pom surface emissions
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB except:
015 bl357 TBL.sh f1.9c5.5h outfrq3s+1850_cam55 9s .............................................FAIL! rc= 7 at Thu May 26 16:23:50 MD
T 2016
027 bl485 TBL.sh f1.9c5.5wtmam4h outfrq3s+waccm_tsmlt_2000_cam55 9s ...........................FAIL! rc= 7 at Thu May 26 17:17:40 MD
T 2016
-- These two tests differ due to changes in the use_cases
yellowstone/intel/aux_cam: Ran test by hand: F2000_DONOTUSE test is BFB with run made with cam5_4_64
hobart/nag: all BFB
hobart/pgi or jaguar/pgi: all BFB except:
041 bl317 TBL.sh f10c5cdm outfrq3s 9s .........................................................FAIL! rc= 7 at Thu May 26 11:52:49 MDT 2016
-- answer changes expected for COSP when using MG2
===============================================================
===============================================================
Tag name: cam5_4_65
Originator(s): eaton, cacraig
Date: May, 27, 2016
One-line Summary: fix restart when ridge scheme enabled
Purpose of changes:
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/src/control/cam_initfiles.F90
M components/cam/src/physics/cam/gw_drag.F90
- fix to allow ridge scheme to restart properly
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB
hobart/nag: all BFB
hobart/pgi or jaguar/pgi: all BFB
===============================================================
===============================================================
Tag name: cam5_4_64
Originator(s): juliob, eaton
Date: Tue May 24 07:56:47 MDT 2016
One-line Summary: add beljaars and ridge to gravity wave param.
Purpose of changes:
. Add Beljaars Sub-Grid Orographic (SGO) Form drag parameterization.
- Returns drag profile and integrated stress associated with subgrid
mountains with horizontal length scales nominally below 3km. Similar
to TMS but drag is distributed in the vertical (Beljaars et al., 2003,
QJRMS).
- This option is controlled by the namelist variable do_beljaars. When
turned on it replaces the existing turbulent mountain stress
parameterization, i.e., if do_beljaars is true then do_tms must be
false.
. Add generation of gravity waves from ridges.
- This option is controlled by the namelist variable use_gw_rdg_beta.
- Code was also added for gamma ridges, but this option is not yet
validated.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist:
. add namelist variables for beljaars and for ridges. see details
below in namelist_definition.xml
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do:
components/cam/src/physics/cam/beljaars_drag.F90
. portable implementation of Beljaars scheme.
components/cam/src/physics/cam/beljaars_drag_cam.F90
. CAM interfaces to Beljaars scheme.
components/cam/src/physics/cam/gw_rdg.F90
. portable implementation of gravity wave sources from ridge scheme.
List all existing files that have been modified, and describe the changes:
components/cam/bld/build-namelist
. check that do_tms and do_beljaars not both set to true
. add defaults for use_gw_rdg_beta and use_gw_rdg_gamma
. modify default for gw_prndl when beta ridges are being used
. add defaults for new vars depending on use_gw_rdg_beta and
use_gw_rdg_gamma
components/cam/bld/namelist_files/namelist_defaults_cam.xml
. set default for do_beljaars to false
components/cam/bld/namelist_files/namelist_definition.xml
. add following vars for beta ridges to group gw_drag_nl:
- use_gw_rdg_beta
- n_rdg_beta
- effgw_rdg_beta
- effgw_rdg_beta_max
- rdg_beta_cd_llb
- trpd_leewv_rdg_beta
. add following vars for gamma ridges to group gw_drag_nl:
- use_gw_rdg_gamma
- n_rdg_gamma
- effgw_rdg_gamma
- effgw_rdg_gamma_max
- rdg_gamma_cd_llb
- trpd_leewv_rdg_gamma
- bnd_rdggm
. add do_beljaars to switch the beljaars scheme
components/cam/src/physics/cam/diffusion_solver.F90
. add intent(in) arg dragblg to compute_vdiff
. modify calcs of tauresx and tauresy
components/cam/src/physics/cam/eddy_diff_cam.F90
. add intent(in) arg dragblg to eddy_diff_tend
. add actual arg dradblj to compute_eddy_diff call
. add intent(in) arg dragblg to compute_eddy_diff
. add actual arg dragblj to compute_vdiff call
components/cam/src/physics/cam/gw_common.F90
. remove limit_tau_without_eff and apply_tndmax. the conditionals for
these variables have been combined to be controlled by a single variable
which is now an optional argument to gw_drag_prof (lapply_effgw)
. gw_drag_prof
- add optional arg kwvrdg
. if present use it in expression for tausat
- add optional arg satfac_in
. if not present then satfac=2.0 which was previously hardcoded
- add optional arg lapply_effgw_in
. if not present the lapply_effgw=.true. This applies efficiency to
completed stress profile.
. lapply_effgw replaces apply_tndmax. it also replaces
(.not. limit_tau_without_eff)
components/cam/src/physics/cam/gw_drag.F90
. Add option for ridge scheme
. Add module data for the ridge scheme. This data is read from the topo
file for the beta ridges, and from file specified by bnd_rdggm for the
gamma ridges.
. Add namelist variables for ridge schemes. Update the namelist reader
to eliminate the SPMD ifdef.
. pass optional arg lapply_effgw_in=gw_apply_tndmax. to gw_drag_prof when
inside conditionals: use_gw_convect_dp, use_gw_convect_sh, use_gw_front,
use_gw_front_igw, use_gw_oro
components/cam/src/physics/cam/vertical_diffusion.F90
. add calls to beljaars readnl, register, init, tend
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS
hobart/nag: All PASS
hobart/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: NONE (the new schemes are not yet
turned on by default for cam5.5)
===============================================================
===============================================================
Tag name: cam5_4_63
Originator(s): bardeenc, fvitt
Date: 19 May 2016
One-line Summary: Corrections to ice nucleation processes
Purpose of changes:
These changes are needed to improve the water vapor and relative
humidity biases in the TTL, particularly when using specified
dynamics (which corrects for an offsetting cold temperature bias
in free running CAM). This affects ice clouds in the TTL and controls
stratospheric water vapor. Ice particle size and ice nucleation were
adjusted to better represent thin cirrus in the UTLS, Type II (ice)
polar stratospheric clouds, and to improve the season dehydration in
the winter polar stratosphere. The PSCs and the dehydration are important
for polar ozone chemistry. The changes will also make sure that we are
the relative humidity with respect to liquid never exceeds 1, which was
happening after CLUBB.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
- look for 'SNOWQM' constituent to add to cldice
M components/cam/src/physics/cam/macrop_driver.F90
- enforce maximum rhliq of 1 everywhere
M components/cam/src/physics/cam/micro_mg2_0.F90
- scale the water saturation values to reflect subgrid scale, where
ice clouds begin forming at a gridbox average relative humidity of
rhmini
M components/cam/src/physics/cam/microp_aero.F90
- Correct coarse dust fraction to include coarse sulfate mass. Also
support cloudborne aerosols for ice clouds.
M components/cam/src/physics/cam/nucleate_ice.F90
M components/cam/src/physics/cam/nucleate_ice_cam.F90
- Added option for ice nucleation using the gridbox average relative
humidity, and added support for cloudbourne aerosols for ice clouds.
M components/cam/src/physics/cam/clubb_intr.F90
- use ice_micro_tend to enforce RHliq<=1 everywhere if clubb_do_liqsupersat is true
- use Goff and Gratch saturation equation
- read clubb_do_liqsupersat namelist option
M components/cam/src/physics/cam/micro_mg_cam.F90
- added support for subgrid growth of ice clouds (qsatfac) and calculation
of SADICE and SADSNOW from micro_mg_tend2_0 interface
M components/cam/src/physics/cam/phys_control.F90
- removed micro_do_icesupersat
M components/cam/src/physics/cam/physpkg.F90
- removed micro_do_icesupersat code block
M components/cam/src/physics/cam/cldfrc2m.F90
- Introduce subgrid scale saturation factor (qsatfac) to allow ice growth beginning at
rhmini if cldfrc2m_do_subgrid_growth is true
M components/cam/src/physics/cam/micro_mg_utils.F90
- allow for smaller ice particle sizes and adjust fall velocity for small ice
particles
M components/cam/src/physics/cam/ndrop.F90
- support for cloudbourne aerosol in ice clouds
- return aerosols to interstitual as cloud fracions decrease
- prevent cloudbourne aerosols from being released when no liquid clould is present
M components/cam/bld/namelist_files/namelist_definition.xml
- added:
nucleate_ice_incloud
nucleate_ice_strat
cldfrc2m_do_subgrid_growth
clubb_do_liqsupersat
- removed:
micro_do_icesupersat
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
M components/cam/bld/build-namelist
- added check to error out if use_preexisting_ice is selected if the aerosol model is MAM7
- set defaults for cldfrc2m_do_subgrid_growth, nucleate_ice_incloud, nucleate_ice_strat and
clubb_do_liqsupersat
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
011 bl353 TBL.sh f1.9c5.4dh outfrq3s 9s .......................................................FAIL! rc= 7 at Wed May 18 19:40:55 MDT 2016
015 bl357 TBL.sh f1.9c5.5h outfrq3s+1850_cam55 9s .............................................FAIL! rc= 7 at Wed May 18 19:46:40 MDT 2016
018 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Wed May 18 19:59:34 MDT 2016
024 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...............................................FAIL! rc= 7 at Wed May 18 20:18:28 MDT 2016
027 bl485 TBL.sh f1.9c5.5wtmam4h outfrq3s+waccm_tsmlt_2000_cam55 9s ...........................FAIL! rc= 7 at Wed May 18 20:40:25 MDT 2016
036 bl421 TBL.sh f1.9c5mam4dh outfrq3s 9s .....................................................FAIL! rc= 7 at Wed May 18 21:03:03 MDT 2016
041 bl430 TBL.sh f1.9c5mzmam3dh outfrq3s_NEUwetdep 9s .........................................FAIL! rc= 7 at Wed May 18 21:20:30 MDT 2016
047 bl470 TBL.sh fsd1.9c5.4wtclbh outfrq3s_sd+sd_waccm5_geos5 9s ..............................FAIL! rc= 7 at Wed May 18 21:59:44 MDT 2016
050 bl482 TBL.sh f1.9c5wscmam3dh outfrq3s_2005+waccm_sc_1955-2005_cam5 9s .....................FAIL! rc= 7 at Wed May 18 22:43:52 MDT 2016
063 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ........................................FAIL! rc= 7 at Wed May 18 23:08:00 MDT 2016
068 bl740 TBL.sh h30c5h outfrq3s_lb0 9s .......................................................FAIL! rc= 7 at Wed May 18 23:12:42 MDT 2016
hobart/nag:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s ...........................................................FAIL! rc= 7 at Wed May 18 16:57:17 MDT 2016
010 bl112 TBL.sh e8c5dm ghgrmp_e8 9s ..........................................................FAIL! rc= 7 at Wed May 18 17:01:15 MDT 2016
014 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Wed May 18 17:04:18 MDT 2016
018 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Wed May 18 17:09:16 MDT 2016
031 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .........................................FAIL! rc= 7 at Wed May 18 17:20:45 MDT 2016
040 bl325 TBL.sh f10c5t5mdmsc outfrq3s_subcol 9s ..............................................FAIL! rc= 7 at Wed May 18 17:33:04 MDT 2016
046 bl338 TBL.sh f10c5dm rad_diag_mam 9s ......................................................FAIL! rc= 7 at Wed May 18 17:36:16 MDT 2016
049 bl354 TBL.sh f1.9c5.4dm outfrq3s_npryz 9s .................................................FAIL! rc= 7 at Wed May 18 17:48:30 MDT 2016
052 bl356 TBL.sh fsd1.9c5clbdm off1.9x2.5 9s ..................................................FAIL! rc= 7 at Wed May 18 18:07:14 MDT 2016
055 bl368 TBL.sh f1.9c5mg2dm outfrq3s_macmic2_classnuc 9s .....................................FAIL! rc= 7 at Wed May 18 18:17:47 MDT 2016
058 bl369 TBL.sh f1.9c5clbdm outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Wed May 18 18:30:58 MDT 2016
061 bl734 TBL.sh h16c5naqdm outfrq3s_bam+aquaplanet_cam5 9s ...................................FAIL! rc= 7 at Wed May 18 18:36:58 MDT 2016
064 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Wed May 18 18:41:50 MDT 2016
070 bl471 TBL.sh f10c5wmam3dm outfrq3s_modalstrat 9s ..........................................FAIL! rc= 7 at Wed May 18 19:11:45 MDT 2016
073 bl805 TBL.sh f4c5carmtt1dm outfrq3s_carma 9s ..............................................FAIL! rc= 7 at Wed May 18 19:27:17 MDT 2016
hobart/pgi:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s ...........................................................FAIL! rc= 7 at Wed May 18 17:00:12 MDT 2016
010 bl112 TBL.sh e8c5dm ghgrmp_e8 9s ..........................................................FAIL! rc= 7 at Wed May 18 17:06:46 MDT 2016
012 bl113 TBL.sh e8c5paqdm aqpgro+aquaplanet_cam5 3s ..........................................FAIL! rc= 7 at Wed May 18 17:11:46 MDT 2016
016 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Wed May 18 17:14:02 MDT 2016
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Wed May 18 17:25:07 MDT 2016
038 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .........................................FAIL! rc= 7 at Wed May 18 17:50:57 MDT 2016
041 bl317 TBL.sh f10c5cdm outfrq3s 9s .........................................................FAIL! rc= 7 at Wed May 18 17:58:11 MDT 2016
045 bl319 TBL.sh f10c5nonedm outfrq3s_bam 9s ..................................................FAIL! rc= 7 at Wed May 18 18:03:37 MDT 2016
048 bl320 TBL.sh f10c5nonedm rad_diag 9s ......................................................FAIL! rc= 7 at Wed May 18 18:04:54 MDT 2016
051 bl321 TBL.sh f10c5cdm atrain 9s ...........................................................FAIL! rc= 7 at Wed May 18 18:07:50 MDT 2016
054 bl334 TBL.sh f10c5udm outfrq3s 9s .........................................................FAIL! rc= 7 at Wed May 18 18:14:26 MDT 2016
057 bl354 TBL.sh f1.9c5.4dm outfrq3s_npryz 9s .................................................FAIL! rc= 7 at Wed May 18 18:32:40 MDT 2016
060 bl420 TBL.sh f4c5mam7dm outfrq3s 9s .......................................................FAIL! rc= 7 at Wed May 18 18:41:58 MDT 2016
063 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Wed May 18 18:50:17 MDT 2016
066 bl734 TBL.sh h16c5naqdm outfrq3s_bam+aquaplanet_cam5 9s ...................................FAIL! rc= 7 at Wed May 18 19:01:20 MDT 2016
All of the cam5+ baseline tests are expected to fail.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change : larger than roundoff but same climate
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
https://svn-ccsm-models.cgd.ucar.edu/cam1/branch_tags/clubbq_cam5_4_51_tags/clubbq07_cam5_4_57
- platform/compilers: yellowstone, intel
- configure commandline:
- build-namelist command (or complete namelist):
For cam5:
create_newcase -mach yellowstone -res f19_f19 -case /glade/p/acd/fvitt/cesm/cases/F2000C5_nucleate_test02 -user_compset 2000_CAM5_CLM40%SP_CICE%PRES_DOCN%DOM_RTM_SGLC_SWAV -user_pes_setby cam
For cam5.5:
create_newcase -case /glade/u/home/bardeenc/ccsm/case/f2000c55_test05f -compset FC55CLUBB -res f09_f09 -mach yellowstone
- MSS location of output:
For cam5:
/home/bardeenc/csm/f2000c5_test05f
For cam5.5:
/home/bardeenc/csm/f2000c55_test05f
MSS location of control simulations used to validate new climate:
For cam5:
/home/fvitt/csm/F2000C5_nucleate_cntl01
For cam5.5:
/home/bardeenc/csm/f2000c55_control
URL for AMWG diagnostics output used to validate new climate:
For cam5:
http://webext.cgd.ucar.edu/FCLIMO/f2000c5_test05f/atm/
For cam5.5
https://acomstaff.acom.ucar.edu/bardeenc/waccm55/f2000c55_test05f-obs/index.html
https://acomstaff.acom.ucar.edu/bardeenc/waccm55/f2000c55_test05f-f2000c55_control/index.html
===============================================================
===============================================================
Tag name: cam5_4_62
Originator(s): bogensch, cacraig
Date: May 18, 2016
One-line Summary: Introduced CLUBB liquid supersat, new CLUBB heat diffusivity
namelist param and updated CLUBB/MG namelist defaults
Purpose of changes:
- Update CLUBB svn external to head of UWM trunk.
- Introduces liquid supersaturation closure after the 2nd PDF call to
avoid relative humidities greater than 100% before CLUBB's equations
are advanced.
- Introduces tunable parameter for the diffusivities of heat in CLUBB.
- Sets the CLUBB/MG namelist default parameters to what Cecile is using
in her current pre-CAM6 "62" testing
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
- added clubb_c_K10h -- Thermo of Kh_zm
- namelist default settings for MG and CLUBB were changed to match Cecile's test "62"
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: cacraig
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M .
M components/cam
M components/cam/SVN_EXTERNAL_DIRECTORIES
- added liquid supersaturation closure after the 2nd PDF call in CLUBB
- merged to the head of the CLUBB library trunk
- updated SILHS to use the same revision number as CLUBB for consistency
M components/cam/bld/build-namelist
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
M components/cam/bld/namelist_files/namelist_definition.xml
M components/cam/src/physics/cam/clubb_intr.F90
- add clubb_c_K10h
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
Answer changes are expected for CLUBB
yellowstone/intel: all BFB except for the tests which use CLUBB
015 bl357 TBL.sh f1.9c5.5h outfrq3s+1850_cam55 9s .............................................FAIL! rc= 7 at Wed May 18 11:20:24 MDT 2016
018 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Wed May 18 11:32:45 MDT 2016
027 bl485 TBL.sh f1.9c5.5wtmam4h outfrq3s+waccm_tsmlt_2000_cam55 9s ...........................FAIL! rc= 7 at Wed May 18 12:14:11 MDT 2016
047 bl470 TBL.sh fsd1.9c5.4wtclbh outfrq3s_sd+sd_waccm5_geos5 9s ..............................FAIL! rc= 7 at Wed May 18 13:35:23 MDT 2016
hobart/nag: all BFB except for the tests which use CLUBB
052 bl356 TBL.sh fsd1.9c5clbdm off1.9x2.5 9s ..................................................FAIL! rc= 7 at Tue May 17 17:24:44 MDT 2016
058 bl369 TBL.sh f1.9c5clbdm outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Tue May 17 17:46:48 MDT 2016
hobart/pgi or jaguar/pgi: all BFB (no CLUBB tests are run using PGI)
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: All configurations which use CLUBB
- what platforms/compilers: all
- nature of change (roundoff; larger than roundoff but same climate; new
climate): new climate
This is part of the modifications which are leading up to CAM6 and has not
been tested independently, but rather as a group with the next several
upcoming tags.
===============================================================
===============================================================
Tag name: cam5_4_61
Originator(s): hannay, cacraig
Date: May 12, 2016
One-line Summary: Add namelist var zmconv_num_cin and cleanup clubb_intr
Purpose of changes:
-- Added namelist var zmconv_num_cin to control the number of negative buoyancy regions
that are allowed before the convection top and CAPE calculations are completed
-- When CAM was run on cori, it core dumped with a floating point exception in clubb.
Removed the on-the-fly arithmetic for outfld calls which operated on all columns
in the array, even undefined ones.
-- Removed unused variables in clubb_intr.F90
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
-- Added zmconv_num_cin (see description above)
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton, cacraig
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
M components/cam/bld/namelist_files/namelist_definition.xml
M components/cam/bld/build-namelist
M components/cam/src/physics/cam/zm_conv_intr.F90
M components/cam/src/physics/cam/zm_conv.F90
-- Added zmconv_num_cin to namelist
M components/cam/src/physics/cam/clubb_intr.F90
-- outfld calls had on-the-fly arithmetic being done on the
entire array, regardless of how many columns were being used.
Scaled variables before passing to outfld
-- Removed unused variables
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: BFB
hobart/nag: BFB
hobart/pgi or jaguar/pgi: BFB
===============================================================
===============================================================
Tag name: cam5_4_60
Originator(s): bdobbins, dennis, cacraig
Date: May 9, 2016
One-line Summary: Remove all references to fortran elementals for performance improvement
Purpose of changes:
- All elementals within the CAM code were removed and according to performance
testing by John Dennis' group, this speeds up CAM by 1.2%
- scalar and vector versions of these routines are provided with a single
interface via fortran overloading
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
- According to tests by John Dennis' group this simple change enables a
1.2% speedup to CAM.
Code reviewed by: cacraig, eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/src/physics/cam/clubb_intr.F90
M components/cam/src/physics/cam/eddy_diff_cam.F90
M components/cam/src/physics/cam/hb_diff.F90
M components/cam/src/physics/cam/pbl_utils.F90
M components/cam/src/physics/cam/vertical_diffusion.F90
- loops were pushed down into lower level subroutines and elemental
subroutines/functions were eliminated.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB
hobart/nag: all BFB
hobart/pgi or jaguar/pgi: all BFB
===============================================================
===============================================================
Tag name: cam5_4_59
Originator(s): ASAP, sacks, eaton
Date: Wed May 4 07:18:50 MDT 2016
One-line Summary: RRTMG performance mods and cleanup; unit test fixes
Purpose of changes:
. Incorporate performance mods (to improve vectorization) from John Dennis'
ASAP group.
. Up to now RRTMG has been included in CAM as an SVN external. Since we
will no longer receive updates to this code from AER the external is not
providing any benefit. The code has been moved to the CAM trunk and the
three subdirectories merged into one. This eliminates the confusion of
files with the same name in more than one directory (which is dealt with
in the CAM build by the Filepath ordering). At the same time all the
unused code to compute aerosol and cloud optical properties was removed.
In CAM the optical properties are computed externally to RRTMG and are
passed to it.
. Fixes to unit tests from Bill Sacks
. Fix to allow changes to namelist definition, defaults, and use case files
to be recognized when put in $CASEROOT/SourceMods/src.cam/ when running
with CESM scripts (it previously only worked with CAM standalone
scripts).
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes:
. The longwave and shortwave calculations have each been sped up by about
25% on yellowstone/intel. Still need to look at overall speedup.
Code reviewed by: cacraig, eaton
List all subroutines eliminated:
SVN external for RRTMG was eliminated:
https://svn-ccsm-models.cgd.ucar.edu/rrtmg/trunk_tags/rrtmg_160210/src
components/cam/test/system/README
. not useful
components/cam/test/system/gen_test_table.sh
. replaced by perl script in gen-test-table
List all subroutines added and what they do:
These files originated from the ASAP branch
https://svn-ccsm-models.cgd.ucar.edu/rrtmg/branch_tags/asap_rrtmg_lw_optimization_tags/asap_rrtmg_n08_format_fix/
components/cam/src/physics/rrtmg/aer_src/mcica_subcol_gen_lw.f90
components/cam/src/physics/rrtmg/aer_src/mcica_subcol_gen_sw.f90
components/cam/src/physics/rrtmg/aer_src/parrrsw.f90
components/cam/src/physics/rrtmg/aer_src/parrrtm.f90
components/cam/src/physics/rrtmg/aer_src/rrlw_con.f90
components/cam/src/physics/rrtmg/aer_src/rrlw_kg01.f90
components/cam/src/physics/rrtmg/aer_src/rrlw_kg02.f90
components/cam/src/physics/rrtmg/aer_src/rrlw_kg03.f90
components/cam/src/physics/rrtmg/aer_src/rrlw_kg04.f90
components/cam/src/physics/rrtmg/aer_src/rrlw_kg05.f90
components/cam/src/physics/rrtmg/aer_src/rrlw_kg06.f90
components/cam/src/physics/rrtmg/aer_src/rrlw_kg07.f90
components/cam/src/physics/rrtmg/aer_src/rrlw_kg08.f90
components/cam/src/physics/rrtmg/aer_src/rrlw_kg09.f90
components/cam/src/physics/rrtmg/aer_src/rrlw_kg10.f90
components/cam/src/physics/rrtmg/aer_src/rrlw_kg11.f90
components/cam/src/physics/rrtmg/aer_src/rrlw_kg12.f90
components/cam/src/physics/rrtmg/aer_src/rrlw_kg13.f90
components/cam/src/physics/rrtmg/aer_src/rrlw_kg14.f90
components/cam/src/physics/rrtmg/aer_src/rrlw_kg15.f90
components/cam/src/physics/rrtmg/aer_src/rrlw_kg16.f90
components/cam/src/physics/rrtmg/aer_src/rrlw_ref.f90
components/cam/src/physics/rrtmg/aer_src/rrlw_tbl.f90
components/cam/src/physics/rrtmg/aer_src/rrlw_wvn.f90
components/cam/src/physics/rrtmg/aer_src/rrsw_con.f90
components/cam/src/physics/rrtmg/aer_src/rrsw_kg16.f90
components/cam/src/physics/rrtmg/aer_src/rrsw_kg17.f90
components/cam/src/physics/rrtmg/aer_src/rrsw_kg18.f90
components/cam/src/physics/rrtmg/aer_src/rrsw_kg19.f90
components/cam/src/physics/rrtmg/aer_src/rrsw_kg20.f90
components/cam/src/physics/rrtmg/aer_src/rrsw_kg21.f90
components/cam/src/physics/rrtmg/aer_src/rrsw_kg22.f90
components/cam/src/physics/rrtmg/aer_src/rrsw_kg23.f90
components/cam/src/physics/rrtmg/aer_src/rrsw_kg24.f90
components/cam/src/physics/rrtmg/aer_src/rrsw_kg25.f90
components/cam/src/physics/rrtmg/aer_src/rrsw_kg26.f90
components/cam/src/physics/rrtmg/aer_src/rrsw_kg27.f90
components/cam/src/physics/rrtmg/aer_src/rrsw_kg28.f90
components/cam/src/physics/rrtmg/aer_src/rrsw_kg29.f90
components/cam/src/physics/rrtmg/aer_src/rrsw_ref.f90
components/cam/src/physics/rrtmg/aer_src/rrsw_tbl.f90
components/cam/src/physics/rrtmg/aer_src/rrsw_wvn.f90
components/cam/src/physics/rrtmg/aer_src/rrtmg_lw_init.f90
components/cam/src/physics/rrtmg/aer_src/rrtmg_lw_k_g.f90
components/cam/src/physics/rrtmg/aer_src/rrtmg_lw_rad.f90
components/cam/src/physics/rrtmg/aer_src/rrtmg_lw_rtrnmc.f90
components/cam/src/physics/rrtmg/aer_src/rrtmg_lw_setcoef.f90
components/cam/src/physics/rrtmg/aer_src/rrtmg_lw_taumol.f90
components/cam/src/physics/rrtmg/aer_src/rrtmg_sw_cldprmc.f90
components/cam/src/physics/rrtmg/aer_src/rrtmg_sw_init.f90
components/cam/src/physics/rrtmg/aer_src/rrtmg_sw_k_g.f90
components/cam/src/physics/rrtmg/aer_src/rrtmg_sw_rad.f90
components/cam/src/physics/rrtmg/aer_src/rrtmg_sw_reftra.f90
components/cam/src/physics/rrtmg/aer_src/rrtmg_sw_setcoef.f90
components/cam/src/physics/rrtmg/aer_src/rrtmg_sw_spcvmc.f90
components/cam/src/physics/rrtmg/aer_src/rrtmg_sw_taumol.f90
components/cam/src/physics/rrtmg/aer_src/rrtmg_sw_vrtqdr.f90
. Removed unused code for computing cloud and aerosol optics
components/cam/test/system/gen-test-style.css
. style sheet used by perl scripts that generate regression test web pages
List all existing files that have been modified, and describe the changes:
components/cam/SVN_EXTERNAL_DIRECTORIES
. The external for RRTMG was eliminated
components/cam/bld/configure
. Adjust Filepath for new directory containing RRTMG source
components/cam/cime_config/buildnml
. add arg to configure command to pass the cam source mods directory
location. This allows changes to namelist definition, defaults, and use
case files to be recognized when put in $CASEROOT/SourceMods/src.cam/
when running with CESM scripts (it previously only worked with CAM
standalone scripts).
components/cam/src/physics/rrtmg/radlw.F90
. rad_rrtmg_lw
- remove inflglw, iceflglw, liqflglw args used to specify options for
computing cloud optics
components/cam/src/physics/rrtmg/radsw.F90
. rad_rrtmg_sw
- remove inflgsw, iceflgsw, liqflgsw args used to specify options for
computing cloud optics
components/cam/test/unit/CMakeLists.txt
components/cam/test/unit/linear_1d_operators/CMakeLists.txt
components/cam/test/unit/micro_mg_utils/CMakeLists.txt
. unit test fixes from Bill Sacks
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS
hobart/nag: All PASS
hobart/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_4_58
Originator(s): goldy
Date: 2016-05-03
One-line Summary: Change writing of FV history coords for PIO2
Purpose of changes: Make FV history coord writing PIO2 compatible
Bugs fixed (include bugzilla ID): NA
Describe any changes made to build system: NA
Describe any changes made to the namelist: NA
List any changes to the defaults for the boundary datasets: NA
Describe any substantial timing or memory changes: NA
Code reviewed by: goldy, katec
List all subroutines eliminated: NA
List all subroutines added and what they do: NA
List all existing files that have been modified, and describe the changes:
M components/cam/doc/ChangeLog
- File was not long enough -- added more lines
M components/cam/src/dynamics/fv/dyn_grid.F90
- Modified filemap for horizontal coordinates to avoid issues with PIO2
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All pass
hobart/nag: All pass
hobart/pgi: All pass
CAM tag used for the baseline comparison tests if different than previous
tag:
===============================================================
===============================================================
Tag name: cam5_4_57
Originator(s): bogensch, raut@uwm.edu, cacraig
Date: April 15, 2016
One-line Summary: Updated CLUBB/SILHS external libraries and introduced a number of CLUBB namelist parameters
Purpose of changes:
- Eric Raut at University of Wisconsin/Madison (UWM) made modifications to
the CLUBB library to allow the current version at UWM to have BFB answers
with the CAM version of the CLUBB library. This allowed CAM to eliminate
the need to maintain a separate branch of the CLUBB library and to use
an unmodified copy from the UWM CLUBB trunk directly.
- A number of CLUBB tunable parameters were introduced into the CLUBB
namelist by Pete Bogenschutz
- Reintroduced the check in build-namelist to prohibit running with more than
one thread if clubb_history is enabled. This feature does not run
properly when more than when thread is used.
- Removed the ability to turn on clubb_do_deep as it was an unsupported
feature
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
- Removed the clubb_do_deep option from clubb_opts
Describe any changes made to the namelist:
- Introduced the following namelist parameters for CLUBB tunings
clubb_c11: Low Skewness in C11 Skw. Function
clubb_c11b: High Skewness in C11 Skw. Function
clubb_mult_coef: Coef. applied to log(avg dz/thresh)
clubb_gamma_coef: Low Skw.: gamma coef. Skw. Fnct.
clubb_c_K10: Coefficient of Kh_zm
clubb_beta: Plume widths for theta_l and rt
clubb_C2rt: C2 coef. for the rtp2_dp1 term
clubb_C2thl: C2 coef. for the thlp2_dp1 term
clubb_C2rtthl: C2 coef. for the rtpthlp_dp1 term
clubb_C8: Coef. #1 in C8 Skewness Equation
clubb_C7: Low Skewness in C7 Skw. Function
clubb_C7b: High Skewness in C7 Skw. Function
clubb_Skw_denom_coef: Factor to decrease sensitivity in the denominator of Skw calculation
clubb_lambda0_stability_coef: Intensity of stability correction applied to C1 and C6
clubb_l_lscale_plume_centered: Uses PDF to compute perturbed values for l_avg_Lscale code
clubb_l_use_ice_latent: Include the effects of ice latent heating in turbulence terms
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: cacraig, eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M .
M components/cam
M components/cam/SVN_EXTERNAL_DIRECTORIES
- updated the CLUBB and SILHS externals to use an unmodified
copy of a recent version from UWM
M components/cam/bld/build-namelist
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
M components/cam/bld/namelist_files/namelist_definition.xml
M components/cam/src/physics/cam/clubb_intr.F90
- added a number of CLUBB tuning parameters
- removed the obsolete clubb_do_deep
M components/cam/bld/configure
- removed clubb_do_deep
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB
hobart/nag: all BFB
hobart/pgi or jaguar/pgi: all BFB
** An additional test was run by Cecile with her current CESM setup and she indicated
that the answers were identical using this version.
===============================================================
===============================================================
Tag name: cam5_4_56
Originator(s): raghuraj@ucar.edu, chris.kerr.llc@gmail.com, cacraig
Date: April 6, 2016
One-line Summary: Optimized binterp in modal_aer_opt.F90
Purpose of changes:
- Optimization changes for binterp inside modal_aer_opt.F90
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
- According to Chris Kerr, binterp running using the yellowstone intel 16 compiler has a 30% speedup with this version.
Code reviewed by: cacraig, eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/src/physics/cam/modal_aer_opt.F90
- optimizations to binterp
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB (bl357 was tested by hand as a baseline did not exist for it in cam5_4_54 -- it was BFB)
hobart/nag: all BFB
hobart/pgi or jaguar/pgi: all BFB
CAM tag used for the baseline comparison tests if different than previous
tag: cam5_4_54 (cam5_4_55 did not have regression tests run for it)
===============================================================
===============================================================
Tag name: cam5_4_55
Originator(s): fvitt
Date: 5 Apr 2016
One-line Summary: Fix pre-industrial CAM5.5 build-namelist use case
Purpose of changes:
. Adjust the 1850_cam55 use case for proper inclusion of H20 forcings
due to oxidation of CH4
. Add regression test for 1850_cam55 use case
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
A + components/cam/test/system/config_files/f1.9c5.5h
List all existing files that have been modified, and describe the changes:
M components/cam/bld/namelist_files/use_cases/1850_cam55.xml
- made adjustments for inclusion of H20 forcings due to oxidation of CH4
M components/cam/test/system/input_tests_master
M components/cam/test/system/tests_pretag_yellowstone
- added test for the 1850_cam55 use case
The standard regression tests were not run since there are no code changes
and the only change is to the 1850_cam55.xml use case file -- only tested
the 1850_cam55 use case -- sm, er, br tests passed.
===============================================================
===============================================================
Tag name: cam5_4_54
Originator(s): chris.kerr.llc@gmail.com, cacraig
Date: March 31, 2016
One-line Summary: optimized MG2
Purpose of changes:
- Improve performance of microphysics routines. A 7-8% improvement
in microphysics is seen with this optimization.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: cacraig, eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/src/physics/cam/micro_mg1_0.F90
M components/cam/src/physics/cam/micro_mg1_5.F90
M components/cam/src/physics/cam/micro_mg2_0.F90
- all arrays are automatic arrays based on the size that micro_mg_get_cols returns
M components/cam/src/physics/cam/micro_mg_cam.F90
- all allocations are removed and arrays are now automatic arrays
- micro_mg_cam_tend simply calls micro_mg_get_cols and micro_mg_cam_tend_pack
- micro_mg_cam_tend_pack is basically the old micro_mg_cam_tend without
the calls to micro_mg_get_cols.
M components/cam/src/physics/cam/micro_mg_utils.F90
- moved constant code outside of the loops over columns
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: BFB
hobart/nag: BFB
hobart/pgi or jaguar/pgi: BFB
===============================================================
===============================================================
Tag name: cam5_4_53
Originator(s): fvitt, mmills
Date: 26 Mar 2016
One-line Summary: CAM5.5 stratospheric aerosols and methane oxidation
Purpose of changes:
. Changes to default namelist settings to turn on prescribed stratospheric
aerosols and prescribed forcing of H2O due to oxidation of methane in CAM5.5
. Update prescribed stratospheric aerosol input dataset is included for CAM5.5
. Turn on prognostic stratospheric aerosols in configurations
that have the appropriated chemistry
. Update volcanic sulfur emissions input dataset for transient runs with prognostic
stratospheric aerosols
. Misc clean up of default namelist settings for CAM5/CAM5.5
. Add CAM5.5 compsets:
FW2000 - WACCM with CAM5.5 physics and TSMLT chemistry
FWAMIP - WACCM5.5 transient with volcanic SO2 emissions
FSV2000 - CAMChem with CAM5.5 stratospheric chemistry and VBS SOAs
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton, cacraig
List all subroutines eliminated:
D components/cam/test/system/config_files/f10c5wtmam3dm
D components/cam/test/system/config_files/fsd1.9c5wtmam3h
- these are replaced by waccm_tsmlt_mam4 versions
D components/cam/src/chemistry/pp_waccm_tsmlt_mam3
D components/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_lu_solve.F90
D components/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_setrxt.F90
D components/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_adjrxt.F90
D components/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_rxt_rates_conv.F90
D components/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_imp_sol.F90
D components/cam/src/chemistry/pp_waccm_tsmlt_mam3/chem_mods.F90
D components/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_prod_loss.F90
D components/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_lin_matrix.F90
D components/cam/src/chemistry/pp_waccm_tsmlt_mam3/m_rxt_id.F90
D components/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_phtadj.F90
D components/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_nln_matrix.F90
D components/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_lu_factor.F90
D components/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_sim_dat.F90
D components/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_indprd.F90
D components/cam/src/chemistry/pp_waccm_tsmlt_mam3/m_spc_id.F90
D components/cam/src/chemistry/pp_waccm_tsmlt_mam3/chem_mech.doc
D components/cam/src/chemistry/pp_waccm_tsmlt_mam3/chem_mech.in
- removed waccm_tsmlt_mam3 chemistry package
List all subroutines added and what they do:
A components/cam/bld/namelist_files/use_cases/waccm_tsmlt_amip_cam55.xml
A components/cam/bld/namelist_files/use_cases/waccm_tsmlt_1850_cam55.xml
A components/cam/bld/namelist_files/use_cases/waccm_tsmlt_2000_cam55.xml
- use cases added for WACCM5.5 compsets
A components/cam/test/system/config_files/f1.9c5.5wtmam4h
A components/cam/test/system/config_files/f10c5wtmam4dm
- config file for waccm_tsmlt_mam3 regression tests
List all existing files that have been modified, and describe the changes:
M components/cam/src/chemistry/mozart/mo_sulf.F90
- correction to checks for the need for prescribed tropospheric sulfates
M components/cam/bld/configure
M components/cam/bld/config_files/definition.xml
- removed waccm_tsmlt_mam3 chem option
M components/cam/bld/perl5lib/Build/ChemNamelist.pm
- added utility to check for presence of named species in the chemistry mechanism
M components/cam/bld/build-namelist
- turn on stratopheric aerosol forcings for cam5.5/cam6
-- activate prognostic stratopheric aerosols in appropriate
configurations, otherwise use prescribed stratospheric aerosols
- turn on methane oxidation in cam5.5/cam6
- corrections to default emissions for tsmlt and mam4 chemistries
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
- updates to namelist defaults for WACCM and CAMChem with cam5.5/cam6
-- SAD updated for CAM5.5 and CAM6
M components/cam/bld/namelist_files/namelist_definition.xml
- "strat_aero_feedback" renamed "prescribed_strataero_feedback"
M components/cam/bld/namelist_files/use_cases/1950-2010_ccmi_refc1_waccm_ma.xml
M components/cam/bld/namelist_files/use_cases/sd_waccm_geos5.xml
M components/cam/bld/namelist_files/use_cases/2000_cam5_trop_strat_mam3.xml
M components/cam/bld/namelist_files/use_cases/waccmx_2000_cam4.xml
M components/cam/bld/namelist_files/use_cases/2000_cam5_trop_strat_mam7.xml
M components/cam/bld/namelist_files/use_cases/waccm_2000_cam4.xml
M components/cam/bld/namelist_files/use_cases/waccm_2000_cam5.xml
M components/cam/bld/namelist_files/use_cases/sd_1975-2010_ccmi_refc1_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/1950-2010_ccmi_refc1_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/1950-2010_ccmi_refc1_waccm_tsmlt.xml
M components/cam/bld/namelist_files/use_cases/waccm_sc_1955-2005_cam5.xml
M components/cam/bld/namelist_files/use_cases/waccm_1850-2005_cam4.xml
M components/cam/bld/namelist_files/use_cases/waccmxi_2000_cam4.xml
M components/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp26.xml
M components/cam/bld/namelist_files/use_cases/sd_waccm5_geos5.xml
M components/cam/bld/namelist_files/use_cases/sd_1975-2010_ccmi_refc1_waccm_ma.xml
M components/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_waccm_ma.xml
M components/cam/bld/namelist_files/use_cases/sd_1975-2010_ccmi_refc1_waccm_tsmlt.xml
M components/cam/bld/namelist_files/use_cases/2000_cam4_trop_chem.xml
M components/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_waccm_tsmlt.xml
M components/cam/bld/namelist_files/use_cases/sd_cam4_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/1850_waccm_tsmlt_cam5.xml
M components/cam/bld/namelist_files/use_cases/2000_cam4_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/2007-2011_htap_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/waccm_1850_cam4.xml
M components/cam/bld/namelist_files/use_cases/waccm_1850_cam5.xml
M components/cam/bld/namelist_files/use_cases/sd_waccm_sulfur.xml
M components/cam/bld/namelist_files/use_cases/2004-2100_ccmi_refc2_rcp45_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/2000_ccmi_refc1_waccm_tsmlt.xml
M components/cam/bld/namelist_files/use_cases/sd_2007-2011_htap_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp45.xml
M components/cam/bld/namelist_files/use_cases/cam4_chem_radpsv_geos5.xml
M components/cam/bld/namelist_files/use_cases/cam4_trop_strat_chem.xml
M components/cam/bld/namelist_files/use_cases/1979-2010_sd_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_rcp85_waccm_ma.xml
M components/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp85.xml
M components/cam/bld/namelist_files/use_cases/waccm_1955-2005_cam4.xml
M components/cam/bld/namelist_files/use_cases/waccm_1974-2005_cam4.xml
M components/cam/bld/namelist_files/use_cases/waccm_1955-2005_cam5.xml
M components/cam/bld/namelist_files/use_cases/waccm_refb1.xml
M components/cam/bld/namelist_files/use_cases/waccm_carma_bc_2013_cam4.xml
M components/cam/bld/namelist_files/use_cases/2005-2100_cam4_strataero_rcp45.xml
M components/cam/bld/namelist_files/use_cases/waccmx_1996_cam4.xml
M components/cam/bld/namelist_files/use_cases/rcp6.0_waccm_tsmlt_nomegan_cam5.xml
- "strat_aero_feedback" renamed "prescribed_strataero_feedback"
M components/cam/bld/namelist_files/use_cases/volc_waccm_ma_cam5.xml
M components/cam/bld/namelist_files/use_cases/volc_waccm_tsmlt_nomegan_cam5.xml
M components/cam/bld/namelist_files/use_cases/volc_waccm_tsmlt_megan_cam5.xml
- update the volcanic so2 emissions
M components/cam/cime_config/config_compsets.xml
M components/cam/cime_config/testdefs/testlist_cam.xml
M components/cam/cime_config/config_component.xml
- new compsets added:
FW2000 -- WACCM
FWAMIP
FSV2000 -- CAMChem with VBS SOAs
M components/cam/test/system/tests_waccm_hybrid
M components/cam/test/system/config_files/f1.9c5.4wtclbh
M components/cam/test/system/config_files/f4c5.4wtclbdm
M components/cam/test/system/config_files/f1.9c5.4wtclbdh
M components/cam/test/system/config_files/fsd1.9c5.4wtclbh
M components/cam/test/system/tests_pretag_yellowstone
M components/cam/test/system/input_tests_master
- changes to test waccm_tsmlt_mam4 with cam5.5 phys
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
td.165357.status:023 bl485 TBL.sh f1.9c5.5wtmam4h outfrq3s+waccm_tsmlt_2000_cam55 9s ...........................FAIL! rc= 7 at Fri Mar 25 12:28:59 MDT 2016
- new test -- no baseline to compare against
td.165357.status:037 bl430 TBL.sh f1.9c5mzmam3dh outfrq3s_NEUwetdep 9s .........................................FAIL! rc= 7 at Fri Mar 25 13:08:55 MDT 2016
td.165357.status:040 bl440 TBL.sh f1.9c4soadh outfrq3s+soa_chem_megan_emis 9s ..................................FAIL! rc= 7 at Fri Mar 25 13:27:28 MDT 2016
- expected failures due to changes to default LBCs
td.165357.status:043 bl470 TBL.sh fsd1.9c5.4wtclbh outfrq3s_sd+sd_waccm5_geos5 9s ..............................FAIL! rc= 7 at Fri Mar 25 13:49:22 MDT 2016
- expected failure due to changed chem package "waccm_tsmlt_mam3" to "waccm_tsmlt_mam4"
hobart/nag:
067 bl426 TBL.sh f10c4wmxdm outfrq3s_ionos 9s .................................................FAIL! rc= 7 at Fri Mar 25 09:59:00 MDT 2016
- expected failure due to changes in default LBC file and SAD year
076 bl011 TBL.sh f4c5portdm outfrq24h_port 2d .................................................FAIL! rc= 7 at Fri Mar 25 10:31:13 MDT 2016
- actually bit-for-bit -- change in the LBC file changes the green house gas values recorded in history
hobart/pgi:
068 bl011 TBL.sh f4c5portdm outfrq24h_port 2d .................................................FAIL! rc= 7 at Fri Mar 25 10:08:52 MDT 2016
- actually bit-for-bit -- change in the LBC file changes the green house gas values recorded in history
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_4_52
Originator(s): bogensch, cacraig
Date: March 21, 2016
One-line Summary: Added alternative autoconversion scheme
Purpose of changes:
- New switch to use alternative autoconversion scheme in MG2 (following Seifert and
Behang 2001). Currently this is not turned on
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
- added logical micro_do_sb_physics which turns on/off the Seifert and Behang autoconversion.
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/bld/namelist_files/namelist_definition.xml
- added namelist parameter to turn on/off using the Seifert and Behang
autoconversion and accretion physics. It currently defaults to false.
M components/cam/src/physics/cam/micro_mg2_0.F90
M components/cam/src/physics/cam/micro_mg_cam.F90
M components/cam/src/physics/cam/micro_mg_utils.F90
- changes to add the Seifert and Behang autoconversion
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB
hobart/nag: all BFB
hobart/pgi or jaguar/pgi: all BFB
===============================================================
===============================================================
Tag name: cam5_4_51
Originator(s): fvitt, tilmes, mahowald@cornell.edu
Date: 24 Feb 2016
One-line Summary: Adjustments to bulk dust model
Purpose of changes:
To update the following for the bulk aerosol model:
. dust physics properties files (optics files)
. dust emissions scaling factors
. soil erosion input file
. dust scavenging coefficients
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/bld/build-namelist
- changes for setting aer_scav_coef (scavenging coefficients) for bulk aerosols
via ChemNamelist->set_dep_lists
- set default dust_emis_fact when soil_erod_file is in the namelist rather than
using chemistry package name
M components/cam/src/chemistry/bulk_aero/aero_model.F90
- bulk model scavenging factor is now set via namelist
M components/cam/src/chemistry/bulk_aero/dust_model.F90
- changes in dust emissions scaling factors
- removal of the threshold where soil_erod is set to zero
M components/cam/src/chemistry/aerosol/wetdep.F90
- minor white space change
M components/cam/bld/perl5lib/Build/ChemNamelist.pm
- changes for setting scavenging coefficients of bulk aerosols
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
- changes in default optics files (phys props) for bulk dust
- new default soil erosion file for bulk dust
- dust_emis_fact set to 0.37 for cam4
M components/cam/bld/namelist_files/namelist_definition.xml
- added namelist variables for scavenging coefficients of bulk aerosols
M components/cam/bld/namelist_files/use_cases/cam4_chem_radpsv_geos5.xml
M components/cam/bld/namelist_files/use_cases/cam4_bam_radpsv_geos5.xml
- removed dust_emis_fact -- let build-namelist/namelist_defaults_cam set this
- removed soil_erod file
M components/cam/bld/namelist_files/use_cases/2007-2011_htap_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/2000_cam4_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/2004-2100_ccmi_refc2_rcp45_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/2000_ccmi_refc1_waccm_tsmlt.xml
M components/cam/bld/namelist_files/use_cases/sd_2007-2011_htap_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/1979-2010_sd_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/sd_cam5_trop_strat_mam3.xml
M components/cam/bld/namelist_files/use_cases/sd_1975-2010_ccmi_refc1_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/1950-2010_ccmi_refc1_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/1950-2010_ccmi_refc1_waccm_tsmlt.xml
M components/cam/bld/namelist_files/use_cases/sd_1975-2010_ccmi_refc1_waccm_ma.xml
M components/cam/bld/namelist_files/use_cases/sd_1975-2010_ccmi_refc1_waccm_tsmlt.xml
M components/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_waccm_tsmlt.xml
M components/cam/bld/namelist_files/use_cases/sd_cam4_trop_strat_soa.xml
- removed dust_emis_fact -- let build-namelist/namelist_defaults_cam set this
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
004 bl132 TBL.sh e48c4dh ghgrmp_e8+1850_cam4 9s ...............................................FAIL! rc= 7 at Tue Feb 23 16:40:21 MST 2016
008 bl331 TBL.sh f4c4dh co2rmp+1850_cam4 9s ...................................................FAIL! rc= 7 at Tue Feb 23 16:40:25 MST 2016
021 bl373 TBL.sh f1.9c4mozdh outfrq3s+2000_cam4_trop_chem 9s ..................................FAIL! rc= 7 at Tue Feb 23 16:40:49 MST 2016
025 bl376 TBL.sh f1.9c4mozdh+testmech outfrq3s 9s .............................................FAIL! rc= 7 at Tue Feb 23 16:41:01 MST 2016
028 bl374 TBL.sh f1.9c4fastdh outfrq3s+1850-2005_cam4_super_fast_llnl 9s ......................FAIL! rc= 7 at Tue Feb 23 16:41:11 MST 2016
031 bl390 TBL.sh f1.9c4wmxh outfrq3s+waccmx_2000_cam4 9s ......................................FAIL! rc= 7 at Tue Feb 23 16:42:59 MST 2016
034 bl397 TBL.sh f1.9c4cm outfrq3s+1850-2005_cam4 9s ..........................................FAIL! rc= 7 at Tue Feb 23 16:43:09 MST 2016
040 bl429 TBL.sh f1.9c4wmxh outfrq3s+waccmxi_2000_cam4 9s .....................................FAIL! rc= 7 at Tue Feb 23 16:44:59 MST 2016
045 bl440 TBL.sh f1.9c4soadh outfrq3s+soa_chem_megan_emis 9s ..................................FAIL! rc= 7 at Tue Feb 23 16:45:33 MST 2016
055 bl387 TBL.sh f1.9c4dm outfrq3s+1850-2005_cam4 9s ..........................................FAIL! rc= 7 at Tue Feb 23 17:50:10 MST 2016
058 bl388 TBL.sh f1.9c4m outfrq24h+1850_cam4 2d ...............................................FAIL! rc= 7 at Tue Feb 23 17:50:13 MST 2016
061 bl379 TBL.sh f1.9c4wmh outfrq24h+waccm_2005-2100_cam4_rcp45 2d ............................FAIL! rc= 7 at Tue Feb 23 17:51:15 MST 2016
hobart/nag:
028 bl314 TBL.sh f10c4wscdm outfrq3s 9s .......................................................FAIL! rc= 7 at Tue Feb 23 10:14:39 MST 2016
034 bl318 TBL.sh f10c4cdm sat_hist 9s .........................................................FAIL! rc= 7 at Tue Feb 23 10:22:05 MST 2016
067 bl426 TBL.sh f10c4wmxdm outfrq3s_ionos 9s .................................................FAIL! rc= 7 at Tue Feb 23 11:47:34 MST 2016
hobart/pgi:
033 bl314 TBL.sh f10c4wscdm outfrq3s 9s .......................................................FAIL! rc= 7 at Tue Feb 23 10:39:48 MST 2016
These cam4 baseline test failures are expected. Configurations that use bulk dust (prognostic or prescribed)
change answers due to changes in:
* dust phys props files (optics files)
* dust emissions scaling factors
* soil erosion input file
* dust scavenging coefficients
CAM tag used for the baseline comparison tests if different than previous
tag: cam5_4_49
Summarize any changes to answers, i.e.,
- what code configurations: cam4 physics
- what platforms/compilers: all
- nature of change : larger than roundoff
Validation of these changes to cam4 dust was via the CCMI simulations.
===============================================================
===============================================================
Tag name: cam5_4_50
Originator(s): cacraig
Date: February 22, 2016
One-line Summary: Update CAM testlists for CAM and CESM testing
Purpose of changes:
- prune both CAM and CESM testlists
- added CESM testing of CLUBB using CAM5.5 and CLM45
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton, goldy, fvitt, joemci
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/cime_config/config_compsets.xml
- added FC55CLUBB to run CLUBB using cam5.5 and CLM45
M components/cam/cime_config/testdefs/testlist_cam.xml
- pruned and modified the CESM testlist for prealpha and prebeta CAM tests
M components/cam/test/system/tests_pretag_yellowstone
- removed a number of unnecessary CAM regression tests
Since there were no code changes, and no new CAM regression tests were introduced, the CAM
regression testing was not run for this tag.
All variations of yellowstone and hobart compilers and prealpha/prebeta were run through the CESM script
create_test utilizing nobuild and autosubmit off to verify the tests were set up properly.
===============================================================
===============================================================
Tag name: cam5_4_49
Originator(s): cacraig
Date: February 16, 2016
One-line Summary: Revert iulog assignment back to what it was in cam5_4_42
Purpose of changes:
- A request had been made to have QNEG3 lines not be directed to unit 6 (which is used by the CESM log). An mpi_bcast
was introduced in cam5_4_43 to broadcast the CAM iulog to all tasks, but that had issues with the gnu compiler during
FLUSH. After extensive investigation, it was determined that master should really be the only task which routinely
writes to iulog. This tag reverst the change made in cam5_4_43 and a future tage will create statistics for QNEG3
that masterproc can write out.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/src/cpl/atm_comp_mct.F90
- revert iulog logic back to all tasks other than masterproc will write to unit 6
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB
hobart/nag: all BFB
hobart/pgi or jaguar/pgi: all BFB
===============================================================
===============================================================
Tag name: cam5_4_48
Originator(s): eaton
Date: Thu Feb 11 07:35:52 MST 2016
One-line Summary: implement vectorized RNG utils in RRTMG
Purpose of changes:
. Vectorized RNG utilities were implemented in the RRTMG code (from ASAP).
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes:
. ASAP reports about 1% speedup in overall throughput with cam5 physics
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
components/cam/SVN_EXTERNAL_DIRECTORIES:
. rrtmg_150428 -> rrtmg_160210
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS
hobart/nag: All PASS
hobart/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_4_47
Originator(s): fvitt
Date: 9 Feb 2016
One-line Summary: WACCM updates and chemistry preprocessor bug fix
Purpose of changes:
- include O(3P) cooling in the thermosphere
- provide capability to input solar EUV irradiance data
- use electron and ion temperatures in usrrxt if available
- fix for chemistry preprocessor "He molecular weight" bug
- misc code clean up
Bugs fixed (include bugzilla ID):
The chemistry preprocessor was computing the molecular weight of helium ('He')
as the molecular weight of hydrogen ('H') plus an electron ('e'). The solution
here is to designate the name of electron element as an uppercase 'E' rather
than lowercase 'e' in the preprocessor. The names of all other elements begin
with an uppercase character.
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
A components/cam/src/chemistry/utils/solar_euv_data.F90
- reads solar EUV irradiance data
A components/cam/test/system/nl_files/outfrq3s_euv
- use solar EUV irradance data
A components/cam/test/system/config_files/f10c4wmdm
List all existing files that have been modified, and describe the changes:
M components/cam/src/chemistry/utils/solar_data.F90
- deallocate arrays in init routine
- remove unused decared variable
M components/cam/src/chemistry/mozart/mo_jeuv.F90
- master proc reads ascii input and then broadcasts
- use solar EUV irradiance data if available otherwise use EUVAC
- code clean up
M components/cam/src/chemistry/mozart/mo_photo.F90
- include option to use solar EUV irradiance data
- code clean up
M components/cam/src/chemistry/mozart/mo_chemini.F90
M components/cam/src/chemistry/mozart/chemistry.F90
M components/cam/src/chemistry/mozart/euvac.F90
- code clean up -- removal of dead code
M components/cam/src/chemistry/mozart/mo_waccm_hrates.F90
M components/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
- pass electron- and ion-temperature to usrrxt if available
M components/cam/src/control/runtime_opts.F90
- invoke solar_euv_data_readnl
- remove unused "endrun" reference
M components/cam/src/physics/waccm/nlte_fomichev.F90
M components/cam/src/physics/waccm/nlte_lw.F90
- include O(3P) cooling
M components/cam/src/chemistry/pp_waccm_mozart/chem_mech.doc
M components/cam/src/chemistry/pp_waccm_mozart/chem_mech.in
M components/cam/src/chemistry/pp_waccm_mozart_sulfur/chem_mech.doc
M components/cam/src/chemistry/pp_waccm_mozart_sulfur/chem_mech.in
M components/cam/src/chemistry/pp_waccm_tsmlt_mam3/chem_mech.doc
M components/cam/src/chemistry/pp_waccm_tsmlt_mam3/chem_mech.in
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/chem_mech.doc
M components/cam/src/chemistry/pp_waccm_tsmlt_mam4/chem_mech.in
M components/cam/src/chemistry/pp_waccm_mozart_mam3/chem_mech.doc
M components/cam/src/chemistry/pp_waccm_mozart_mam3/chem_mech.in
M components/cam/src/chemistry/pp_waccm_mozart_mam4/chem_mech.doc
M components/cam/src/chemistry/pp_waccm_mozart_mam4/chem_mech.in
M components/cam/src/chemistry/pp_waccm_tsmlt/chem_mech.in
M components/cam/src/chemistry/pp_waccm_tsmlt/chem_mech.doc
M components/cam/src/chemistry/pp_waccm_tsmlt_sulfur/chem_mech.doc
M components/cam/src/chemistry/pp_waccm_tsmlt_sulfur/chem_mech.in
M components/cam/src/chemistry/pp_waccmx_mozart/chem_mech.doc
M components/cam/src/chemistry/pp_waccmx_mozart/chem_mech.in
- chemistry prepocessor fix to correct "He" molecular weight
-- electron is now 'E' in elements table in chem preprocessor
-- map 'e' to 'E' in chem_mech.in, i.e., 'e -> E'
M components/cam/bld/build-namelist
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
M components/cam/bld/namelist_files/namelist_definition.xml
- new "solar_euv_data_file" namelist var
- remove euvacdat_file -- data read in was not used
M components/cam/bld/namelist_files/use_cases/waccm_1850_cam4.xml
M components/cam/bld/namelist_files/use_cases/waccm_1850_cam5.xml
M components/cam/bld/namelist_files/use_cases/sd_waccm_sulfur.xml
M components/cam/bld/namelist_files/use_cases/waccm_sc_2005-2100_cam4_rcp45.xml
M components/cam/bld/namelist_files/use_cases/waccm_sc_2005-2100_cam4_rcp85.xml
M components/cam/bld/namelist_files/use_cases/waccm_sc_1850_cam4.xml
M components/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp45.xml
M components/cam/bld/namelist_files/use_cases/waccm_1955-2005_cam4.xml
M components/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp85.xml
M components/cam/bld/namelist_files/use_cases/waccm_1955-2005_cam5.xml
M components/cam/bld/namelist_files/use_cases/waccm_1974-2005_cam4.xml
M components/cam/bld/namelist_files/use_cases/volc_waccm_tsmlt_nomegan_cam5.xml
M components/cam/bld/namelist_files/use_cases/waccm_refb1.xml
M components/cam/bld/namelist_files/use_cases/waccm_carma_bc_2013_cam4.xml
M components/cam/bld/namelist_files/use_cases/waccmx_1996_cam4.xml
M components/cam/bld/namelist_files/use_cases/sd_waccm_geos5.xml
M components/cam/bld/namelist_files/use_cases/waccmx_2000_cam4.xml
M components/cam/bld/namelist_files/use_cases/waccm_2000_cam4.xml
M components/cam/bld/namelist_files/use_cases/waccm_2000_cam5.xml
M components/cam/bld/namelist_files/use_cases/waccm_sc_2000_cam4.xml
M components/cam/bld/namelist_files/use_cases/waccm_sc_2000_cam5.xml
M components/cam/bld/namelist_files/use_cases/waccm_sc_2005-2100_cam4_rcp26.xml
M components/cam/bld/namelist_files/use_cases/waccm_sc_1955-2005_cam4.xml
M components/cam/bld/namelist_files/use_cases/waccm_sc_1955-2005_cam5.xml
M components/cam/bld/namelist_files/use_cases/waccm_1850-2005_cam4.xml
M components/cam/bld/namelist_files/use_cases/waccmxi_2000_cam4.xml
M components/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp26.xml
M components/cam/bld/namelist_files/use_cases/sd_waccm5_geos5.xml
M components/cam/bld/namelist_files/use_cases/volc_waccm_ma_cam5.xml
M components/cam/bld/namelist_files/use_cases/1850_waccm_tsmlt_cam5.xml
- removed comments that no longer apply
M components/cam/test/system/nl_files/outfrq3s_neu
- removed nsplit specification -- should be set by namelist_defaults
M components/cam/test/system/nl_files/outfrq3s_ionos
- use solar EUV irradance data
- change start date 20031028
M components/cam/test/system/input_tests_master
M components/cam/test/system/tests_waccm_hybrid
M components/cam/test/system/tests_waccm_mpi
- added waccm tests that use solar EUV irradance data
M components/cam/SVN_EXTERNAL_DIRECTORIES
M components/cam
- new chem preprocessor that fixes "He" molecular weight problem
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
009 bl203 TBL.sh fsd1.9c4wtsmlth outfrq3s_sd+sd_1975-2010_ccmi_refc1_waccm_tsmlt 9s ...........FAIL! rc= 7 at Mon Feb 8 15:14:30 MST 2016
016 bl326 TBL.sh f1.9c4wscdm outfrq3s_2005+waccm_sc_2005-2100_cam4_rcp45 9s ...................FAIL! rc= 7 at Mon Feb 8 15:36:26 MST 2016
029 bl336 TBL.sh f4c4wmdh outfrq3s+waccm_1850_cam4 9s .........................................FAIL! rc= 7 at Mon Feb 8 16:08:57 MST 2016
035 bl359 TBL.sh f4c5wmclbdh outfrq3s_clubb+waccm_2000_cam5 9s ................................FAIL! rc= 7 at Mon Feb 8 16:54:35 MST 2016
052 bl390 TBL.sh f1.9c4wmxh outfrq3s+waccmx_2000_cam4 9s ......................................FAIL! rc= 7 at Mon Feb 8 17:49:25 MST 2016
061 bl429 TBL.sh f1.9c4wmxh outfrq3s+waccmxi_2000_cam4 9s .....................................FAIL! rc= 7 at Mon Feb 8 18:12:08 MST 2016
069 bl470 TBL.sh fsd1.9c5.4wtclbh outfrq3s_sd+sd_waccm5_geos5 9s ..............................FAIL! rc= 7 at Mon Feb 8 19:00:57 MST 2016
072 bl482 TBL.sh f1.9c5wscmam3dh outfrq3s_2005+waccm_sc_1955-2005_cam5 9s .....................FAIL! rc= 7 at Mon Feb 8 19:47:37 MST 2016
082 bl379 TBL.sh f1.9c4wmh outfrq24h+waccm_2005-2100_cam4_rcp45 2d ............................FAIL! rc= 7 at Mon Feb 8 20:14:41 MST 2016
hobart/nag:
028 bl314 TBL.sh f10c4wscdm outfrq3s 9s .......................................................FAIL! rc= 7 at Mon Feb 8 11:59:16 MST 2016
067 bl426 TBL.sh f10c4wmxdm outfrq3s_ionos 9s .................................................FAIL! rc= 7 at Mon Feb 8 13:36:48 MST 2016
070 bl471 TBL.sh f10c5wmam3dm outfrq3s_modalstrat 9s ..........................................FAIL! rc= 7 at Mon Feb 8 13:52:21 MST 2016
jaguar/pgi:
033 bl314 TBL.sh f10c4wscdm outfrq3s 9s .......................................................FAIL! rc= 7 at Mon Feb 8 12:27:11 MST 2016
The inclusion of O(3P) cooling has caused all WACCM configurations to change answers.
In addition to (3P) cooling the use of ion and electron temperatures to set reaction
rates has cause WACCMX with ionosphere to change answers.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_4_46
Originator(s): bogensch, eaton
Date: Mon Feb 8 07:33:58 MST 2016
One-line Summary: move reading of dycore namelists; clubb mod.
Purpose of changes:
. Move reading of dycore namelists to subroutine read_namelist (from
dyn_init)
. CLUBB mod to provide horizontal cell size estimates
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist:
. create eul/sld specific versions of all the generic dyn_spectral_*
variables, i.e.,
dyn_spectral_dif2 -> eul_dif2_coef and sld_dif2_coef
dyn_spectral_dif4 -> sld_dif4_coef
dyn_spectral_divdampn -> eul_divdampn and sld_divdampn
dyn_spectral_eps -> eul_tfilt_eps and sld_tfilt_eps
dyn_spectral_kmxhdc -> eul_kmxhdc and sld_kmxhdc
. move all eulerian specific vars to namelist group dyn_eul_inparm
. move all sld specific vars to namelist group dyn_sld_inparm
. create fv specific variable names:
nsplit -> fv_nsplit
nspltrac -> fv_nspltrac
nspltvrm -> fv_nspltvrm
iord -> fv_iord
jord -> fv_jord
kord -> fv_kord
dyn_conservative -> fv_conserve
filtcw -> fv_filtcw
ct_overlap -> fv_ct_overlap
trac_decomp -> fv_trac_decomp
fft_flt -> fv_fft_flt
div24del2flag -> fv_div24del2flag
del2coef -> fv_del2coef
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton, goldy
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
components/cam/bld/build-namelist
components/cam/bld/namelist_files/namelist_defaults_cam.xml
. update variable names for Eul, FV, and SLD dycores
components/cam/bld/namelist_files/namelist_definition.xml
. update variable names for Eul, FV, and SLD dycores
. replace generic dyn_spectral_inparm group by dycore specific versions
dyn_eul_inparm and dyn_sld_inparm.
components/cam/bld/namelist_files/use_cases/1850_waccm_tsmlt_cam5.xml
components/cam/bld/namelist_files/use_cases/aquaplanet_cam3.xml
components/cam/bld/namelist_files/use_cases/dabi_p2004.xml
components/cam/bld/namelist_files/use_cases/waccm_sc_1955-2005_cam5.xml
components/cam/bld/namelist_files/use_cases/waccmx_1996_cam4.xml
components/cam/bld/namelist_files/use_cases/waccmx_2000_cam4.xml
components/cam/bld/namelist_files/use_cases/waccmxi_2000_cam4.xml
. update variable names for Eul and FV dycores
components/cam/src/control/cam_comp.F90
. remove namelist file from arg lists of cam_initial and cam_read_restart
components/cam/src/control/cam_restart.F90
. remove namelist file from arg lists of cam_read_restart and
read_restart_dynamics
components/cam/src/control/runtime_opts.F90
. add call to dyn_readnl
components/cam/src/cpl/atm_comp_esmf.F90
components/cam/src/cpl/atm_comp_mct.F90
. fix comment
components/cam/src/dynamics/eul/dyn_comp.F90
. add subroutine dyn_readnl (copied from dyn_eul_readnl)
- update the mpibcast calls to use mpi_bcast
- change spectral namelist variables to eul specific ones, and change the
namelist group from dyn_spectral_inparm to dyn_eul_inparm
. move spmd_readnl call to dyn_readnl
. remove dyn_readnl call from dyn_init.
components/cam/src/dynamics/eul/eul_control_mod.F90
. move dyn_eul_readnl to dyn_comp (rename dyn_readnl).
components/cam/src/dynamics/eul/inital.F90
. remove namelist file from arg lists of cam_initial and dyn_init
components/cam/src/dynamics/eul/restart_dynamics.F90
. remove namelist file from arg lists of read_restart_dynamics and dyn_init
components/cam/src/dynamics/fv/cd_core.F90
. code cleanup -- mainly adding consistent indentation to loop structure
components/cam/src/dynamics/fv/ctem.F90
. update namelist reader and call it from dyn_readnl rather than ctem_init
components/cam/src/dynamics/fv/dyn_comp.F90
. code cleanup -- mainly adding consistent indentation to loop structure
. add dyn_readnl (copied from fv_control_mod)
- update the mpibcast calls to use mpi_bcast
- change namelist variables to have consistent fv_ prefix:
. move spmd_readnl call to dyn_readnl
. move ctem_readnl call to dyn_readnl
components/cam/src/dynamics/fv/fv_control_mod.F90
. move dyn_readnl to dyn_comp.
components/cam/src/dynamics/fv/inital.F90
. remove namelist file from arg lists of cam_initial and dyn_init
components/cam/src/dynamics/fv/restart_dynamics.F90
. remove namelist file from arg lists of read_restart_dynamics and dyn_init
components/cam/src/dynamics/se/dyn_comp.F90
. remove call to dyn_readnl from dyn_init1.
. move call to native_mapping_readnl from dyn_init1 to dyn_readnl
. remove nlfilename arg from dyn_init1
components/cam/src/dynamics/se/inital.F90
. remove namelist file from arg lists of cam_initial and dyn_init1
components/cam/src/dynamics/se/native_mapping.F90
. update namelist reader to use mpi_bcast
components/cam/src/dynamics/se/restart_dynamics.F90
. remove namelist file from arg lists of read_restart_dynamics and dyn_init1
components/cam/src/dynamics/sld/dyn_comp.F90
. add dyn_readnl (copied from dyn_sld_readnl)
- update the mpibcast calls to use mpi_bcast
- change dyn_spectral_* namelist variables to sld specific ones, and
change the namelist group from dyn_spectral_inparm to dyn_sld_inparm
. move spmd_readnl call to dyn_readnl
components/cam/src/dynamics/sld/inital.F90
. remove namelist file from arg lists of cam_initial and dyn_init
components/cam/src/dynamics/sld/restart_dynamics.F90
. remove namelist file from arg lists of read_restart_dynamics and dyn_init
components/cam/src/dynamics/sld/sld_control_mod.F90
. move dyn_sld_readnl to dyn_comp and rename to dyn_readnl
components/cam/src/physics/cam/clubb_intr.F90
. mod to provide horizontal cell size estimates
components/cam/test/system/nl_files/ttrac
. dyn_spectral_divdampn changed to eul_divdampn
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS
hobart/nag: All PASS
hobart/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_4_45
Originator(s): fvitt
Date: 4 Feb 2016
One-line Summary: Miscellaneous improvements to diagnostics
Purpose of changes:
- provide capability to output standard deviations over the sampling period
- output zonal-mean FV TEM diagnostics to a zonal-mean grid in the history file
- improve performance of the sathist output
- include vertically integrated rate diagnostic in NEU wet deposition
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton, goldy
List all subroutines eliminated:
List all subroutines added and what they do:
A components/cam/test/system/config_files/f4c5wmclbdh
List all existing files that have been modified, and describe the changes:
M components/cam/src/control/cam_history_support.F90
- added sbuf data buffer which stores the accumulation of history fields squared \
for use in standard deviation output
- added sbuf_varid array for use in output of sbuf data to history restart files
M components/cam/src/control/cam_history.F90
- Standard deviation is calculated using the method of computing variance outlined in
http://www.johndcook.com/blog/standard_deviation/
- Changes were made to support the additional data buffer (sbuf) which holds the
accumulating field squared, including persistence in the history restart files
when needed.
- Replaced module variable "i" with "idx". The module variable i was causing confusion
in subroutines that used i. Some subroutines declared a local variable i while others
did not. In subroutine h_normalize i was used without declaring it as a local variable
which led to a threading race condition.
M components/cam/src/control/cam_history_buffers.F90
- new accumulation method added to accumulate the history field squared
and store it in sbuf
M components/cam/bld/namelist_files/namelist_definition.xml
- document the standard deviation option avgflag_pertape ('S')
M components/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
- check for MOZ wet dep method rather than .not.do_neu_wetdep
- clean up
M components/cam/src/chemistry/mozart/mo_neu_wetdep.F90
- added WD_* diagnostics (vertically integrated rates -- kg/s)
M components/cam/src/chemistry/mozart/mo_chm_diags.F90
- add/output WD history fields only if MOZ wet dep method is used
- clean up
M components/cam/src/chemistry/mozart/chemistry.F90
- removed unused chem_name from chemdr argument list
M components/cam/src/dynamics/fv/ctem.F90
- clean up of output of zonal mean fields including removal of the
output of the zonal mean fields on the horizontal grid
M components/cam/src/control/sat_hist.F90
- performance improved by grouping the output of columns of simularly decomposed
variables that share an iodesc object rather than creating a new iodesc object for
each variable
M components/cam/test/system/nl_files/sat_hist
- added standard deviation output test
- '3d' fields renamed to '2d' -- zonal means
M components/cam/test/system/input_tests_master
M components/cam/test/system/tests_pretag_yellowstone
- reduce resolution waccm5+clubb yellowstone regression test
to speedup the test time
M components/cam/bld/namelist_files/use_cases/sd_1975-2010_ccmi_refc1_waccm_tsmlt.xml
M components/cam/bld/namelist_files/use_cases/2000_cam4_trop_chem.xml
M components/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_waccm_tsmlt.xml
M components/cam/bld/namelist_files/use_cases/volc_waccm_ma_cam5.xml
M components/cam/bld/namelist_files/use_cases/1850_waccm_tsmlt_cam5.xml
M components/cam/bld/namelist_files/use_cases/2007-2011_htap_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/waccm_1850_cam4.xml
M components/cam/bld/namelist_files/use_cases/waccm_1850_cam5.xml
M components/cam/bld/namelist_files/use_cases/volc_waccm_tsmlt_megan_cam5.xml
M components/cam/bld/namelist_files/use_cases/1850_cam4_super_fast_llnl.xml
M components/cam/bld/namelist_files/use_cases/sd_waccm_sulfur.xml
M components/cam/bld/namelist_files/use_cases/2004-2100_ccmi_refc2_rcp45_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/2005-2100_cam4_rcp45.xml
M components/cam/bld/namelist_files/use_cases/2000_ccmi_refc1_waccm_tsmlt.xml
M components/cam/bld/namelist_files/use_cases/2005-2100_cam4_rcp85.xml
M components/cam/bld/namelist_files/use_cases/cam4_trop_strat_chem.xml
M components/cam/bld/namelist_files/use_cases/cam4_chem_radpsv_geos5.xml
M components/cam/bld/namelist_files/use_cases/1850-2005_cam4_super_fast_llnl.xml
M components/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp45.xml
M components/cam/bld/namelist_files/use_cases/1979-2010_sd_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_rcp85_waccm_ma.xml
M components/cam/bld/namelist_files/use_cases/2005-2100_cam4_rcp45_bgc.xml
M components/cam/bld/namelist_files/use_cases/waccm_1955-2005_cam4.xml
M components/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp85.xml
M components/cam/bld/namelist_files/use_cases/2005-2100_cam4_rcp85_bgc.xml
M components/cam/bld/namelist_files/use_cases/waccm_1955-2005_cam5.xml
M components/cam/bld/namelist_files/use_cases/waccm_1974-2005_cam4.xml
M components/cam/bld/namelist_files/use_cases/waccm_refb1.xml
M components/cam/bld/namelist_files/use_cases/volc_waccm_tsmlt_nomegan_cam5.xml
M components/cam/bld/namelist_files/use_cases/waccm_carma_bc_2013_cam4.xml
M components/cam/bld/namelist_files/use_cases/2005-2100_cam4_strataero_rcp45.xml
M components/cam/bld/namelist_files/use_cases/rcp6.0_waccm_tsmlt_nomegan_cam5.xml
M components/cam/bld/namelist_files/use_cases/waccmx_1996_cam4.xml
M components/cam/bld/namelist_files/use_cases/1950-2010_ccmi_refc1_waccm_ma.xml
M components/cam/bld/namelist_files/use_cases/2000_cam4_trop_moz_soa.xml
M components/cam/bld/namelist_files/use_cases/sd_waccm_geos5.xml
M components/cam/bld/namelist_files/use_cases/waccmx_2000_cam4.xml
M components/cam/bld/namelist_files/use_cases/waccm_2000_cam4.xml
M components/cam/bld/namelist_files/use_cases/waccm_2000_cam5.xml
M components/cam/bld/namelist_files/use_cases/1850_cam4.xml
M components/cam/bld/namelist_files/use_cases/sd_1975-2010_ccmi_refc1_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/1950-2010_ccmi_refc1_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/2005-2100_cam4_rcp60.xml
M components/cam/bld/namelist_files/use_cases/1850-2005_cam4.xml
M components/cam/bld/namelist_files/use_cases/2005-2100_cam4_rcp26.xml
M components/cam/bld/namelist_files/use_cases/1950-2010_ccmi_refc1_waccm_tsmlt.xml
M components/cam/bld/namelist_files/use_cases/waccm_1850-2005_cam4.xml
M components/cam/bld/namelist_files/use_cases/waccmxi_2000_cam4.xml
M components/cam/bld/namelist_files/use_cases/1850-2005_cam4_bgc.xml
M components/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp26.xml
M components/cam/bld/namelist_files/use_cases/sd_waccm5_geos5.xml
M components/cam/bld/namelist_files/use_cases/sd_1975-2010_ccmi_refc1_waccm_ma.xml
M components/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_waccm_ma.xml
- *2d ctem history fields replaced by *zm fields
- removed unused WDR_* history fields
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
035 bl359 TBL.sh f4c5wmclbdh outfrq3s_clubb+waccm_2000_cam5 9s ................................FAIL! rc= 7 at Thu Feb 4 05:28:10 MST 2016
042 bl373 TBL.sh f1.9c4mozdh outfrq3s+2000_cam4_trop_chem 9s ..................................FAIL! rc= 7 at Thu Feb 4 05:57:04 MST 2016
049 bl374 TBL.sh f1.9c4fastdh outfrq3s+1850-2005_cam4_super_fast_llnl 9s ......................FAIL! rc= 7 at Thu Feb 4 06:12:58 MST 2016
test bl359 is a new test
tests bl373 and bl374 are bit-for-bit -- correction to WD_* wet dep diagnostics cause these tests to fail
hobart/nag:
034 bl318 TBL.sh f10c4cdm sat_hist 9s .........................................................FAIL! rc= 7 at Wed Feb 3 17:49:58 MST 2016
This is bit-for-bit -- the additional history stream for standard deviations causes this baseline test fail
hobart/pgi: all pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_4_44
Originator(s): goldy, cacraig, chris.kerr.llc@gmail.com
Date: 2016-01-30
One-line Summary: FV npr_yz fix plus MG2 optimization
Purpose of changes:
- Fix bug when FV dycore is not using all processors
- Minor MG2 peformance fix contributed by Chri Kerr
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: goldy, cacraig
List all subroutines eliminated: NA
List all subroutines added and what they do: NA
List all existing files that have been modified, and describe the changes:
M components/cam/doc/ChangeLog
- Now with more changes than ever before!
M components/cam/src/dynamics/fv/dyn_grid.F90
- Fixed bug with use of npr_yz namelist when FV dycore ended up using fewer
tasks than CAM overall
M components/cam/src/physics/cam/micro_mg_utils.F90
- Moved non-repeating code out of loop to improve performance.
Noted by Chris Kerr.
M components/cam/src/utils/cam_grid_support.F90
- Fixed possible problems when some PEs are not part of a distributed grid
M components/cam/test/system/input_tests_master
- added npr_yz test to sm354
A components/cam/test/system/nl_files/outfrq3s_npryz
- added npr_yz test
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All pass (but see note about br397 below)
hobart/nag: All pass
hobart/pgi: All pass
CAM tag used for the baseline comparison tests if different than previous
tag: Note: Softlink was provided to allow modified sm354 to pass BFB test
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_4_43
Originator(s): eaton, erik, julio, cacraig
Date: Jan 28, 2016
One-line Summary: Keep CAM output out of cesm.log, revise the COSP regression test, new ncdata file for ne30np4_L32
Purpose of changes:
- QNEG messages which were not on task 0, were being passed into cesm.log. Erik pointed out
where iulog was not being broadcast to the other tasks.
- The COSP setup for CAM regression testing was setting a number of cosp-specific flags. When
debug was turned on, one of these flags caused a floating point exception down deep in cosp. We
have removed the flags and just test cosp with its default settings. This change results in a branch
test run have issues. Per Brian Eaton's suggestion, the branch test has been removed and a
bugzilla has been filed (#2278)
- Julio created a new file for ne30np4_L32 (supports homme cam5.5 runs)
Bugs fixed (include bugzilla ID):
2262: CAM regression testing now tests cosp with the default settings
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
- added file for homme cam5.5 initial data conditions
M components/cam/src/cpl/atm_comp_mct.F90
- added mpi broadcast for iulog
M components/cam/test/system/input_tests_master
- modified cosp test
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB except
056 bl397 TBL.sh f1.9c4cm outfrq3s+1850-2005_cam4 9s ..........................................FAIL! rc= 7 at Wed Jan 27 18:17:09 MS
T 2016
-- This was expected to be different due to changing the settings for cosp
055 br397 TBR.sh f1.9c4cm outfrq3s+1850-2005_cam4 6+3s ........................................FAIL! rc= 12 at Wed Jan 27 18:17:09 M
ST 2016
-- This branch test fails due to removing the settings for cosp. Bugzilla 2278 has been filed
and this test is removed until it is solved
hobart/nag: all BFB
hobart/pgi or jaguar/pgi: all BFB
Summarize any changes to answers, i.e.,
- what code configurations: Testing is now using different settings for cosp - no changes were made to cosp itself
===============================================================
===============================================================
Tag name: cam5_4_42
Originator(s): goldy, eaton
Date: 2016-01-25
One-line Summary:
Purpose of changes: Provide native namelist support for CAM-SE
Bugs fixed (include bugzilla ID): NA
Describe any changes made to build system: NA
Describe any changes made to the namelist:
- Summary: Create formal definitions for all supported HOMME variables.
Changed names for consistency (all start with "se_")
Group name changed from ctl_nl to dyn_se_inparm
Removed items which only had one valid value in CAM
Added support for refined mesh simulations
- hypervis_order ==> se_hypervis_order (rename)
- hypervis_subcycle ==> se_hypervis_subcycle (rename)
- hypervis_subcycle_q ==> se_hypervis_subcycle_q (rename)
- nu ==> se_nu (rename)
- nu_div ==> se_nu_div (rename)
- nu_p ==> se_nu_p (rename)
- nu_q ==> se_nu_q (rename)
- nu_top ==> se_nu_top (rename)
- qsplit ==> se_qsplit (rename)
- rsplit ==> se_rsplit (rename)
- statefreq ==> se_statefreq (rename)
- tstep_type ==> se_tstep_type (rename)
- vert_remap_q_alg ==> se_vert_remap_q_alg (rename)
- integration: Removed
- se_partmethod: Removed
- se_phys_tscale: Removed
- se_topology: Removed
- se_fine_se: Added to support refined mesh simulations
- se_hypervis_power: Added to support refined mesh simulations
- se_hypervis_scaling: Added to support refined mesh simulations
- se_max_hypervis_courant: Added to support refined mesh simulations
- se_mesh_file: Added to support refined mesh simulations
- se_npes: Added to support different number of SE tasks
- se_refined_mesh: Added to support refined mesh simulations
List any changes to the defaults for the boundary datasets: NA
Describe any substantial timing or memory changes: NA
Code reviewed by: eaton
List all subroutines eliminated:
- dynamics/se/share/namelist_mod.F90: readnl -- No longer called from CAM-SE
- dynamics/se/spmd_utils: spmd_readnl -- No longer needed
List all subroutines added and what they do:
- dynamics/se/dyn_comp.F90: dyn_readnl -- New namelist reader for CAM-SE
- dynamics/se/share/namelist_mod.F90: homme_set_defaults -- Called from dyn_readnl
- dynamics/se/share/namelist_mod.F90: homme_postprocess_namelist -- Called from dyn_readnl
List all existing files that have been modified, and describe the changes:
M components/cam/SVN_EXTERNAL_DIRECTORIES
- Switch HOMME external to new tag with namelist and other fixes.
M components/cam/bld/build-namelist
- Rewrite of logic to build namelist for SE dycore to use full CAM namelist support.
M components/cam/bld/configure
- Fixed comment about WACCM physics
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
M components/cam/bld/namelist_files/namelist_definition.xml
- Modified namelist item names (added se_ -- see above) and added new items
to allow CAM to be fully in charge of reading the dynamics namelist
M components/cam/doc/ChangeLog
- New and improved!
M components/cam/src/control/cam_history.F90
- Fixed an incorrect WARNING message
M components/cam/src/dynamics/se/dyn_comp.F90
- Added dyn_readnl (currently called from dyn_init1) and set HOMME variables.
- Do not call HOMME's readnl
D components/cam/src/dynamics/se/physical_constants.F90
- This was a temporary module put in by Brian waiting for a new HOMME tag
(see statement in cam5_4_40 section below).
M components/cam/src/dynamics/se/restart_dynamics.F90
- Took out unsed but now invalid use of the HOMME readnl subroutine
M components/cam/src/dynamics/se/spmd_dyn.F90
- Took out no-longer-needed spmd_readnl routine which was mostly reading
Eulerian dycore variables. The number of pes passed to HOMME is now
properly handled by build_namelist
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All pass
hobart/nag: All pass
hobart/pgi: All pass
CAM tag used for the baseline comparison tests if different than previous
tag: NA
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate): NA
If bitwise differences were observed, how did you show they were no worse
than roundoff? NA
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository): NA
===============================================================
===============================================================
Tag name: cam5_4_41
Originator(s): fvitt foster
Date: 25 Jan 2016
One-line Summary: Updates to WACCM geomagnetic coordinates module
Purpose of changes:
- allow WACCM physics to run on unstructured SE grid
- remove dynamics grid dependencies in mo_apex
- refactor of the legacy apex geomagnetic coordinates code (Ben Foster)
- encapsulate apex routines in a Fortran module
- general cleanup
- add unit test for aurora module
- add CIME test for WACCM on SE grid
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
D components/cam/src/chemistry/mozart/apex_subs.F90
- Ben Foster -- encapsulate apex subroutines in a module
D components/cam/test/system/nl_files/outfrq3s_wse
- should not need special namelist settings to test waccm with SE dycore
List all subroutines added and what they do:
A components/cam/src/chemistry/utils/apex.F90
- Ben Foster -- refactor legacy apex code
- encapsulate apex subroutines in a module
NOTE: produces near-identical results as the legacy code -- not B4B
A components/cam/test/system/config_files/h5c4wmdm
- regression test for waccm with SE dycore
A components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq9s_wse/user_nl_cam
A components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq9s_wse/shell_commands
A components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq9s_wse/user_nl_clm
A components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq9s_wse
- CIME test for WACCM with SE dycore
A components/cam/src/unit_drivers/aur/unit_driver.F90
A components/cam/src/unit_drivers/aur
- unit test for aurora module
List all existing files that have been modified, and describe the changes:
M components/cam/src/control/cam_control_mod.F90
- moved magnetic year to mo_apex
M components/cam/src/physics/cam/physpkg.F90
- initialize APEX if needed
M components/cam/src/physics/waccm/iondrag.F90
M components/cam/src/chemistry/mozart/mo_aurora.F90
M components/cam/src/chemistry/mozart/mo_apex.F90
M components/cam/src/chemistry/mozart/efield.F90
M components/cam/src/chemistry/mozart/exbdrift.F90
M components/cam/src/chemistry/mozart/spedata.F90
M components/cam/test/system/input_tests_master
- regression tests updates
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
- set defaults:
- nu* (hyperviscosity) for ne5np4 resolution
- ncdata for WACCM on ne5np4 and ne30np4 resolutions
- rsplit for ne5np4 resolution
- se_nsplit for WACCM on ne5np4 resolution
M components/cam/cime_config/config_pes.xml
- use a large number of cores with WACCM on SE ne30 grid
M components/cam/cime_config/testdefs/testlist_cam.xml
- add CIME test for WACCM on SE grid
M components/cam/bld/configure
M components/cam/bld/config_files/definition.xml
- unit test for aurora module
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
009 bl203 TBL.sh fsd1.9c4wtsmlth outfrq3s_sd+sd_1975-2010_ccmi_refc1_waccm_tsmlt 9s ...........FAIL! rc= 7 at Fri Jan 22 14:45:31 MST 2016
029 bl336 TBL.sh f4c4wmdh outfrq3s+waccm_1850_cam4 9s .........................................FAIL! rc= 7 at Fri Jan 22 15:37:45 MST 2016
035 bl365 TBL.sh f1.9c5wmclbdh outfrq3s_clubb+waccm_2000_cam5 9s ..............................FAIL! rc= 7 at Fri Jan 22 16:59:01 MST 2016
052 bl390 TBL.sh f1.9c4wmxh outfrq3s+waccmx_2000_cam4 9s ......................................FAIL! rc= 7 at Fri Jan 22 17:53:02 MST 2016
062 bl429 TBL.sh f1.9c4wmxh outfrq3s+waccmxi_2000_cam4 9s .....................................FAIL! rc= 7 at Fri Jan 22 18:16:39 MST 2016
070 bl470 TBL.sh fsd1.9c5.4wtclbh outfrq3s_sd+sd_waccm5_geos5 9s ..............................FAIL! rc= 7 at Fri Jan 22 19:05:48 MST 2016
083 bl379 TBL.sh f1.9c4wmh outfrq24h+waccm_2005-2100_cam4_rcp45 2d ............................FAIL! rc= 7 at Fri Jan 22 20:16:13 MST 2016
hobart/nag:
067 bl426 TBL.sh f10c4wmxdm outfrq3s_ionos 9s .................................................FAIL! rc= 7 at Fri Jan 22 14:54:14 MST 2016
070 bl471 TBL.sh f10c5wmam3dm outfrq3s_modalstrat 9s ..........................................FAIL! rc= 7 at Fri Jan 22 15:08:57 MST 2016
hobart/pgi: All pass
These waccm/waccmx baseline failures are expected. The new apex module produce
near-identical results as the legacy code although larger than roundoff. These
difference are not expect to be climate changing.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: waccm/waccmx
- what platforms/compilers: all
- nature of change : larger than roundoff but same climate
If bitwise differences were observed, how did you show they were no worse
than roundoff?
The new apex module produces near-identical results as the legacy code although
larger than roundoff. These difference are not expect to be climate changing.
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
https://svn-ccsm-models.cgd.ucar.edu/cam1/branch_tags/apex_cam5_4_39_tags/apex01_cam5_4_39
- platform/compilers:
yellowstone/intel
- configure commandline:
create_newcase -compset F55WCN -res f19_f19 -mach yellowstone -case /glade/u/home/fvitt/cesm/cases/F55WCN_cam5_4_39_f19_cntl.001
create_newcase -compset F55WCN -res f19_f19 -mach yellowstone -case /glade/u/home/fvitt/cesm/cases/F55WCN_cam5_4_39_f19_test.001
- build-namelist command (or complete namelist): default for F55WCN compset with these changes
user_nl_cam:
ncdata = '/glade/p/work/fvitt/f.e11.FWTREFC1.f19_f19.ccmi31.001.cam.i.1989-01-01-00000.nc'
fincl1 = 'QSUM','ALONM','ALATM','QRS_TOT','QRL_TOT','QNO','QRLNLTE',
'QRS_AUR','EPOTEN'
fincl2 = ' '
fincl3 = ' '
fincl4 = ' '
fincl5 = ' '
fincl6 = ' '
env_run.xml:
- MSS location of output:
/home/fvitt/csm/F55WCN_cam5_4_39_f19_test.001
MSS location of control simulations used to validate new climate:
/home/fvitt/csm/F55WCN_cam5_4_39_f19_cntl.001
===============================================================
===============================================================
Tag name: cam5_4_40
Originator(s): eaton
Date: Fri Jan 22 09:19:04 MST 2016
One-line Summary: Add fixes and compset for simplified model test; update cime external.
Purpose of changes:
. The simplified model test case has been renamed to DABI (dry adiabatic
baroclinic instability). This is reflected in the name of the use case
file which implements the test: dabi_p2004.xml. A compset, FDABIP04, has
also been defined for running this test from the CESM scripts.
. The parameter that controls the top level for the del*N horizontal
diffusion in the Eulerian dycore can now be set using a namelist
variable (eul_hdif_kmnhdn).
. Fix constants in dabi_p2004 use case file to match Table 1 in Polvani et
al. (2004)
. Update to cime4.4.0 and add option to build pio2 (-pio2 switch to
configure). pio1 remains the default for cam standalone builds.
Bugs fixed (include bugzilla ID):
. The type of eul_hdif_coef in the namelist_definition.xml file was
integer. Change it to type real.
. Add initializer in fv/metdata.F90 to fix debug mode failures in some FV
specified dynamics configurations.
Describe any changes made to build system:
. Add option to build pio2 library via setting -pio2 switch to configure. Note
that this library does not currently work with the cam standalone build
which uses a serial netcdf library. When -pio2 is set then pio2 is built
as a separate library using pio's cmake. This location can be installed
in a user specified location by setting the new configure option
-pio2_install_dir.
Describe any changes made to the namelist:
. add eul_hdif_kmnhdn
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
components/cam/bld/namelist_files/use_cases/adiabatic_LC1.xml
. renamed to dabi_p2004.xml
List all subroutines added and what they do:
components/cam/bld/namelist_files/use_cases/dabi_p2004.xml
. use case adiabatic_LC1 renamed to dabi_p2004 (dry adiabatic baroclinic
instability, Polvani et al. 2004)
src/dynamics/se/physical_constants.F90
. copy of se/share/physical_constants.F90 with parameter attribute removed
from Cp so that it can be set from a namelist variable.
List all existing files that have been modified, and describe the changes:
SVN_EXTERNAL_DIRECTORIES:
. cime4.2.3 -> cime4.4.0
components/cam/bld/build-namelist
. add default for eul_hdif_kmnhdn
components/cam/bld/config_files/definition.xml
. add pio_build_dir and pio_install_dir
components/cam/bld/configure
. update the Filepath file for pio1 source code
. add -pio2 as switch to enable building pio2 as a separate library. The
default is to use pio1 and to build it as part of the cam executable.
. add -pio2_install_dir option to allow user to specify an install location
for the pio2 library. This location is checked by configure for the
presence of an existing library, and if a library already exists then it
is used.
. Update to the use of nc-config utility for constructing the makefile
options needed to link netcdf libraries.
. Add code to run the PIO2 cmake configuration for building PIO2 as a
separate library.
components/cam/bld/Makefile.in
. Add macros PIO_BUILD_DIR and PIO_INSTALL_DIR. If these are set then PIO2
will be built and installed as a separate library.
. Add rules to build and install PIO2 as separate library.
components/cam/bld/namelist_files/namelist_defaults_cam.xml
. add default value for eul_hdif_kmnhdn
components/cam/bld/namelist_files/namelist_definition.xml
. change eul_hdif_coef type from integer to real
. clarify definition of dyn_spectral_dif2
. add definition for eul_hdif_kmnhdn
. add definition for cpair
. add pio_rearranger for pio2
components/cam/cime_config/buildnml
. strip level info from grid name and set up -nlev arg for configure
components/cam/cime_config/config_component.xml
components/cam/cime_config/config_compsets.xml
. add definition for FDABIP04
components/cam/src/dynamics/eul/eul_control_mod.F90
. change kmnhdn from hardcoded parameter value to namelist variable
(eul_hdif_kmnhdn)
. add eul_hdif_kmnhdn to log output
components/cam/src/dynamics/fv/metdata.F90
. add initializer for met_q so that it may safely be used in an array
expression.
components/cam/src/physics/cam/co2_cycle.F90
. c_cp now set in the co2_register method rather than with an initializer
in the declaration statement. The initializer is no longer valid since
cpair is no longer a parameter.
components/cam/src/physics/carma/cam/carma_constants_mod.F90
components/cam/src/physics/carma/cam/carma_intr.F90
. move init of CP to carma_init
components/cam/src/utils/physconst.F90
. add protected attribute to public variables
. change cpair from parameter to variable, and set via namelist
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS
hobart/nag: All PASS
hobart/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_4_39
Originator(s): eaton
Date: Tue Jan 19 08:13:10 MST 2016
One-line Summary: Refactor radiation interfaces.
Purpose of changes:
. Refactor radiation interfaces in preparation for adding the new RRTMGP
parameterization.
- move the old cam_rt radiation code into its own directory
(src/physics/cam_rt) parallel to the rrtmg directory
- make radiation modules responsible for the data they need to read and
write from restart files (needed to do this to allow moving radae into
a directory which is not built unless cam_rt is active)
- make cam_rt radiation module responsible for calling
initialize_radbuffer (this was tangled up with restart functionality)
- redesign the RRTMG output type to only contain the diagnostics for a
single climate or diagnostic calculation (i.e., remove the N_DIAG
dimension). Remove components not needed for output, or which were
duplicated in the pbuf or cam_out objects.
Bugs fixed (include bugzilla ID):
. resolve bug 2254 by adding scale factors to the fields FUSC, FDSC in the
camrt code, and removing the scale factors applied to FUS, FDS in the
rrtmg code. These fields are only output for scm_crm_mode which is not
tested.
. create_newcase was failing on FADIAB compset
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: didn't check
Code reviewed by: review session on 6 Jan 2016
List all subroutines eliminated:
components/cam/src/physics/cam/cloud_rad_props.F90
. remove stub module
components/cam/src/physics/cam/hirsbt.f90
components/cam/src/physics/cam/hirsbtpar.f90
. remove unused diagnostic
components/cam/src/physics/cam/rad_solar_var.F90
. move file to physics/camrt/rad_solar_var.F90
components/cam/src/physics/cam/radae.F90
. move file to physics/camrt/radae.F90
components/cam/src/physics/cam/radiation.F90
. move file to physics/cam_rt/radiation.F90
components/cam/src/physics/cam/radiation_utils.F90
. remove file - move definition of rad_diagdata_type to radiation module
components/cam/src/physics/rrtmg/radiation_utils.F90
. remove file - move definition of rad_diagdata_type to radiation module
components/cam/src/physics/cam/radlw.F90
. move file to physics/camrt/radlw.F90
components/cam/src/physics/cam/radsw.F90
. move file to physics/camrt/radsw.F90
List all subroutines added and what they do:
components/cam/src/physics/camrt/rad_solar_var.F90
. moved here from physics/cam/rad_solar_var.F90
components/cam/src/physics/camrt/radae.F90
. moved here from physics/cam/radae.F90
. call initializae_radbuffer from radae_init. Need to check that the
allocatable arrays haven't already been allocated (if restart or branch
run).
components/cam/src/physics/camrt/radconstants.F90
. moved here from physics/cam/radconstants.F90
components/cam/src/physics/camrt/radiation.F90
. moved here from physics/cam/radiation.F90
. remove use of mpishorthand from namelist reader
. move definition of rad_diagdata_type into this module, shorten name to rad_out_t
. radiation_init
- add call to rad_solar_var_init
. radiation_tend
- remove unused args icefrac and snowh
- make rd optional
- code cleanup
. add radiation_define_restart
- move code here from init_restart_physics
. add radiation_write_restart
- move code here from write_restart_physics
. add radiation_read_restart
- move code here from read_restart_physics
components/cam/src/physics/camrt/radlw.F90
. moved here from physics/cam/radlw.F90
components/cam/src/physics/camrt/radsw.F90
. moved here from physics/cam/radsw.F90
. multiply fusc and fdsc by 1.e-3 in the outfld call. This is consistent
with the treatment of fsntc (done in radiation_tend).
List all existing files that have been modified, and describe the changes:
components/cam/bld/configure
. add components/cam/src/physics/camrt to filepaths unless using rrtmg.
components/cam/cime_config/config_compsets.xml
. remove the docn and clm entries since these variables are not defined if
using stub components as the ideal, adiabatic, and aquaplanet compsets do.
Move entries to confi_component.xml
components/cam/cime_config/config_component.xml
. move entries for docn and clm here from config_compsets.xml
. Supplement the new entries with definition elements. Find this info in
the docn and clm config_compsets.xml files and add it to the new entries.
components/cam/src/control/cam_restart.F90
. remove use of get_abs_restart_filepath which was writing nonsense into
the rpointer file.
. remove reading/writing of aeres_int. This was only being used to control
the writing of nonsense (previous bullet) into the rpointer file
components/cam/src/control/startup_initialconds.F90
. initial_conds -- remove initialize_radbuffer
components/cam/src/physics/cam/micro_mg_cam.F90
. moved addfld/outfld calls for CLDFSNOW here from rrtmg/radiation.F90.
Also changed the default averaging from 'I' to 'A'.
components/cam/src/physics/cam/physpkg.F90
. remove unused ref to radiation_do
. phys_init
- remove call to cloud_rad_props_init
- remove call to rad_solar_var_init
. tphysbc
- remove local radiation output object since this was changed to be an
optional arg in the radiation_tend interface
- get rid of unused icefrac and snowhland args from radiation_tend call
components/cam/src/physics/cam/radconstants.F90
. This is now a dummy module which contains module variables that are
referenced by modal_aero_data, modal_aer_opt, aer_rad_props, phys_prop,
radiation_data, rad_constituents, carma_constants_mod, carma_intr. When
radiation is active the version of this file from the appropriate
radiation subdirectory will be compiled instead.
components/cam/src/physics/cam/restart_physics.F90
. move code for radiation to new routines in the radiation interface
modules.
. make calls to new routines: radiation_define_restart,
radiation_write_restart, radiation_read_restart
. remove broken/unneeded routine get_abs_restart_filepath. This was used
when the camrt abs-ems restart data was written to a separate restart
file. It appears the separate file was eliminated when PIO was
implemented to write restarts.
components/cam/src/physics/rrtmg/radiation.F90
. remove use of mpishorthand from namelist reader
. radiation_init
- add call to cloud_rad_props_init
- add call to rad_solar_var_init
. move definition of rad_diagdata_type into this module, shorten name to rad_out_t
. radiation_tend
- remove unused args icefrac and snowh
- make rd optional
- remove the unused hirsbt code
- code cleanup
. add radiation_define_restart, radiation_write_restart,
radiation_read_restart
. remove N_DIAG dimension from components of rad_out_t
. add diag(icall) suffix to FSNR and FLNR
. move addfld/outfld for CLDFSNOW to micro_mg_cam.F90
. add routine radiation_output_cld to output the cloud optics fields
. refactor radiation_output_sw and radiation_output_lw to allow them to be
called inside the loop over climate and diagnostic calcs.
. remove ice_tau, liq_tau, cld_tau, snow_tau, c_cld_tau, cldfprime,
aer_tau, cld_lw_abs, snow_lw_abs from rad_out_t.
These are local vars not used for diagnostic output.
. remove qrs, fsns, fsnt, fsds, qrl, flnt, flns, su, sd, lu, ld, from
rad_out_t. These are in pbuf.
. remove sols, soll, solsd, solld from rad_out_t. These are in cam_out.
. remove swcf, lwcf from rad_out_t. Only need to compute in output routines.
components/cam/src/physics/rrtmg/radsw.F90
. remove scaling of fus and fds by 1.e-3 in the outfld calls.
components/cam/src/unit_drivers/drv_input_data.F90
. fix log output for driving data file
components/cam/src/unit_drivers/rad/unit_driver.F90
. update radiation interfaces
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS
hobart/nag: All PASS
hobart/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_4_38
Originator(s): jedwards, cecile, cacraig
Date: Jan 14, 2016
One-line Summary: Changes to support pio2; turnoff icesupersat as default for CLUBB and CAM5.5
Purpose of changes:
- Jim Edwards made minor changes to CAM to support pio2
- It was decided that at this point, CAM did not want to bring in a new climate with
the turning on of ice supersaturation by default. It can still be turned on by users
as it is a namelist parameter.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
- ice_supersat is set to .false. for CLUBB runs (reverses the setting implemented in cam5_4_37)
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
- set ice_supersat to .false. for CLUBB runs
M components/cam/src/control/sat_hist.F90
M components/cam/src/dynamics/fv/metdata.F90
- changes implemented by Jim Edwards to support pio2
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB except:
035 bl365 TBL.sh f1.9c5wmclbdh outfrq3s_clubb+waccm_2000_cam5 9s ..............................FAIL! rc= 7 at Wed Jan 13 21:12:59 MST 2016
038 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Wed Jan 13 21:26:07 MST 2016
070 bl470 TBL.sh fsd1.9c5.4wtclbh outfrq3s_sd+sd_waccm5_geos5 9s ..............................FAIL! rc= 7 at Wed Jan 13 23:13:20 MST 2016
hobart/nag: all BFB except:
052 bl356 TBL.sh fsd1.9c5clbdm off1.9x2.5 9s ..................................................FAIL! rc= 7 at Wed Jan 13 17:20:36 MST 2016
058 bl369 TBL.sh f1.9c5clbdm outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Wed Jan 13 17:44:15 MST 2016
hobart/pgi or jaguar/pgi: all BFB
All CLUBB tests are expected to be different due to turning off ice_supersat when running CLUBB. This reverses the
"new climate" that was introduced in cam5_4_37. According to Pete Bogenschutz the net effect of cam5_4_37 and
cam5_4_38 taken together will be "answer changing" and NOT "climate changing"
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_4_37
Originator(s): bogensch, cacraig
Date: January 13, 2016
One-line Summary: Ice supersat on by default for CAM5.5 and CLUBB; mods to energy conservation fixer
Purpose of changes:
Turn on ice supersaturation parameterization when CAM5.5 physics is called
to remove access RH bias in the upper troposphere and stratosphere. In addition,
modify the energy conservation fixer in the CAM CLUBB interface to remove small
tendencies CLUBB may produce in layers above which it is active to the layers
where it is active, to avoid dehydration in the stratosphere. This also
includes a minor bug fix to the pre-existing energy fixer.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
- ice supersaturation is now turned on be default anytime CLUBB is turned on
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: cacraig
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
- turn on ice supersaturation by default when using clubb
M components/cam/src/physics/cam/clubb_intr.F90
- modifications to energy conservation fixing
M components/cam/src/physics/cam/macrop_driver.F90
- minor bug fix for ice supersaturation
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB except:
035 bl365 TBL.sh f1.9c5wmclbdh outfrq3s_clubb+waccm_2000_cam5 9s ..............................FAIL! rc= 7 at Tue Jan 12 16:05:56 MST 2016
038 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Tue Jan 12 16:19:41 MST 2016
070 bl470 TBL.sh fsd1.9c5.4wtclbh outfrq3s_sd+sd_waccm5_geos5 9s ..............................FAIL! rc= 7 at Tue Jan 12 18:20:40 MST 2016
hobart/nag: all BFB except:
052 bl356 TBL.sh fsd1.9c5clbdm off1.9x2.5 9s ..................................................FAIL! rc= 7 at Tue Jan 12 12:00:43 MST 2016
058 bl369 TBL.sh f1.9c5clbdm outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Tue Jan 12 12:25:48 MST 2016
All test which use CLUBB are climate changing
hobart/pgi or jaguar/pgi: all BFB
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: CAM5.5 and anything that turns on CLUBB
- what platforms/compilers: all
- nature of change (roundoff; larger than roundoff but same climate; new
climate): new climate
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
http://webext.cgd.ucar.edu/FCLIMO/cam55_FC5_f09f09_version6_052/atm/cam55_FC5_f09f09_version6_052-cam55_FC5_f09f09_cam5_4_22_001/
===============================================================
===============================================================
Tag name: cam5_4_36
Originator(s): eaton
Date: January 12, 2016
One-line Summary: update to cesm1_5_beta03 externals
Purpose of changes:
. Update to cesm1_5_beta03 externals.
. Add new RNG code in cime/share/shr_RandNum/ to the CAM standalone build.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system:
. New optimized code for random number generation is included in the CAM
standalone build. The CAM code doesn't yet reference the new RNG code.
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: not checked
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
SVN_EXTERNAL_DIRECTORIES
. update externals as follows:
cime4.0.1 -> cime4.2.3
mosart_cime3_n02_clm4_5_3_r134 -> clm4_5_6_r156
cice5_20150924 -> lone_cam_tags/lone01_c5435
cime_refactor3_n01_rtm1_0_53 -> rtm1_0_54
cime_refactor_n01_aquap_150301 -> aquap_151016
The cice tag in beta03 (cice5_20151113) needs a check for an empty
CASEROOT variable in order to work with the cam standalone build. That
is the sole change in the branch tag lone01_c5435.
components/cam/bld/Makefile.in
. add option '-std=gnu99' to CFLAGS for gnu, intel, and nag (which uses
gcc) compilers. This was needed for C code that's been added to the cime
external.
components/cam/bld/configure
. add following to Filepath file for new RNG code:
cime/share/shr_RandNum/include
cime/share/shr_RandNum/src
cime/share/shr_RandNum/src/dsfmt_f03
cime/share/shr_RandNum/src/kissvec
cime/share/shr_RandNum/src/mt19937
components/cam/test/system/TCB_ccsm.sh
. remove -silent option to xmlchange commands
. replace case_setup by case.setup
components/cam/test/system/TER_ccsm.sh
components/cam/test/system/TSM_ccsm.sh
. remove -silent option to xmlchange commands
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS except:
009 bl203 TBL.sh fsd1.9c4wtsmlth outfrq3s_sd+sd_1975-2010_ccmi_refc1_waccm_tsmlt 9s ...........FAIL! rc= 7 at Mon Jan 11 13:00:31 MST 2016
042 bl373 TBL.sh f1.9c4mozdh outfrq3s+2000_cam4_trop_chem 9s ..................................FAIL! rc= 7 at Mon Jan 11 15:25:15 MST 2016
046 bl376 TBL.sh f1.9c4mozdh+testmech outfrq3s 9s .............................................FAIL! rc= 7 at Mon Jan 11 15:34:24 MST 2016
064 bl430 TBL.sh f1.9c5mzmam3dh outfrq3s_NEUwetdep 9s .........................................FAIL! rc= 7 at Mon Jan 11 16:22:20 MST 2016
067 bl440 TBL.sh f1.9c4soadh outfrq3s+soa_chem_megan_emis 9s ..................................FAIL! rc= 7 at Mon Jan 11 16:41:17 MST 2016
070 bl470 TBL.sh fsd1.9c5.4wtclbh outfrq3s_sd+sd_waccm5_geos5 9s ..............................FAIL! rc= 7 at Mon Jan 11 17:02:09 MST 2016
These failures are expected due to a change in the deposition parameters
for some terpene species.
hobart/nag: All PASS
hobart/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB except for chem configs that include
terpene species with updated deposition.
===============================================================
===============================================================
Tag name: cam5_4_35
Originator(s): hannay, cacraig
Date: Jan 8, 2016
One-line Summary: tunings to TOA based on coupled runs
Purpose of changes:
- Cecile tuned TOA in coupled runs
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam
M components/cam/SVN_EXTERNAL_DIRECTORIES
- parameter gamma_coef = 0.287 in CLUBB library
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
- dust_emis_fact = 0.60D0 and micro_mg_dcs = 140.D-6 for the CAM5.5 setup
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All BFB except:
035 bl365 TBL.sh f1.9c5wmclbdh outfrq3s_clubb+waccm_2000_cam5 9s ..............................FAIL! rc= 7 at Thu Jan 7 20:57:00 MST 2016
038 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Thu Jan 7 21:10:37 MST 2016
070 bl470 TBL.sh fsd1.9c5.4wtclbh outfrq3s_sd+sd_waccm5_geos5 9s ..............................FAIL! rc= 7 at Thu Jan 7 23:03:36 MST 2016
All CAM 5.5 runs are climate changing
hobart/nag: All BFB except:
052 bl356 TBL.sh fsd1.9c5clbdm off1.9x2.5 9s ..................................................FAIL! rc= 7 at Thu Jan 7 14:02:06 MST 2016
058 bl369 TBL.sh f1.9c5clbdm outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Thu Jan 7 14:24:47 MST 2016
All CAM 5.5 runs are climate changing
hobart/pgi or jaguar/pgi: All BFB (No CAM5.5 tests in this suite)
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: cam5.5
- what platforms/compilers: all
- nature of change (roundoff; larger than roundoff but same climate; new
climate): new climate
----------------
DOCUMENTATION FROM CECILE ON COUPLED SIMULATION TO VALIDATE THIS RUN
----------------
Case directory:
/glade/p/cesmdata/cseg/runs/cesm1_5/b.e15.B1850G.f09_g16.pi_control.28
Local history file:
/glade/p/cesm0005/archive/b.e15.B1850G.f09_g16.pi_control.28
Later in HPPS:
/CCSM/csm/b.e15.B1850G.f09_g16.pi_control.28
Diags:
http://webext.cgd.ucar.edu/B1850/b.e15.B1850G.f09_g16.pi_control.28/
-----------------------------------
# Sandbox (Mariana will create experimental tag at some point)
cd /glade/p/work/hannay/cesm_tags
svn co https://svn-ccsm-models.cgd.ucar.edu/cesm1/tags/cesm1_5_beta03 cesm1_5_beta03_clm_r162_pop_20151215_mosart1_0_14
cd /glade/p/work/hannay/cesm_tags/cesm1_5_beta03_clm_r162_pop_20151215_mosart1_0_14/components/clm
svn switch https://svn-ccsm-models.cgd.ucar.edu/clm2/branch_tags/respmods_tags/respmods_n05_clm4_5_7_r162/components/clm
cd /glade/p/work/hannay/cesm_tags/cesm1_5_beta03_clm_r162_pop_20151215_mosart1_0_14/components/pop
svn switch https://svn-ccsm-models.cgd.ucar.edu/pop2/trunk_tags/cesm_pop_2_1_20151215
cd /glade/p/work/hannay/cesm_tags/cesm1_5_beta03_clm_r162_pop_20151215_mosart1_0_14/components/mosart
svn switch https://svn-ccsm-models.cgd.ucar.edu/mosart/trunk_tags/mosart1_0_14/
cp /glade/scratch/erik/clm_upcime/cime/cime_config/cesm/archive.xml /glade/p/work/hannay/cesm_tags/cesm1_5_beta03_clm_r162_pop_20151215_mosart1_0_14/cime/cime_config/cesm
# compset:
cd /glade/p/work/hannay/cesm_tags/cesm1_5_beta03_clm_r162_pop_20151215_mosart1_0_14/cime/scripts
./create_newcase -case /glade/p/cesmdata/cseg/runs/cesm1_5/b.e15.B1850G.f09_g16.pi_control.28 -user_compset 1850_CAM55_CLM50%BGC_CICE_POP2%ECO_MOSART_CISM1%NOEVOLVE_SWAV_BGC%BDRD -res f09_g16 -mach yellowstone -user_pes_setby allactive -project CESM0005
# clm mods
/glade/p/cesmdata/cseg/runs/cesm1_5/b.e15.B1850G.f09_g16.pi_control.28/SourceMods/src.clm/LunaMod.F90
# cam mods (are in this tag)
/glade/p/cesmdata/cseg/runs/cesm1_5/b.e15.B1850G.f09_g16.pi_control.28/SourceMods/src.cam/parameters_tunable.F90
# hybrid run
./xmlchange RUN_TYPE=hybrid,RUN_REFCASE=b.e15.B1850G.f09_g16.pi_control.25,RUN_REFDATE=0041-01-01,GET_REFCASE=TRUE
# grid mapping file
./xmlchange ROF2OCN_RMAPNAME=cpl/gridmaps/r05/map_r05_to_gx1v6_e1000r300_151109.nc
# run the following before running
./xmlchange OCN_CHL_TYPE=diagnostic
# user_nl_clm
suplnitro = 'NONE'
use_flexiblecn=.true.
use_luna=.true.
use_fun=.true.
fsurdat = '/glade/p/cesmdata/cseg/inputdata/lnd/clm2/surfdata_map/surfdata_0.9x1.25_16pftsmodwshrub_simyr1850_c151210.nc'
paramfile = '/glade/p/cesmdata/cseg/inputdata/lnd/clm2/paramdata/clm5_params.c151217.nc'
leaf_mr_vcm = 0.015d00
use_dynroot = .false.
lower_boundary_condition = 2
use_bedrock = .true.
rooting_profile_method=1
soil_layerstruct = '20SL_8.5m'
limit_irrigation = .false.
use_init_interp = .true.
init_interp_fill_missing_with_natveg = .true.
hist_dov2xy = .true.,.false.
hist_fincl2 = 'TLAI','EFLX_LH_TOT','Vcmx25Z','TSA'
hist_mfilt = 1,1
hist_nhtfrq = 0,0
# user_nl_cpl
flux_diurnal = .true.
# user_nl_cice
ice_ic = 'b.e15.B1850G.f09_g16.pi_control.25.cice.r.0041-01-01-00000.nc'
# user_nl_cam
bnd_topo = '/glade/p/cesmdata/cseg/inputdata/atm/cam/topo/fv_0.9x1.25-gmted2010_modis-cam_fv_smooth-intermediate_ncube3000-no_anisoSGH_c151029.nc'
fincl1='dst_a1','dst_c1','dst_a3','dst_c3','dst_a1_SRF','dst_a3_SRF','dst_a1DDF','dst_a1SFWET',
'dst_c1DDF','dst_c1SFWET','dst_a3DDF','dst_a3SFWET','dst_c3DDF','dst_c3SFWET',
dust_emis_fact = 0.60D0
micro_mg_dcs = 140.D-6
# user_nl_pop
alk_riv_flux_input%filename = '/glade/p/cesm/bgcwg/forcing/BEC_gx1v6_forcing_no_ms/river_nutrients_GNEWS2000_gx1v6_no_ms_c150702.nc'
dfe_riv_flux_input%filename = '/glade/p/cesm/bgcwg/forcing/BEC_gx1v6_forcing_no_ms/river_nutrients_GNEWS2000_gx1v6_no_ms_c150702.nc'
dic_riv_flux_input%filename = '/glade/p/cesm/bgcwg/forcing/BEC_gx1v6_forcing_no_ms/river_nutrients_GNEWS2000_gx1v6_no_ms_c150702.nc'
din_riv_flux_input%filename = '/glade/p/cesm/bgcwg/forcing/BEC_gx1v6_forcing_no_ms/river_nutrients_GNEWS2000_gx1v6_no_ms_c150702.nc'
dip_riv_flux_input%filename = '/glade/p/cesm/bgcwg/forcing/BEC_gx1v6_forcing_no_ms/river_nutrients_GNEWS2000_gx1v6_no_ms_c150702.nc'
doc_riv_flux_input%filename = '/glade/p/cesm/bgcwg/forcing/BEC_gx1v6_forcing_no_ms/river_nutrients_GNEWS2000_gx1v6_no_ms_c150702.nc'
don_riv_flux_input%filename = '/glade/p/cesm/bgcwg/forcing/BEC_gx1v6_forcing_no_ms/river_nutrients_GNEWS2000_gx1v6_no_ms_c150702.nc'
dop_riv_flux_input%filename = '/glade/p/cesm/bgcwg/forcing/BEC_gx1v6_forcing_no_ms/river_nutrients_GNEWS2000_gx1v6_no_ms_c150702.nc'
dsi_riv_flux_input%filename = '/glade/p/cesm/bgcwg/forcing/BEC_gx1v6_forcing_no_ms/river_nutrients_GNEWS2000_gx1v6_no_ms_c150702.nc'
dust_flux_input%filename = '/glade/p/cesm/bgcwg/forcing/BEC_gx1v6_forcing_no_ms/dst79gnx_gx1v6_090416_no_ms_c150702.nc'
fesedflux_input%filename = '/glade/p/cesm/bgcwg/forcing/BEC_gx1v6_forcing_no_ms/fesedflux_gx1v6_etopo2v2_Nov2015_vents_no_ms_c151218.nc'
iron_flux_input%filename = '/glade/p/cesm/bgcwg/forcing/BEC_gx1v6_forcing_no_ms/solFe_scenario4_current_gx1v6_no_ms_c150702.nc'
nhy_flux_monthly_input%filename = '/glade/p/cesm/bgcwg/forcing/BEC_gx1v6_forcing_no_ms/ndep_ocn_1850_gx1v6_no_ms_c150702.nc'
nox_flux_monthly_input%filename = '/glade/p/cesm/bgcwg/forcing/BEC_gx1v6_forcing_no_ms/ndep_ocn_1850_gx1v6_no_ms_c150702.nc'
===============================================================
===============================================================
Tag name: cam5_4_34
Originator(s): goldy, eaton, fvitt
Date: 2015-12-30
One-line Summary: Add zonal_mean option for FV history output
Purpose of changes:
- Zonal mean output can save considerable space and post-processing time,
however, the workaround which has been used for FV does not work in
many WACCM runs (when there are more levels than longitudes). This
change allows for native zonal-mean output with standard metadata.
Bugs fixed (include bugzilla ID): NA
Describe any changes made to build system: NA
Describe any changes made to the namelist: NA
List any changes to the defaults for the boundary datasets: NA
Describe any substantial timing or memory changes: NA
Code reviewed by: eaton
List all subroutines eliminated: NA
List all subroutines added and what they do:
- cam_grid_is_zonal (cam_grid_support): Return .true. iff input grid is zonal
List all existing files that have been modified, and describe the changes:
M components/cam/bld/namelist_files/namelist_definition.xml
M components/cam/doc/ChangeLog
- Now 0.09% heavier!
M components/cam/src/control/cam_history.F90
- Make sure new cell_methods field property restarts
- addfld constructs a cell_methods string if the input grid is zonal
- Do not allow (endrun) zonal fields for local time processing ('L')
- Do not allow (endrun) zonal fields on satellite history file
M components/cam/src/control/cam_history_support.F90
- Add a cell_methods property to the field structure
M components/cam/src/dynamics/fv/ctem.F90
- Modify output of zonally-averaged fields to use new zonal grid and
native output methods
M components/cam/src/dynamics/fv/dyn_grid.F90
- Add new zonal grid, fv_centers_zonal, for history zonal mean output
M components/cam/src/utils/cam_grid_support.F90
- Add bounds attribute to horizontal coordinate structure, entered via a new
argument to horiz_coord_create
- Add zonal_grid attribute to grid structure.
M components/cam/test/system/nl_files/sat_hist
- Change processing of zonal fields from 'L' to 'I'
- Remove zonal field from sat_hist file (no longer valid)
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
hobart/nag: NA (but only because the missing field (UW3d from
sathist_fincl) does not trigger a baseline failure).
hobart/pgi or jaguar/pgi: NA
CAM tag used for the baseline comparison tests if different than previous
tag: NA
Summarize any changes to answers:
- The failed test outputs zonally-averaged fields (e.g., 'VTH3d') which
are now output as proper 2-D fields (they were copied plon times into
3-D fields). The actual values are unchanged.
If bitwise differences were observed, how did you show they were no worse
than roundoff? NA
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate: NA
URL for AMWG diagnostics output used to validate new climate: NA
===============================================================
===============================================================
Tag name: cam5_4_33
Originator(s): fvitt, mmills, jrichter, aksmith, joemci
Date: 18 Dec 2015
One-line Summary: WACCM updates and bug fixes
Purpose of changes:
- update gravity wave drag for WACCM to be consistent with latest CCMI
-- turn off inertial gravity waves
-- set inverse Prandtl number to 0.5
-- multiply tau by a factor of 2.0 in southern hemisphere
-- turn off land fraction scaling
-- switch order of limiter and when efficiencies are applied
- add namelist tuning parameter which effective reduces GW phase speeds
to aid in model generated QBO -- gw_qbo_hdepth_scaling
- correction to EUV heating for WACCM
- addition of CO2 cooling to space for WACCM
- correction to troposphere/stratosphere heterogeneous chemistry rates
-- minimum tropopause pressure of 300 hPa poleward of 50 degrees latitude
-- corrected double counting at the tropopause level, by calculating tropospheric
rates only below the tropopause, and not at the tropopause level
- add chemistry packages:
waccm_mozart_mam4
waccm_tsmlt_mam4
- add build-namelist use case:
volc_waccm_ma_cam5
- add compset F55W5SC
- CIME tests updates
Bugs fixed (include bugzilla ID):
- correction to EUV heating for WACCM
- addition of CO2 cooling to space for WACCM
- correction to trop/strat heterogeneous chemistry rates
-- minimum tropopause pressure of 300 hPa poleward of 50 degrees latitude
-- corrected double counting at the tropopause level, by calculating tropospheric
rates only below the tropopause, and not at the tropopause level
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton, cacraig, joemci
List all subroutines eliminated:
D components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq3d_sc
D components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq3d_sc/user_nl_clm
D components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq3d_sc/user_nl_cam
D components/cam/cime_config/testdefs/testmods_dirs/cam/outfrq3d_sc/shell_commands
- not needed and were removed -- there should be no need to set RUN_STARTDATE for waccm-sc
List all subroutines added and what they do:
A components/cam/src/chemistry/pp_waccm_mozart_mam4/mo_lu_solve.F90
A components/cam/src/chemistry/pp_waccm_mozart_mam4/mo_setrxt.F90
A components/cam/src/chemistry/pp_waccm_mozart_mam4/mo_adjrxt.F90
A components/cam/src/chemistry/pp_waccm_mozart_mam4/mo_rxt_rates_conv.F90
A components/cam/src/chemistry/pp_waccm_mozart_mam4/mo_imp_sol.F90
A components/cam/src/chemistry/pp_waccm_mozart_mam4/chem_mods.F90
A components/cam/src/chemistry/pp_waccm_mozart_mam4/mo_prod_loss.F90
A components/cam/src/chemistry/pp_waccm_mozart_mam4/mo_lin_matrix.F90
A components/cam/src/chemistry/pp_waccm_mozart_mam4/m_rxt_id.F90
A components/cam/src/chemistry/pp_waccm_mozart_mam4/mo_phtadj.F90
A components/cam/src/chemistry/pp_waccm_mozart_mam4/mo_nln_matrix.F90
A components/cam/src/chemistry/pp_waccm_mozart_mam4/mo_lu_factor.F90
A components/cam/src/chemistry/pp_waccm_mozart_mam4/mo_sim_dat.F90
A components/cam/src/chemistry/pp_waccm_mozart_mam4/mo_indprd.F90
A components/cam/src/chemistry/pp_waccm_mozart_mam4/m_spc_id.F90
A components/cam/src/chemistry/pp_waccm_mozart_mam4/chem_mech.doc
A components/cam/src/chemistry/pp_waccm_mozart_mam4/chem_mech.in
A components/cam/src/chemistry/pp_waccm_mozart_mam4
- new chem package
A components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_setrxt.F90
A components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_adjrxt.F90
A components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_rxt_rates_conv.F90
A components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_imp_sol.F90
A components/cam/src/chemistry/pp_waccm_tsmlt_mam4/chem_mods.F90
A components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_prod_loss.F90
A components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_lin_matrix.F90
A components/cam/src/chemistry/pp_waccm_tsmlt_mam4/m_rxt_id.F90
A components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_phtadj.F90
A components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_nln_matrix.F90
A components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_lu_factor.F90
A components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_sim_dat.F90
A components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_indprd.F90
A components/cam/src/chemistry/pp_waccm_tsmlt_mam4/m_spc_id.F90
A components/cam/src/chemistry/pp_waccm_tsmlt_mam4/chem_mech.doc
A components/cam/src/chemistry/pp_waccm_tsmlt_mam4/chem_mech.in
A components/cam/src/chemistry/pp_waccm_tsmlt_mam4/mo_lu_solve.F90
A components/cam/src/chemistry/pp_waccm_tsmlt_mam4
- new chem package
A components/cam/bld/namelist_files/use_cases/volc_waccm_ma_cam5.xml
- new use case
A components/cam/cime_config/testdefs/testmods_dirs/cam/sd_leapday/shell_commands
A components/cam/cime_config/testdefs/testmods_dirs/cam/sd_leapday/user_nl_cam
A components/cam/cime_config/testdefs/testmods_dirs/cam/sd_leapday
- added leap day test
List all existing files that have been modified, and describe the changes:
M components/cam/bld/build-namelist
- set default for new gw_* namelist vars
M components/cam/bld/configure
M components/cam/bld/config_files/definition.xml
- new chem options -- waccm_mozart_mam4, waccm_tsmlt_mam4
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
- defaults for new gw_* namelist vars
- set fv nsplit default to 12 for 0.9x1.25 resolution
M components/cam/bld/namelist_files/namelist_definition.xml
- new gw namelist options:
gw_limit_tau_without_eff :
Apply limiters to tau before applying the efficiency factor, rather than afterward.
Default: .false.
gw_apply_tndmax :
Apply limiter on maximum wind tendency from stress divergence in gravity wave drag scheme.
Default: .true.
gw_oro_south_fac :
Factor to multiply tau by, for orographic waves in the southern hemisphere.
Default: 1._r8
gw_prndl :
Inverse Prandtl number used in gravity wave diffusion
Default: 0.5
gw_qbo_hdepth_scaling :
Scaling factor for heating depth in gravity waves from convection. If less than 1.0
this acts as an effective reduction of the gravity wave phase speeds needed to drive
the QBO.
Default: 1.0
M components/cam/bld/namelist_files/use_cases/waccm_sc_1955-2005_cam5.xml
- include gw and cld frac tuning settings
- update emissions and other changes from Mike Mills
- remove use of IGW
M components/cam/bld/namelist_files/use_cases/sd_2007-2011_htap_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/sd_waccm5_geos5.xml
M components/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_waccm_ma.xml
M components/cam/bld/namelist_files/use_cases/sd_1975-2010_ccmi_refc1_waccm_tsmlt.xml
M components/cam/bld/namelist_files/use_cases/1850_waccm_tsmlt_cam5.xml
M components/cam/bld/namelist_files/use_cases/2007-2011_htap_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/sd_1975-2010_ccmi_refc1_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/2004-2100_ccmi_refc2_rcp45_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/1950-2010_ccmi_refc1_waccm_tsmlt.xml
M components/cam/bld/namelist_files/use_cases/2000_ccmi_refc1_waccm_tsmlt.xml
M components/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_rcp85_waccm_ma.xml
M components/cam/bld/namelist_files/use_cases/1979-2010_sd_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/sd_1975-2010_ccmi_refc1_waccm_ma.xml
M components/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_waccm_tsmlt.xml
M components/cam/bld/namelist_files/use_cases/1950-2010_ccmi_refc1_trop_strat_soa.xml
M components/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_trop_strat_soa.xml
- remove use of IGW
M components/cam/bld/namelist_files/use_cases/volc_waccm_tsmlt_megan_cam5.xml
M components/cam/bld/namelist_files/use_cases/volc_waccm_tsmlt_nomegan_cam5.xml
- add IC file for 1-degrees res
- use updated volcanic SO2 emissions
- remove use of IGW
M components/cam/bld/namelist_files/use_cases/rcp6.0_waccm_tsmlt_nomegan_cam5.xml
- remove use of IGW
- fix bug in external forcing specifier
M components/cam/bld/namelist_files/use_cases/waccm_sc_2000_cam4.xml
- added flbc_list
- remove use of IGW
M components/cam/cime_config/config_component.xml
- use waccm_sc_2000_cam5 build-namelist use case for FW5SC compset
- use waccm_sc_mam3 chem configure option for waccm-sc-cam5 compsets
- new compset F55W5SC
M components/cam/cime_config/config_compsets.xml
- new compset F55W5SC
- use GREGORIAN calendar for SDC1_* compsets
- use updated SST files that have data through 2014 for CCMI and SD compsets
- use a transient CLM use case
M components/cam/cime_config/testdefs/testlist_cam.xml
- aux_wcm tests added
- use cam/outfrq3d testmods rather than outfrq3d_sc for waccm-sc tests
- leap day test added
M components/cam/src/physics/waccm/nlte_lw.F90
M components/cam/src/physics/waccm/nlte_fomichev.F90
- added co2-to-spacing cooling term and associated diagnostics
M components/cam/src/physics/cam/gw_diffusion.F90
M components/cam/src/physics/cam/gw_common.F90
M components/cam/src/physics/cam/gw_front.F90
M components/cam/src/physics/cam/gw_convect.F90
M components/cam/src/physics/cam/gw_drag.F90
- namelist options to enable waccm GWs to be consistent with CCMI
- add namelist tuning parameter which effective reduces GW phase speeds
to aid in model generated QBO -- gw_qbo_hdepth_scaling
M components/cam/src/chemistry/modal_aero/aero_model.F90
- correction in comments regarding stratospheric wet surface area in stratosphere
M components/cam/src/chemistry/mozart/mo_jeuv.F90
- correction to EUV heating - wavelenths corrected
M components/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
M components/cam/src/chemistry/mozart/mo_usrrxt.F90
- correction to trop/strat heterogeneous chemistry rates
-- minimum tropopause pressure of 300 hPa poleward of 50 degrees latitude
-- corrected double counting at the tropopause level, by calculating tropospheric
rates only below the tropopause, and not at the tropopause level
M components/cam/src/chemistry/pp_waccm_mozart_mam3/chem_mech.in
M components/cam/src/chemistry/pp_waccm_mozart_mam3/mo_lu_solve.F90
M components/cam/src/chemistry/pp_waccm_mozart_mam3/mo_setrxt.F90
M components/cam/src/chemistry/pp_waccm_mozart_mam3/mo_adjrxt.F90
M components/cam/src/chemistry/pp_waccm_mozart_mam3/mo_rxt_rates_conv.F90
M components/cam/src/chemistry/pp_waccm_mozart_mam3/chem_mods.F90
M components/cam/src/chemistry/pp_waccm_mozart_mam3/mo_prod_loss.F90
M components/cam/src/chemistry/pp_waccm_mozart_mam3/mo_lin_matrix.F90
M components/cam/src/chemistry/pp_waccm_mozart_mam3/m_rxt_id.F90
M components/cam/src/chemistry/pp_waccm_mozart_mam3/mo_phtadj.F90
M components/cam/src/chemistry/pp_waccm_mozart_mam3/mo_nln_matrix.F90
M components/cam/src/chemistry/pp_waccm_mozart_mam3/mo_lu_factor.F90
M components/cam/src/chemistry/pp_waccm_mozart_mam3/mo_sim_dat.F90
M components/cam/src/chemistry/pp_waccm_mozart_mam3/mo_indprd.F90
M components/cam/src/chemistry/pp_waccm_mozart_mam3/m_spc_id.F90
M components/cam/src/chemistry/pp_waccm_mozart_mam3/chem_mech.doc
- updates -- consistent with waccm_mozart
- short-lived species declared as not-transported
M components/cam/test/system/tests_waccm_mpi
- changed "br" test to "er"
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
009 bl203 TBL.sh fsd1.9c4wtsmlth outfrq3s_sd+sd_1975-2010_ccmi_refc1_waccm_tsmlt 9s ...........FAIL! rc= 7 at Thu Dec 17 20:15:00 MST 2015
016 bl326 TBL.sh f1.9c4wscdm outfrq3s_2005+waccm_sc_2005-2100_cam4_rcp45 9s ...................FAIL! rc= 7 at Thu Dec 17 20:36:05 MST 2015
029 bl336 TBL.sh f4c4wmdh outfrq3s+waccm_1850_cam4 9s .........................................FAIL! rc= 7 at Thu Dec 17 20:58:05 MST 2015
035 bl365 TBL.sh f1.9c5wmclbdh outfrq3s_clubb+waccm_2000_cam5 9s ..............................FAIL! rc= 7 at Thu Dec 17 22:08:44 MST 2015
052 bl390 TBL.sh f1.9c4wmxh outfrq3s+waccmx_2000_cam4 9s ......................................FAIL! rc= 7 at Thu Dec 17 23:01:38 MST 2015
062 bl429 TBL.sh f1.9c4wmxh outfrq3s+waccmxi_2000_cam4 9s .....................................FAIL! rc= 7 at Thu Dec 17 23:23:47 MST 2015
064 bl430 TBL.sh f1.9c5mzmam3dh outfrq3s_NEUwetdep 9s .........................................FAIL! rc= 7 at Thu Dec 17 23:32:51 MST 2015
070 bl470 TBL.sh fsd1.9c5.4wtclbh outfrq3s_sd+sd_waccm5_geos5 9s ..............................FAIL! rc= 7 at Fri Dec 18 00:09:18 MST 2015
073 bl482 TBL.sh f1.9c5wscmam3dh outfrq3s_2005+waccm_sc_1955-2005_cam5 9s .....................FAIL! rc= 7 at Fri Dec 18 00:43:43 MST 2015
083 bl379 TBL.sh f1.9c4wmh outfrq24h+waccm_2005-2100_cam4_rcp45 2d ............................FAIL! rc= 7 at Fri Dec 18 01:09:47 MST 2015
hobart/nag:
028 bl314 TBL.sh f10c4wscdm outfrq3s 9s .......................................................FAIL! rc= 7 at Thu Dec 17 14:38:11 MST 2015
067 bl426 TBL.sh f10c4wmxdm outfrq3s_ionos 9s .................................................FAIL! rc= 7 at Thu Dec 17 16:54:17 MST 2015
070 bl471 TBL.sh f10c5wmam3dm outfrq3s_modalstrat 9s ..........................................FAIL! rc= 7 at Thu Dec 17 17:05:19 MST 2015
hobart/pgi or jaguar/pgi:
033 bl314 TBL.sh f10c4wscdm outfrq3s 9s .......................................................FAIL! rc= 7 at Thu Dec 17 15:18:19 MST 2015
These baseline failures are expected due to:
- update gravity wave drag for WACCM
- correction to troposphere/stratosphere heterogeneous chemistry rates
===============================================================
===============================================================
Tag name: cam5_4_32
Originator(s): jedwards, cacraig
Date: Dec 17, 2015
One-line Summary: Bug fix in restart_physics when using pio2
Purpose of changes:
- Jim Edwards discovered a bug and provided a fix in restart_physics.
This bug only occurs when using pio2 (which is not yet the CAM default)
and using the SE dycore.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: cacraig
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/src/physics/cam/restart_physics.F90
- bug fix when using pio2
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB
hobart/nag: all BFB
hobart/pgi or jaguar/pgi: all BFB
===============================================================
===============================================================
Tag name: cam5_4_31
Originator(s): pel, hannay, rneale, mmills, fvitt, cacraig
Date: Dec 15, 2015
One-line Summary: Update topo files for 1 and 2 degree FV runs
Purpose of changes:
- all FV 1 and 2 degree runs will have answer changes
-- this includes both CAM and WACCM runs
- new topo files are based on GMTED2010 and MODIS
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: okayed by rneale
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
- added new topo files for FV 1 and 2 degree runs
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
As expected, all 1 and 2 degree FV runs are answer changing due to the new topo files.
yellowstone/intel:
012 bl322 TBL.sh f1.9c5nonedh outfrq3s_lb0 9s .................................................FAIL! rc= 7 at Mon Dec 14 20:11:42 MST 2015
016 bl326 TBL.sh f1.9c4wscdm outfrq3s_2005+waccm_sc_2005-2100_cam4_rcp45 9s ...................FAIL! rc= 7 at Mon Dec 14 20:25:46 MST 2015
032 bl353 TBL.sh f1.9c5.4dh outfrq3s 9s .......................................................FAIL! rc= 7 at Mon Dec 14 20:58:38 MST 2015
035 bl365 TBL.sh f1.9c5wmclbdh outfrq3s_clubb+waccm_2000_cam5 9s ..............................FAIL! rc= 7 at Mon Dec 14 21:55:48 MST 2015
038 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Mon Dec 14 22:09:08 MST 2015
042 bl373 TBL.sh f1.9c4mozdh outfrq3s+2000_cam4_trop_chem 9s ..................................FAIL! rc= 7 at Mon Dec 14 22:24:52 MST 2015
044 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...............................................FAIL! rc= 7 at Mon Dec 14 22:27:27 MST 2015
046 bl376 TBL.sh f1.9c4mozdh+testmech outfrq3s 9s .............................................FAIL! rc= 7 at Mon Dec 14 22:33:42 MST 2015
049 bl374 TBL.sh f1.9c4fastdh outfrq3s+1850-2005_cam4_super_fast_llnl 9s ......................FAIL! rc= 7 at Mon Dec 14 22:40:34 MST 2015
052 bl390 TBL.sh f1.9c4wmxh outfrq3s+waccmx_2000_cam4 9s ......................................FAIL! rc= 7 at Mon Dec 14 22:49:25 MST 2015
056 bl397 TBL.sh f1.9c4cm outfrq3s_cosp+1850-2005_cam4 9s .....................................FAIL! rc= 7 at Mon Dec 14 22:53:13 MST 2015
059 bl421 TBL.sh f1.9c5mam4dh outfrq3s 9s .....................................................FAIL! rc= 7 at Mon Dec 14 23:03:42 MST 2015
062 bl429 TBL.sh f1.9c4wmxh outfrq3s+waccmxi_2000_cam4 9s .....................................FAIL! rc= 7 at Mon Dec 14 23:11:58 MST 2015
064 bl430 TBL.sh f1.9c5mzmam3dh outfrq3s_NEUwetdep 9s .........................................FAIL! rc= 7 at Mon Dec 14 23:21:06 MST 2015
067 bl440 TBL.sh f1.9c4soadh outfrq3s+soa_chem_megan_emis 9s ..................................FAIL! rc= 7 at Mon Dec 14 23:38:18 MST 2015
073 bl482 TBL.sh f1.9c5wscmam3dh outfrq3s_2005+waccm_sc_1955-2005_cam5 9s .....................FAIL! rc= 7 at Tue Dec 15 00:21:02 MST 2015
077 bl387 TBL.sh f1.9c4dm outfrq3s+1850-2005_cam4 9s ..........................................FAIL! rc= 7 at Tue Dec 15 00:27:54 MST 2015
080 bl388 TBL.sh f1.9c4m outfrq24h+1850_cam4 2d ...............................................FAIL! rc= 7 at Tue Dec 15 00:29:28 MST 2015
083 bl379 TBL.sh f1.9c4wmh outfrq24h+waccm_2005-2100_cam4_rcp45 2d ............................FAIL! rc= 7 at Tue Dec 15 00:47:02 MST 2015
hobart/nag: All PASS
049 bl354 TBL.sh f1.9c5.4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Mon Dec 14 16:59:16 MST 2015
055 bl368 TBL.sh f1.9c5mg2dm outfrq3s_macmic2_classnuc 9s .....................................FAIL! rc= 7 at Mon Dec 14 17:34:58 MST 2015
058 bl369 TBL.sh f1.9c5clbdm outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Mon Dec 14 17:56:04 MST 2015
hobart/pgi or jaguar/pgi:
057 bl354 TBL.sh f1.9c5.4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Mon Dec 14 18:11:08 MST 2015
===============================================================
===============================================================
Tag name: cam5_4_30
Originator(s): cacraig, andrew, katec, eaton
Date: Mon Dec 7 09:00:08 MST 2015
One-line Summary: Refactoring init sequence; remove rgrid; bug fixes
Purpose of changes:
. Continue refactoring initialization sequence
- More cleanup in runtime_opts: move call of sat_hist_readnl to
history_readnl.
- Move call to trunc from cam_init to the dyn_init calls in
the EUL and SLD dycores
- Remove call to init_masterlinkedlist from cam_init. This call was not
needed.
. Finish removing code for unsupported rgrid functionality. This is
responsible for the bulk of the changes in this tag.
. Fix threading problem in clubb when micro_do_icesupersat=.true.
. Fix in MG code for analytic radar reflectivity diagnostic calc
. Update SILHS external.
Bugs fixed (include bugzilla ID):
bugzilla 2247: fix test in analytic radar reflectivity calc
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
components/cam/src/control/pspect.F90
. moved to eul and sld dycore directories
components/cam/src/control/rgrid.F90
. removed -- unsupported functionality
components/cam/src/dynamics/eul/trunc.F90
components/cam/src/dynamics/sld/trunc.F90
. subroutine moved into
components/cam/src/dynamics/fv/comspe.F90
components/cam/src/dynamics/fv/comsta.h
. remove unused files
components/cam/src/dynamics/fv/trunc.F90
components/cam/src/dynamics/se/trunc.F90
. refactoring initialization allows removing these stubs
List all subroutines added and what they do:
components/cam/src/dynamics/eul/pspect.F90
components/cam/src/dynamics/sld/pspect.F90
. copied from control directory. These are only used by the global
spectral dycores.
List all existing files that have been modified, and describe the changes:
components/cam/SVN_EXTERNAL_DIRECTORIES
. update silhs from silhs_07_n02 to silhs_07_n04
components/cam/src/advection/slt/bandij.F90
. remove rgrid
components/cam/src/control/cam_comp.F90
. remove calls to init_masterlinkedlist() and trunc()
. remove unused arg in call to init_pio_subsystem
components/cam/src/control/cam_history.F90
. remove init_masterlinkedlist(). Not needed since nullifying the pointers
masterlinkedlist and tape is now done in the declaration statements.
. remove public attribute from namelist variables that are passed as args to the
sat_hist_readnl routine. Now sat_hist_readnl is called from history_readnl.
components/cam/src/control/cam_restart.F90
. remove unused ref to pmgrid. remove unused var tmp_rgrid
components/cam/src/control/runtime_opts.F90
. move call of sat_hist_readnl to history_readnl.
components/cam/src/dynamics/eul/comspe.F90
components/cam/src/dynamics/sld/comspe.F90
. Remove public attribute for specific variables since whole module is
public data.
components/cam/src/dynamics/eul/dp_coupling.F90
components/cam/src/dynamics/eul/dyn.F90
components/cam/src/dynamics/eul/dyn_grid.F90
components/cam/src/dynamics/eul/dynpkg.F90
components/cam/src/dynamics/eul/grcalc.F90
components/cam/src/dynamics/eul/herxin.F90
components/cam/src/dynamics/eul/inidat.F90
components/cam/src/dynamics/eul/initcom.F90
components/cam/src/dynamics/eul/linemsdyn.F90
components/cam/src/dynamics/eul/quad.F90
components/cam/src/dynamics/eul/scan2.F90
components/cam/src/dynamics/eul/scandyn.F90
components/cam/src/dynamics/eul/scanslt.F90
components/cam/src/dynamics/eul/spegrd.F90
components/cam/src/dynamics/eul/spetru.F90
components/cam/src/dynamics/eul/spmd_dyn.F90
components/cam/src/dynamics/eul/stepon.F90
components/cam/src/dynamics/sld/dp_coupling.F90
components/cam/src/dynamics/sld/dyn.F90
components/cam/src/dynamics/sld/dyn_grid.F90
components/cam/src/dynamics/sld/grcalc.F90
components/cam/src/dynamics/sld/initcom.F90
components/cam/src/dynamics/sld/quad.F90
components/cam/src/dynamics/sld/scan2.F90
components/cam/src/dynamics/sld/scandyn.F90
components/cam/src/dynamics/sld/scanslt.F90
components/cam/src/dynamics/sld/spegrd.F90
components/cam/src/dynamics/sld/spetru.F90
components/cam/src/dynamics/sld/spmd_dyn.F90
components/cam/src/dynamics/sld/stepon.F90
components/cam/src/physics/cam/phys_gmean.F90
components/cam/src/utils/cam_dom/sst_data.F90
. remove rgrid
. replace refs to nlon by plon
. replace refs to nmmax by pmmax
. replace refs to beglatpair by 1
. remove any conditional code in (.not. fullgrid) branches
components/cam/src/dynamics/eul/dyn_comp.F90
. remove unused dyndecomp_set
. add call to trunc()
. subroutine trunc as private routine in this module
components/cam/src/dynamics/eul/pmgrid.F90
components/cam/src/dynamics/fv/pmgrid.F90
components/cam/src/dynamics/se/pmgrid.F90
components/cam/src/dynamics/sld/pmgrid.F90
. remove unused dyndecomp_set
components/cam/src/dynamics/fv/dp_coupling.F90
components/cam/src/dynamics/fv/dyn_grid.F90
components/cam/src/dynamics/fv/initcom.F90
. remove unused rgrid
components/cam/src/dynamics/fv/dyn_comp.F90
components/cam/src/dynamics/se/dyn_comp.F90
. remove unused dyndecomp_set
components/cam/src/dynamics/sld/dyn_comp.F90
. remove unused dyndecomp_set
. add call to trunc()
. subroutine trunc as private routine in this module
. bug fix for addfld calls missing gridname='gauss_grid' arg
components/cam/src/dynamics/sld/inidat.F90
. bug fix -- read PS from fh_ini, not fh_topi
. remove rgrid
. replace refs to nlon by plon
components/cam/src/physics/cam/clubb_intr.F90
. Fix threading problem with micro_do_icesupersat=.true. Put init of lq2
in init method.
components/cam/src/physics/cam/micro_mg1_0.F90
components/cam/src/physics/cam/micro_mg2_0.F90
. fix test in analytic radar reflectivity calc
components/cam/src/physics/cam/radiation.F90
components/cam/src/physics/rrtmg/radiation.F90
. remove unused ref to pspect
components/cam/src/utils/cam_pio_utils.F90
. remove unused dummy arg in routine init_pio_subsystem
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS
hobart/nag: All PASS
hobart/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_4_29
Originator(s): fvitt, jedwards, cacraig, aliceb, goldy
Date: November 24, 2015
One-line Summary: cam_pio_get_var bug fix, mira compiler bug fix, chem_preproc bug fix, fix typo in config_component.xml
Purpose of changes:
- cam_pio_get_var bug fix (supplied by Jim, approved by Steve)
- mira compiler bug fix (changed and tested by Cheryl and Alice, approved by Francis)
- chem_preproc bug fix (supplied by Francis)
- correct typo in PTS_MODE group identifier (supplied by Jim)
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: see above
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam
M components/cam/SVN_EXTERNAL_DIRECTORIES
- chem_preproc bug fix : was unable to compile with DEBUG flag enabled - a new
chem_proc external fixes this
M components/cam/cime_config/config_component.xml
- correct typo in PTS_MODE group identifier
M components/cam/src/chemistry/modal_aero/modal_aero_data.F90
- mira compiler bug fix: mira did not like '??????' - change it to 'UNDEF '
M components/cam/src/utils/cam_pio_utils.F90
- time dimension needed to be changed from a hardwired 3 to ndims in two places
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB
hobart/nag: all BFB
hobart/pgi or jaguar/pgi: all BFB
===============================================================
===============================================================
Tag name: cam5_4_28
Originator(s): goldy
Date: 2015-11-15
One-line Summary: Fix IO bug and PIO2 incompatiblity
Purpose of changes: Fix IO bug and PIO2 incompatiblity
Bugs fixed (include bugzilla ID): NA
Describe any changes made to build system: NA
Describe any changes made to the namelist: NA
List any changes to the defaults for the boundary datasets: NA
Describe any substantial timing or memory changes: NA
Code reviewed by: goldy, jedwards
List all subroutines eliminated: NA
List all subroutines added and what they do: NA
List all existing files that have been modified, and describe the changes:
M components/cam/doc/ChangeLog
- Updated according to ancient rules of the CAMbal
M components/cam/src/utils/cam_pio_utils.F90
- Fixed but in cam_pio_get_var for initial files with multiple timeslices
M components/cam/src/physics/cam/physics_buffer.F90.in
- Output buffers with a singleton mdim as 2-D variable for PIO2 compatibility
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All pass
hobart/nag: All pass
hobart/pgi or jaguar/pgi: All pass (hobart)
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_4_27
Originator(s): fvitt, tilmes
Date: 30 Oct 2015
One-line Summary: Enable MAM to have multiple SOA species per mode
Purpose of changes:
- refactor how data objects in modal aerosol model are set to provide
flexibility needed for multiple SOA species (or bins) per mode
- support VBS (Volatility Basis-Set) aerosol chemistry schemes
- add trop_strat_mam4_vbs chemistry package and compset FSTRATVBSL45
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
D components/cam/src/chemistry/modal_aero/modal_aero_initialize_data.F90
- initialization routines moved to modal_aero_data module
List all subroutines added and what they do:
A components/cam/src/chemistry/pp_trop_strat_mam4_vbs/m_rxt_id.F90
A components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_phtadj.F90
A components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_nln_matrix.F90
A components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_lu_factor.F90
A components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_sim_dat.F90
A components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_indprd.F90
A components/cam/src/chemistry/pp_trop_strat_mam4_vbs/m_spc_id.F90
A components/cam/src/chemistry/pp_trop_strat_mam4_vbs/chem_mech.doc
A components/cam/src/chemistry/pp_trop_strat_mam4_vbs/chem_mech.in
A components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_lu_solve.F90
A components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_setrxt.F90
A components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_adjrxt.F90
A components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_rxt_rates_conv.F90
A components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_imp_sol.F90
A components/cam/src/chemistry/pp_trop_strat_mam4_vbs/chem_mods.F90
A components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_prod_loss.F90
A components/cam/src/chemistry/pp_trop_strat_mam4_vbs/mo_lin_matrix.F90
A components/cam/src/chemistry/pp_trop_strat_mam4_vbs
- new VBS chemistry package
A components/cam/test/system/config_files/f10c5mam4vbsdm
A components/cam/test/system/config_files/f1.9c5mam4vbsdh
- tests added for trop_strat_mam4_vbs
List all existing files that have been modified, and describe the changes:
M components/cam/src/chemistry/modal_aero/aero_model.F90
M components/cam/src/chemistry/modal_aero/modal_aero_rename.F90
M components/cam/src/chemistry/modal_aero/modal_aero_coag.F90
M components/cam/src/chemistry/modal_aero/modal_aero_convproc.F90
M components/cam/src/chemistry/modal_aero/modal_aero_data.F90
M components/cam/src/chemistry/modal_aero/modal_aero_gasaerexch.F90
M components/cam/src/chemistry/modal_aero/seasalt_model.F90
M components/cam/src/chemistry/modal_aero/dust_model.F90
M components/cam/src/chemistry/modal_aero/modal_aero_newnuc.F90
M components/cam/src/chemistry/utils/modal_aero_calcsize.F90
- refactored modal aerosol data structures to provide flexibility for
multiple SOA species per mode
M components/cam/src/chemistry/utils/aerodep_flx.F90
M components/cam/src/chemistry/utils/modal_aero_deposition.F90
- generalized/refactored to sum over all soa species
NOTE: these changes result in round-off level changes in deposition fluxes
of carbon aerosol exported to surface models which may feedback on climate
M components/cam/bld/namelist_files/namelist_definition.xml
- trop_strat_mam4_vbs added to list of acceptable cam_chempkg names
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
- defaults added for trop_strat_mam4_vbs chemistry
M components/cam/bld/namelist_files/master_aer_drydep_list.xml
M components/cam/bld/namelist_files/master_aer_wetdep_list.xml
M components/cam/bld/namelist_files/master_gas_drydep_list.xml
M components/cam/bld/namelist_files/master_gas_wetdep_list.xml
- gas-phase and aerosol species added to deposition lists
M components/cam/bld/build-namelist
M components/cam/bld/perl5lib/Build/ChemNamelist.pm
- use utility routine to construct modal aerosol objects for mode_defs
based on species in chemistry mechanism file
M components/cam/bld/configure
- new chem option added: trop_strat_mam4_vbs
M components/cam/cime_config/config_component.xml
M components/cam/cime_config/config_compsets.xml
M components/cam/cime_config/testdefs/testlist_cam.xml
- added FSTRATVBSL45 compset and prealpha test
M components/cam/src/chemistry/mozart/mo_photo.F90
- removed alias of dust_nbins to appease PGI compiler
M components/cam/src/chemistry/mozart/mo_drydep.F90
M components/cam/src/chemistry/mozart/mo_sethet.F90
M components/cam/src/chemistry/mozart/mo_neu_wetdep.F90
M components/cam/src/chemistry/mozart/mo_usrrxt.F90
- mods to handle new SOA and "TS1" species
M components/cam/test/system/input_tests_master
- tests added for trop_strat_mam4_vbs
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
009 bl203 TBL.sh fsd1.9c4wtsmlth outfrq3s_sd+sd_1975-2010_ccmi_refc1_waccm_tsmlt 9s ...........FAIL! rc= 7 at Thu Oct 29 16:30:34 MDT 2015
042 bl373 TBL.sh f1.9c4mozdh outfrq3s+2000_cam4_trop_chem 9s ..................................FAIL! rc= 7 at Thu Oct 29 18:52:23 MDT 2015
046 bl376 TBL.sh f1.9c4mozdh+testmech outfrq3s 9s .............................................FAIL! rc= 7 at Thu Oct 29 19:01:07 MDT 2015
064 bl430 TBL.sh f1.9c5mzmam3dh outfrq3s_NEUwetdep 9s .........................................FAIL! rc= 7 at Thu Oct 29 19:47:50 MDT 2015
067 bl440 TBL.sh f1.9c4soadh outfrq3s+soa_chem_megan_emis 9s ..................................FAIL! rc= 7 at Thu Oct 29 20:04:55 MDT 2015
070 bl470 TBL.sh fsd1.9c5.4wtclbh outfrq3s_sd+sd_waccm5_geos5 9s ..............................FAIL! rc= 7 at Thu Oct 29 20:25:31 MDT 2015
- expected failures due to change in deposition of terpene species in tropospheric chemistry
hobart/nag: All pass
hobart/pgi:
057 bl354 TBL.sh f1.9c5.4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Thu Oct 29 15:49:26 MDT 2015
060 bl420 TBL.sh f4c5mam7dm outfrq3s 9s .......................................................FAIL! rc= 7 at Thu Oct 29 16:02:09 MDT 2015
063 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Thu Oct 29 16:14:28 MDT 2015
- these failures are caused by round-off level differences in aerosol-gas exchange routine due to refactoring
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_4_26
Originator(s): hannay, cacraig, jshollen
Date: October 29, 2015
One-line Summary: Add 1850_CAM55 use case
Purpose of changes:
- Add a use_case for 1850_CAM55. For now, this is an exact copy of 1850_CAM5
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
A components/cam/bld/namelist_files/use_cases/1850_cam55.xml
List all existing files that have been modified, and describe the changes:
M components/cam/cime_config/config_component.xml
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: none run as this was only adding a new use_case for CESM runs
hobart/nag: none run as this was only adding a new use_case for CESM runs
hobart/pgi or jaguar/pgi: none run as this was only adding a new use_case for CESM runs
- Ran a test using a replica of this in alpha02f with Jay and verified by hand that
the proper use case was selected
===============================================================
===============================================================
Tag name: cam5_4_25
Originator(s): Chris Kerr, John Dennis, cacraig
Date: 0ct 28, 2015
One-line Summary: Optimized MG2
Purpose of changes:
- optimizations which according to John Dennis' testing yields 27% speedup in MG2
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton, cacraig
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/src/physics/cam/micro_mg2_0.F90
M components/cam/src/physics/cam/micro_mg_utils.F90
- optimizations
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB
hobart/nag: all BFB
hobart/pgi or jaguar/pgi: all BFB
===============================================================
===============================================================
Tag name: cam5_4_24
Originator(s): eaton
Date: Thu Oct 22 14:58:07 MDT 2015
One-line Summary: Cleanup namelist reading; build system update.
Purpose of changes:
. cleanup namelist reading code in runtime_opts.F90. Move all remaining
variables out of the generic cam_inparm group and make the appropriate
modules responsible for reading their own namelist data
. add '-phys cam5.5' option. It's equivalent to '-phys cam5.4 -clubb_sgs'
. fix CAM standalone build of timing code so that dependencies are
automatically generated.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system:
. add '-phys cam5.5' option
. fix CAM standalone build of timing code so that dependencies are
automatically generated.
Describe any changes made to the namelist:
. move namelist variables out of cam_inparm an into module specific
namelist groups
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: cam developers
List all subroutines eliminated:
components/cam/src/physics/cam/scyc.F90
components/cam/src/physics/cam/tsinti.F90
. no longer used
List all subroutines added and what they do:
components/cam/src/physics/cam/dadadj_cam.F90
. cam interface layer added for dry adiabatic adjustment parameterization
List all existing files that have been modified, and describe the changes:
components/cam/bld/Makefile.in
. modify build of timing code so that only files in the source directories
that should not be built are specified rather than specifying files that
should be built. The dependency generation can now be done by mkDepends
rather than be hardcoded in the Makefile.
components/cam/bld/configure
. add '-phys cam5.5' option
. add gptl and timing source directories to Filepath file.
components/cam/bld/mkSrcfiles
. add option to allow excluding a list of files from the Srcfiles file.
components/cam/bld/namelist_files/namelist_definition.xml
. move variables out of cam_inparm group and into the groups read by the
local namelist readers.
components/cam/src/control/cam_comp.F90
. Extract dtime from the eclock object passed by the driver. Then can use
it to initialize the time manager and don't need to put the read_namelist
call in front of the time manager init.
. move call to read_namelist after call to timemgr_init
components/cam/src/control/cam_control_mod.F90
. remove unused variables: rair, stebol, snwedp, latice, tmelt, latvap
. move other constants to modules that should be responsible for them
components/cam/src/control/cam_history.F90
. history_readnl
- access dtime from time manager rather than passing it as an arg. Can
do this because time manager is initialized before reading namelists.
components/cam/src/control/cam_initfiles.F90
. add pertlim to this module
. replace mpishorthand by std mpi code
components/cam/src/control/cam_restart.F90
. replace the default_opts, setopts, and printopts methods by
cam_restart_readnl.
components/cam/src/control/camsrfexch.F90
. access rair from physconst rather than cam_control_mod
components/cam/src/control/filenames.F90
. module data for absems_data moved to cam4 radiation module
components/cam/src/control/rgrid.F90
. hardcode fullgrid=.true. -- will be removing this entirely in the near
future.
components/cam/src/control/runtime_opts.F90
. remove dtime from cam_inparm namelist -- now gotten from coupler
. remove dtime as actual arg in history_readnl call
. remove absems_data, iradae, iradlw, iradsw, irad_always, spectralflux,
use_rad_dt_cosz from cam_inparm
. replace the radiation_defaultopts, radiation_setopts, radiation_printopts
with radiation_readnl.
. remove absems_data, iradae, iradlw, iradsw, irad_always, spectralflux,
use_rad_dt_cosz
. remove indirect
. remove print_step_cost -- this functionality has moved to the driver
. remove pertlim from cam_inparms namelist group
. remove nlvdry and add call to dadadj_readnl
. remove cam_branch_file and replace calls to restart default_opts,
setopts, and printopts methods by single call to cam_restart_readnl
. remove direct access to readtrace via constituents module and instead
call the new cnst_readnl
. remove rayk0, raykrange, raytau0 from cam_inparms and add call to
rayleigh_friction_readnl
. move pbuf_global_allocate to the pbuf_nl namelist
*** the pbuf_{defaultopts,setopts} are not present, which means this option
hasn't been getting tested.
- add pbuf_readnl call. leave it commented out for now. Need to fix
the pbuf_global_allocate=.false. option
. replace scam defaultopts/setopts routines by scam_readnl
. move print_energy_errors to the check_energy namelist
. replace check_energy_{defaultopts,setopts} by check_energy_readnl
components/cam/src/control/scamMod.F90
. replace scam defaultopts/setopts routines by scam_readnl
components/cam/src/dynamics/eul/inidat.F90
components/cam/src/dynamics/fv/inidat.F90
components/cam/src/dynamics/se/inidat.F90
components/cam/src/dynamics/sld/inidat.F90
. access pertlim from cam_initfiles rather than cam_control_mod
components/cam/src/physics/cam/check_energy.F90
. replace check_energy_{defaultopts,setopts} by check_energy_readnl
components/cam/src/physics/cam/cldwat.F90
. remove unused ref to nlvdry
components/cam/src/physics/cam/constituents.F90
. add cnst_readnl to read constituents_nl namelist group
components/cam/src/physics/cam/dadadj.F90
. put this subroutine into a module
. change subroutine dadadj to dadadj_calc -- move error handling for the
routine to interface layer
. new subroutine dadadj_initial
components/cam/src/physics/cam/phys_grid.F90
. replace phys_grid defaultopts and setopts by readnl
components/cam/src/physics/cam/physics_buffer.F90.in
. replace pbuf_{defaultopts,setopts} by pbuf_readnl
components/cam/src/physics/cam/physpkg.F90
. remove call to tsinti from routine phys_init
. replace call to dadadj by new call to cam interface routine dadadj_tend
components/cam/src/physics/cam/radae.F90
. pass absems_data through radae_init args. remove access via module
filenames.
components/cam/src/physics/cam/radiation.F90
. remove routine radiation_get -- not used
. add absems_data namelist variable to module data and pass it to
radae_init.
. remove spectralflux option which is not valid in camrt
. replace the radiation_defaultopts, radiation_setopts, radiation_printopts
with radiation_readnl.
components/cam/src/physics/cam/rayleigh_friction.F90
. add namelist reader
components/cam/src/physics/cam/rk_stratiform.F90
. remove unused addfld calls for EFFLIQ, EFFLIQ_IND, EFFICE
components/cam/src/physics/rrtmg/radiation.F90
. remove routine radiation_get -- not used
. remove iradae option which is not valid in rrtmg
. replace the radiation_defaultopts, radiation_setopts, radiation_printopts
with radiation_readnl.
components/cam/src/utils/time_manager.F90
. remove timemgr_set_step_size. Move setting dtime to the timemgr_init
method.
components/cam/src/utils/units.F90
. remove unused association of endrun
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS
hobart/nag: All PASS
hobart/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag: cam5_4_22 (no testing done for cam5_4_23)
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_4_23
Originator(s): cacraig
Date: October 14, 2015
One-line Summary: Add CAM55 definition for CESM compsets
Purpose of changes: Add CAM55 definition for CESM compsets
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes: Improves timing
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/cime_config/config_component.xml
- add the CAM55 configuration definition for CESM compset use
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: Changes not testable in CAM regression testing - not run
hobart/nag: Changes not testable in CAM regression testing - not run
hobart/pgi or jaguar/pgi: Changes not testable in CAM regression testing - not run
Jim Edwards ran a CESM test using these changes and deemed them correct
===============================================================
===============================================================
Tag name: cam5_4_22
Originator(s): goldy
Date: 2015-10-08
One-line Summary: Fix performance issue with regional output
Purpose of changes: Fix performance issue with regional output
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes: Improves timing
Code reviewed by: goldy
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/doc/ChangeLog
- Can't remember what I did to this file
M components/cam/src/utils/cam_pio_utils.F90
. Added optional argument for unlimited dim ID in cam_pio_var_info
M components/cam/src/control/cam_history_support.F90
. Do not include unliminted dimension in PIO decomp calls
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB
hobart/nag: all BFB
hobart/pgi or jaguar/pgi: all BFB
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_4_21
Originator(s): cacraig, fvitt
Date: October 7, 2015
One-line Summary: Update to CESM alpha02a externals, fix failing WACCM ionosphere CESM test, remove goldbach from testing
Purpose of changes:
- Bring the CAM externals up to the current CESM tag (1_5_alpha02a)
- CESM test ERP_Ln9.f19_f19.FWXI.yellowstone_intel.cam-outfrq9s is now working properly
- All references to goldbach are removed from CAM testing due to its decommissioning. This included
removing the special nag53 compilation flag
- CESM now requires the submit script to be run which issues a bsub command. CAM regression testing
now changes the value of RUN_WITH_SUBMIT to allow the run script to be used directly (and not issue
a bsub). This circumvents calling short-term archiving which CAM regression testing does not use.
Bugs fixed (include bugzilla ID): CESM ERP_Ln9.f19_f19.FWXI.yellowstone_intel.cam-outfrq9s is now working
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: cacraig
List all subroutines eliminated:
D components/cam/test/system/tests_posttag_goldbach
D components/cam/test/system/tests_pretag_goldbach_nag
D components/cam/test/system/tests_pretag_goldbach_pgi
- remove goldbach testing
List all subroutines added and what they do:
A components/cam/test/system/nl_files/outfrq3s_ionos
- namelist for waccm ionosphere test
List all existing files that have been modified, and describe the changes:
M .
M SVN_EXTERNAL_DIRECTORIES
- update externals to CESM1_5_alpha02a
M components/cam/bld/Makefile.in
M components/cam/bld/config_files/definition.xml
M components/cam/bld/configure
- remove special nag53 compilation flag (was needed for goldbach)
M components/cam/src/physics/waccmx/ionosphere.F90
- Fixed reproducibility issue -- changing number of MPI tasks produced different results
M components/cam/test/system/CAM_runcmnd.sh
- removed goldbach testing now that this computer is decommissioned
M components/cam/test/system/TCB_ccsm.sh
M components/cam/test/system/TER_ccsm.sh
M components/cam/test/system/TSM_ccsm.sh
- add xmlchange for flag RUN_WITH_SUBMIT, to allow CAM regression testing to use the run
script instead of submit (which issues an independent bsub command)
M components/cam/test/system/input_tests_master
- add additional waccm tests including one for ionosphere
M components/cam/test/system/test_driver.sh
- removed goldbach testing
M components/cam/test/system/tests_pretag_hobart_nag
- changed waccm test to one which includes ionosphere
M components/cam/test/system/tests_pretag_yellowstone
- add waccm test for ionosphere
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB except for new test:
062 bl429 TBL.sh f1.9c4wmxh outfrq3s+waccmxi_2000_cam4 9s .....................................FAIL! rc= 7 at Tue Oct 6 23:14:56 MDT 2015
hobart/nag: all BFB except for new test:
067 bl426 TBL.sh f10c4wmxdm outfrq3s_ionos 9s .................................................FAIL! rc= 7 at Tue Oct 6 18:17:23 MDT 2015
hobart/pgi or jaguar/pgi: all BFB
===============================================================
===============================================================
Tag name: cam5_4_20
Originator(s): sacks, cacraig
Date: October 6, 2015
One-line Summary: Passing CAM's topographic height to the coupler
Purpose of changes:
- Pass the topographic height from CAM to the coupler so that it is available
for other components to use
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
M .
M SVN_EXTERNAL_DIRECTORIES
- temporarily for this tag, use a cime branch off of cime3.0.7 which includes
just the cime changes needed to support the additional topographic field
to the coupler
M components/cam/src/control/camsrfexch.F90
M components/cam/src/cpl/atm_import_export.F90
M components/cam/src/cpl/cam_cpl_indices.F90
- add the topographic height to the atmosphere/coupler exchange structures
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB
hobart/nag: all BFB
hobart/pgi or jaguar/pgi: all BFB
===============================================================
===============================================================
Tag name: cam5_4_19
Originator(s): fvitt
Date: 1 Oct 2015
One-line Summary: Bug fixes for SD CLUBB and WACCM with CLUBB
Purpose of changes:
- fix restart bug which occurs when CLUBB is used in specified dynamics mode
- fix bug in CLUBB when used with WACCM which set the constituent mixing ratios
to zero above the active CLUBB region
- correct emissoins of number density tracers in WACCM5 build-namelist use case
volc_waccm_tsmlt_nomegan_cam5
- add compsets (and corresponding cime tests):
FSDW5 -- WACCM5 with TSMLT chemistry, cam5.4 physics, and specified dynamics
FWTC5L45CCMIR1 -- free running WACCM5 with TSMLT chemistry and cam5.4 physics
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/src/cpl/atm_comp_esmf.F90
M components/cam/src/cpl/atm_comp_mct.F90
M components/cam/src/cpl/atm_import_export.F90
- add optional argument to atm_import which is used during the initialization
phase of a restarted run so that cam_in fields wsx, wsy, shf, and cflx are not
overwritten by fields read in from the rs restart file. These fields are
set by the data from the *cam.r.* restart file. When nudging to specified
dynamics the nudged surface fields presist in the *cam.r.* restart so that
CLUBB inputs are bit-for-bit the same on the first step of a restarted run
as a continuous run.
M components/cam/src/physics/cam/clubb_intr.F90
- enforce zero mixing ratio tendencies above the top_lev level so the clubb
does not effect tracer above its active region when used in WACCM
M components/cam/bld/namelist_files/use_cases/volc_waccm_tsmlt_nomegan_cam5.xml
- correction to num_a1 and num_a2 emissions
M components/cam/cime_config/config_component.xml
M components/cam/cime_config/config_compsets.xml
M components/cam/cime_config/testdefs/testlist_cam.xml
- add compsets (and corresponding cime tests):
FSDW5 -- WACCM5 with TSMLT chemistry, cam5.4 physics, and specified dynamics
FWTC5L45CCMIR1 -- free running WACCM5 with TSMLT chemistry and cam5.4 physics
M components/cam/src/physics/cam/restart_physics.F90
- cam_in fields wsx, wsy, shf added to *cam.r.* restart file
M components/cam/test/system/input_tests_master
M components/cam/test/system/nl_files/off1.9x2.5
M components/cam/test/system/tests_pretag_hobart_nag
M components/cam/test/system/tests_pretag_yellowstone
- add SD CAM/WACCM with CLUBB tests
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
035 bl365 TBL.sh f1.9c5wmclbdh outfrq3s_clubb+waccm_2000_cam5 9s ..............................FAIL! rc= 7 at Thu Oct 1 09:33:27 MDT 2015
- new test -- no baseline to compare to
hobart/nag:
052 bl356 TBL.sh fsd1.9c5clbdm off1.9x2.5 9s ..................................................FAIL! rc= 7 at Thu Oct 1 09:59:15 MDT 2015
- new test -- no baseline to compare to
hobart/pgi: All pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_4_18
Originator(s): mvertens, jedwards, cacraig
Date: Sept 29, 2015
One-line Summary: Bring in CESM cime refactor
Purpose of changes:
- Bring in the CESM cime refactor modifications
- For CAM regression testing, have gmake sync its compilation output
- use the externals from cesm1_5_alpha01d
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
D components/cam/bld/cam.buildlib
D components/cam/bld/cam.buildnml
D components/cam/bld/user_nl_cam
D components/cam/cimetest
D components/cam/cimetest/testlist_cam.xml
D components/cam/cimetest/testmods_dirs
D components/cam/cimetest/testmods_dirs/cam
D components/cam/cimetest/testmods_dirs/cam/cam4_port
D components/cam/cimetest/testmods_dirs/cam/cam4_port/user_nl_cam
D components/cam/cimetest/testmods_dirs/cam/cam5_port
D components/cam/cimetest/testmods_dirs/cam/cam5_port/user_nl_cam
D components/cam/cimetest/testmods_dirs/cam/cam5_port_f45
D components/cam/cimetest/testmods_dirs/cam/cam5_port_f45/user_nl_cam
D components/cam/cimetest/testmods_dirs/cam/cam5_port_ne30
D components/cam/cimetest/testmods_dirs/cam/cam5_port_ne30/user_nl_cam
D components/cam/cimetest/testmods_dirs/cam/chemproc
D components/cam/cimetest/testmods_dirs/cam/chemproc/shell_commands
D components/cam/cimetest/testmods_dirs/cam/chemproc/user_nl_cam
D components/cam/cimetest/testmods_dirs/cam/cosp
D components/cam/cimetest/testmods_dirs/cam/cosp/shell_commands
D components/cam/cimetest/testmods_dirs/cam/fire_emis
D components/cam/cimetest/testmods_dirs/cam/fire_emis/user_nl_cam
D components/cam/cimetest/testmods_dirs/cam/fire_emis/user_nl_clm
D components/cam/cimetest/testmods_dirs/cam/mam4_fire_emis
D components/cam/cimetest/testmods_dirs/cam/mam4_fire_emis/user_nl_cam
D components/cam/cimetest/testmods_dirs/cam/mam4_fire_emis/user_nl_clm
D components/cam/cimetest/testmods_dirs/cam/outfrq1d
D components/cam/cimetest/testmods_dirs/cam/outfrq1d/shell_commands
D components/cam/cimetest/testmods_dirs/cam/outfrq1d/user_nl_cam
D components/cam/cimetest/testmods_dirs/cam/outfrq1d/user_nl_clm
D components/cam/cimetest/testmods_dirs/cam/outfrq3d
D components/cam/cimetest/testmods_dirs/cam/outfrq3d/shell_commands
D components/cam/cimetest/testmods_dirs/cam/outfrq3d/user_nl_cam
D components/cam/cimetest/testmods_dirs/cam/outfrq3d/user_nl_clm
D components/cam/cimetest/testmods_dirs/cam/outfrq3d_sc
D components/cam/cimetest/testmods_dirs/cam/outfrq3d_sc/shell_commands
D components/cam/cimetest/testmods_dirs/cam/outfrq3d_sc/user_nl_cam
D components/cam/cimetest/testmods_dirs/cam/outfrq3d_sc/user_nl_clm
D components/cam/cimetest/testmods_dirs/cam/outfrq9s
D components/cam/cimetest/testmods_dirs/cam/outfrq9s/shell_commands
D components/cam/cimetest/testmods_dirs/cam/outfrq9s/user_nl_cam
D components/cam/cimetest/testmods_dirs/cam/outfrq9s/user_nl_clm
D components/cam/cimetest/testmods_dirs/cam/outfrq9s_sat_hist
D components/cam/cimetest/testmods_dirs/cam/outfrq9s_sat_hist/shell_commands
D components/cam/cimetest/testmods_dirs/cam/outfrq9s_sat_hist/user_nl_cam
D components/cam/cimetest/testmods_dirs/cam/outfrq9s_sat_hist/user_nl_clm
D components/cam/cimetest/testmods_dirs/cam/reduced_hist3s
D components/cam/cimetest/testmods_dirs/cam/reduced_hist3s/user_nl_cam
D components/cam/cimetest/testmods_dirs/cam/tssoa_fire_emis
D components/cam/cimetest/testmods_dirs/cam/tssoa_fire_emis/user_nl_cam
- files moved to components/cam/cime_config
List all subroutines added and what they do:
A + components/cam/cime_config
- directory replaces cimetest
List all existing files that have been modified, and describe the changes:
M .
M SVN_EXTERNAL_DIRECTORIES
- use the externals from cesm1_5_alpha01d
M components/cam/bld/build-namelist
- scriptsroot for CICE build-namelist is now cimeroot
M components/cam/test/system/TCB_ccsm.sh
- cesm_setup is now case_setup
M components/cam/test/system/TSM_ccsm.sh
- allow for checking of final atm.log file to occur in either /run or /logs
M components/cam/test/system/test_driver.sh
- update intel compiler on yellowstone to 15.0.3 (to match CESM compiler setting)
- force gmake to synchronize output so that NAG compiler warnings/errors are not scrambled
M components/cam/cime_config/testdefs/testlist_cam.xml
- add gl10 to F1850C5L4BGC f19_f19 grid_name (per Jim Edwards)
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB
hobart/nag: all BFB
hobart/pgi or jaguar/pgi: all BFB
===============================================================
===============================================================
Tag name: cam5_4_17
Originator(s): cchen, cacraig
Date: September 25, 2015
One-line Summary: Bug fix for hetfrz_classnuc and update CLUBB external
Purpose of changes:
- With the original value of pdf_n_theta set to 101 the dust activation
fraction between -15 and 0 C could be overestimated. This problem was
eliminated by increasing pdf_n_theta to 301. To reduce the expense of
computing the dust activation fraction the integral is only evaluated
where dim_theta is non-zero. This was determined to be between
dim_theta index values of 53 through 113. These loop bounds are
hardcoded in the variables i1 and i2.
- PGI compiler had issue with variable in CLUBB external declared threadprivate before defined - fixed
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam
M components/cam/SVN_EXTERNAL_DIRECTORIES
- updated CLUBB library external
M components/cam/src/physics/cam/hetfrz_classnuc.F90
- modifications to fix hetfrz bug
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
** All files which use hetfrz_classnuc are expected to have different results.
** Default for cam5.4 is for hetfrz_classnuc to be used
yellowstone/intel:
032 bl353 TBL.sh f1.9c5.4dh outfrq3s 9s .......................................................FAIL! rc= 7 at Thu Sep 24 22:55:42 MDT 2015
067 bl474 TBL.sh f1.9c5.4wtclbh outfrq3s+1850_waccm_tsmlt_cam5 9s .............................FAIL! rc= 7 at Fri Sep 25 01:42:46 MDT 2015
hobart/nag:
049 bl354 TBL.sh f1.9c5.4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Thu Sep 24 18:34:20 MDT 2015
052 bl368 TBL.sh f1.9c5mg2dm outfrq3s_macmic2_classnuc 9s .....................................FAIL! rc= 7 at Thu Sep 24 18:50:52 MDT 2015
hobart/pgi or jaguar/pgi:
057 bl354 TBL.sh f1.9c5.4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Thu Sep 24 19:14:32 MDT 2015
URL for AMWG diagnostics output used to validate new climate:
http://www.cgd.ucar.edu/cms/cchen/cam5_4_12_test-cam5_4_12_control
There are no significant differences between the two simulations
===============================================================
===============================================================
Tag name: cam5_4_16
Originator(s): goldy
Date: 2015-09-16
One-line Summary: Bug fixes related to new I/O infrastructure
Purpose of changes:
- Fix issue with history coordinate writes which causes PIO2 to crash
- Fix wrong-grid error in specified dynamics history
Bugs fixed (include bugzilla ID):
- Bugzilla 2201, Output of history fields
Describe any changes made to build system: NA
Describe any changes made to the namelist: NA
List any changes to the defaults for the boundary datasets: NA
Describe any substantial timing or memory changes: NA
Code reviewed by: goldy, jedwards
List all subroutines eliminated: NA
List all subroutines added and what they do: NA
List all existing files that have been modified, and describe the changes:
M components/cam/doc/ChangeLog
- Added text around this line (30 above / 45 below)
M components/cam/src/dynamics/eul/dyn_grid.F90
. Fix write statements and comments
M components/cam/src/dynamics/eul/initcom.F90
. Initialize lat and lon values for restart
M components/cam/src/dynamics/fv/metdata.F90
. Fix bugzilla 2201
M components/cam/src/physics/cam/phys_grid.F90
. Make sure coordinate dofs are one-to-one for physgrid
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: None
hobart/nag: None
hobart/pgi or jaguar/pgi: None
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate): NA
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_4_15
Originator(s): fvitt
Date: 16 Sep 2015
One-line Summary: WACCM5.5 updates
Purpose of changes:
- chemistry mechanism changes for waccm_tsmlt_mam3 including 'not-transported'
species
- correction to surface area densities of stratospheric aerosols
- bug fix in background ionization for waccm/waccmx
- changes to namelist use_case files for waccm5.5 compsets
- change FSDS phys buffer field scope from 'physpkg' to 'global' to solve
restart problem in chemistry
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
D components/cam/test/system/config_files/f1.9c5wtmam3dh
D components/cam/test/system/config_files/f1.9c5wtmam3h
D components/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_waccm_tsmlt_cam5.xml
D components/cam/bld/namelist_files/use_cases/sd_waccm_tsmlt_geos5.xml
D components/cam/bld/namelist_files/use_cases/1850-2005_waccm_tsmlt_cam5.xml
D components/cam/bld/namelist_files/use_cases/sd_waccm_tslt_geos5.xml
List all subroutines added and what they do:
A components/cam/test/system/config_files/f4c5.4wtmam3dm
A components/cam/test/system/config_files/f1.9c5.4wtmam3h
A components/cam/test/system/config_files/f1.9c5.4wtmam3dh
A components/cam/test/system/config_files/fsd1.9c5wtmam3h
- regression tests
A components/cam/bld/namelist_files/use_cases/rcp6.0_waccm_tsmlt_nomegan_cam5.xml
A components/cam/bld/namelist_files/use_cases/volc_waccm_tsmlt_nomegan_cam5.xml
A components/cam/bld/namelist_files/use_cases/volc_waccm_tsmlt_megan_cam5.xml
- new use cases for waccm5.5
List all existing files that have been modified, and describe the changes:
M components/cam/src/physics/cam/radiation.F90
M components/cam/src/physics/rrtmg/radiation.F90
- change FSDS phys buffer field from 'physpkg' to 'global' to solve
restart problem in chemistry
M components/cam/src/chemistry/modal_aero/aero_model.F90
- calculate wet surface area densities of stratospheric aerosols
M components/cam/src/chemistry/mozart/photo_bkgrnd.F90
- corrections to waccm/waccmx back ground ionization
M components/cam/src/physics/waccm/qbo.F90
- restrict log messages to master process
M components/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_rxt_rates_conv.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam3/chem_mods.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_prod_loss.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_lin_matrix.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_nln_matrix.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_lu_factor.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_sim_dat.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_indprd.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam3/m_spc_id.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam3/chem_mech.doc
M components/cam/src/chemistry/pp_waccm_tsmlt_mam3/chem_mech.in
M components/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_lu_solve.F90
- updates to chemistry mechanism including declaring some species as
"not-transported"
M components/cam/cimetest/testlist_cam.xml
- added cime tests for waccm5.5
M components/cam/test/system/tests_waccm_mpi
M components/cam/test/system/input_tests_master
M components/cam/test/system/tests_waccm_hybrid
- changes to regression testing for waccm5.5
M components/cam/bld/namelist_files/use_cases/1950-2010_ccmi_refc1_waccm_tsmlt.xml
M components/cam/bld/namelist_files/use_cases/sd_1975-2010_ccmi_refc1_waccm_ma.xml
M components/cam/bld/namelist_files/use_cases/sd_waccm_geos5.xml
M components/cam/bld/namelist_files/use_cases/sd_waccm5_geos5.xml
M components/cam/bld/namelist_files/use_cases/1950-2010_ccmi_refc1_waccm_ma.xml
M components/cam/bld/namelist_files/use_cases/sd_1975-2010_ccmi_refc1_waccm_tsmlt.xml
- changes to use cases for waccm5.5
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
009 bl203 TBL.sh fsd1.9c4wtsmlth outfrq3s_sd+sd_1975-2010_ccmi_refc1_waccm_tsmlt 9s ...........FAIL! rc= 7 at Tue Sep 15 09:52:22 MDT 2015
029 bl336 TBL.sh f4c4wmdh outfrq3s+waccm_1850_cam4 9s .........................................FAIL! rc= 7 at Tue Sep 15 09:55:56 MDT 2015
035 bl365 TBL.sh f1.9c5wmclbdh outfrq3s_clubb+waccm_2000_cam5 9s ..............................FAIL! rc= 7 at Tue Sep 15 09:57:34 MDT 2015
052 bl390 TBL.sh f1.9c4wmxh outfrq3s+waccmx_2000_cam4 9s ......................................FAIL! rc= 7 at Tue Sep 15 09:59:18 MDT 2015
067 bl474 TBL.sh f1.9c5.4wtclbh outfrq3s+1850_waccm_tsmlt_cam5 9s .............................FAIL! rc= 7 at Tue Sep 15 09:59:19 MDT 2015
080 bl379 TBL.sh f1.9c4wmh outfrq24h+waccm_2005-2100_cam4_rcp45 2d ............................FAIL! rc= 7 at Tue Sep 15 10:00:20 MDT 2015
- expected baseline failures due to :
-- corrections to back ground ionization
-- correction to surface area densities of stratospheric aerosols (WACCM5)
-- chemistry mechanism changes for waccm_tsmlt_mam3 including not-transported species (WACCM5)
hobart/nag:
064 bl424 TBL.sh f10c4wmxdm outfrq3s 9s .......................................................FAIL! rc= 7 at Tue Sep 15 16:04:30 MDT 2015
067 bl471 TBL.sh f10c5wmam3dm outfrq3s_modalstrat 9s ..........................................FAIL! rc= 7 at Tue Sep 15 16:30:17 MDT 2015
- expected baseline failures due to :
-- corrections to back ground ionization
-- correction to surface area densities of stratospheric aerosols (WACCM5)
-- chemistry mechanism changes for waccm_tsmlt_mam3 including not-transported species (WACCM5)
hobart/pgi: all pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_4_14
Originator(s): bogensch, cacraig
Date: Sept 14, 2015
One-line Summary: CLUBB total energy adjustment modification
Purpose of changes:
- Change of the vertical extent of the CLUBB total energy adjustment to limit
the altitudes to where CLUBB is creating a significant tendency. CLUBB and CAM
have different total energy definitions, and CLUBB is adjusted to match CAM.
Previously energy adjustment was over the whole depth of the model, this
limits it mostly to the troposphere.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/src/physics/cam/clubb_intr.F90
- changes to CLUBB total energy adjustment
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB except:
035 bl365 TBL.sh f1.9c5wmclbdh outfrq3s_clubb+waccm_2000_cam5 9s ..............................FAIL! rc= 7 at Fri Sep 11 22:07:37 MD
T 2015
038 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Fri Sep 11 22:20:44 MD
T 2015
hobart/nag: all BFB except:
055 bl369 TBL.sh f1.9c5clbdm outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Fri Sep 11 18:41:44 MDT 2015
hobart/pgi or jaguar/pgi: all BFB
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: all CLUBB
- what platforms/compilers: all
- nature of change : larger than roundoff but same climate
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
https://svn-ccsm-models.cgd.ucar.edu/cam1/branch_tags/camclubb_version4_tags/camclubb_version4_n03_cam5_4_03
URL for AMWG diagnostics output used to validate new climate:
http://webext.cgd.ucar.edu/FCLIMO/clubb_energy/cam55_FC5_f09f09_014/atm/cam55_FC5_f09f09_014-cam55_FC5_f09f09_001/
===============================================================
===============================================================
Tag name: cam5_4_13
Originator(s): jedwards, santos, gettelman, cacraig
Date: Sept 11, 2015
One-line Summary: incorporate CESM branch fixes and misc minor updates
Purpose of changes:
- incorporate mods to support pio2
- fixes problem with ESMF interface
- Temporary workaround fix for failing CESM test with SE dycore on mira
- update MG2 header
- add missing clubb/waccm test to yellowstone
- Added perf_mod dependencies to allow NAG6 to compile on hobart for CAM standalone
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M .
M SVN_EXTERNAL_DIRECTORIES
- update to cime tag required for the pio changes
M components/cam
M components/cam/SVN_EXTERNAL_DIRECTORIES
- Modification to allow CESM mira test to pass
- Detailed diagnostics calculation and log output skipped in prim_printstate due to use
of uninitialized variables. Output will be corrected and restored in a later release.
M components/cam/bld/Makefile.in
- NAG6.0 was having issues with perf_mod not being compiled before it was required
Added four routines which were compiling when NAG aborted which have dependencies on perf_mod
M components/cam/src/cpl/atm_comp_esmf.F90
- changes for compatibility with pio2
- fix for ESMF interface
M components/cam/src/dynamics/fv/restart_dynamics.F90
M components/cam/src/physics/cam/restart_physics.F90
M components/cam/src/utils/cam_grid_support.F90
M components/cam/src/utils/cam_pio_utils.F90
- changes for compatibility with pio2
M components/cam/src/physics/cam/micro_mg2_0.F90
- updated header to better document MG2
M components/cam/test/system/tests_pretag_yellowstone
- added missing CLUBB/WACCM test
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB except for newly added bl365 as it had no comparison file
hobart/nag: all BFB
hobart/pgi or jaguar/pgi: all BFB
CAM tag used for the baseline comparison tests if different than previous
tag:
===============================================================
===============================================================
Tag name: cam5_4_12
Originator(s): santos
Date: 2015/09/02
One-line Summary: Changes to adapt new CAM physics for WACCM5.5
Purpose of changes:
- Allow WACCM to run with CLUBB, which requires the vertical diffusion solver
to run in order to perform molecular diffusion, as well as adding a lid to
CLUBB that matches the one on most of the CAM5 and CAM5.4 cloud physics.
(Currently, this lid is at 1 hPa.)
- Some minor progress has been made toward getting WACCM, CAM5.4 physics, and
UNICON to run together. This should not be considered functional yet.
- The turbulent mountain stress parameterization (TMS) now has its own CAM
interface module and namelist.
- Upper boundary conditions are now set in a somewhat more consistent way.
- The eddy_diff module is now "mostly" portable (very minor CAM dependencies).
This is accomplished in part through introducing a new CAM interface module,
eddy_diff_cam.
- The "comsrf" module, which was originally the COMmon block for the SuRFace,
has finally been removed. Its variables are either removed entirely (dead
code), or added to pbuf.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
- Variables do_tms, tms_orocnst, and tms_z0fac have been moved to the new TMS
namelist. Variables specific to the UW eddy diffusion have been moved to the
eddy_diff namelist.
- The eddy_diff variables are only added to the namelist if the diag_TKE (UW)
scheme is actually enabled.
- WACCM5 use_cases no longer add "QPERT" to any fincl, since it is only
available when the diag_TKE moist turbulence scheme is running.
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
D components/cam/src/physics/cam/comsrf.F90
- This was a common block that had been turned into a module without
much improvement in the design. According to its comments:
"Currently this is a hodge-podge of 2D arrays without a lot of thought
or design. We are under the process of removing this completely and
moving the relevent arrays to the modules that actually use the data."
Now this is done.
List all subroutines added and what they do:
A components/cam/src/physics/cam/eddy_diff_cam.F90
- Move compute_eddy_diff, and much of the data and initialization logic
it requires, to this new module, which is now the CAM interface to the
eddy_diff module.
- diag_TKE-specific setup, including reading the namelist, is now pushed
down into this module.
A components/cam/src/physics/cam/trb_mtn_stress_cam.F90
- Add CAM interface for the TMS code, which has its own namelist read
and history outputs.
- Outputs used by the vertical diffusion are placed into pbuf rather
than being passed in and out as arguments to the "tend" routine.
List all existing files that have been modified, and describe the changes:
M components/cam/bld/build-namelist
M components/cam/bld/namelist_files/namelist_definition.xml
M components/cam/src/physics/cam/phys_control.F90
M components/cam/src/utils/physconst.F90
- Moved TMS and eddy_diff variables to their respective namelists.
- Only add eddy_diff variables if the PBL scheme is UW (i.e. the
eddy_scheme is diag_TKE).
M components/cam/bld/namelist_files/use_cases/waccm_2000_cam5.xml
M components/cam/bld/namelist_files/use_cases/waccm_sc_2000_cam5.xml
- Remove "QPERT" from WACCM5 fincl settings.
M components/cam/src/chemistry/mozart/chemistry.F90
M components/cam/src/chemistry/pp_none/chemistry.F90
M components/cam/src/control/camsrfexch.F90
M components/cam/src/control/startup_initialconds.F90
M components/cam/src/dynamics/eul/iop.F90
M components/cam/src/physics/cam/cam_diagnostics.F90
M components/cam/src/physics/cam/cldwat.F90
M components/cam/src/physics/cam/convect_shallow.F90
M components/cam/src/physics/cam/radiation.F90
M components/cam/src/physics/cam/rk_stratiform.F90
M components/cam/src/physics/cam/unicon_cam.F90
M components/cam/src/physics/rrtmg/radiation.F90
M components/cam/src/unit_drivers/rad/unit_driver.F90
- For variables previously kept in comsrf, which were passed directly
into many routines from physpkg, instead have each scheme retrieve any
needed variables from pbuf.
- Unused comsrf variables (and use statements and copies of those
variables) are simply removed.
- Where possible, reduce the argument list of top-level routines so that
the arguments are just a few CAM objects (state, pbuf, cam_in,
cam_out) rather than including lists of r8 arrays.
M components/cam/src/chemistry/mozart/upper_bc.F90
M components/cam/src/physics/cam/molec_diff.F90
M components/cam/src/physics/cam/ref_pres.F90
M components/cam/src/physics/waccmx/majorsp_diffusion.F90
- Remove the variable "ntop_molec", since setting it to a value other
than 1 would not actually work, and would not be useful in any current
configuration anyway.
- Consistently use ubc_get_vals to set the top-of-model boundary
conditions.
- The molec_diff module no longer handles boundary conditions, except
for the data used specifically to calculate molecular diffusion. The
ubc calls are moved to vertical_diffusion.
- molec_diff is becoming "more" portable, but still largely depends on
physconst arrays for operations performed in WACCM-X.
M components/cam/src/control/history_defaults.F90
- SGH/SGH30 from the topography file are no longer available as
outputs. SGH30 was not even output properly, and there is not much
point in adding these to history files, since they are simply exact
copies of data in the topography file, and do not change from timestep
to timestep.
M components/cam/src/dynamics/se/dyn_comp.F90
- Remove clearing of topography variables for aquaplanet runs, since it
is redundant with a block in physpkg that already does this for all
dycores.
M components/cam/src/physics/cam/clubb_intr.F90
- Remove clubb_surface and the TMS calculations from CLUBB, in favor of
allowing the existing CAM code in vertical_diffusion to handle these.
- Use the new CLUBB API module (which is intended to eventually serve as
a stable interface?) instead of clubb_core and other CLUBB internal
routines, where possible.
- Put a lid on CLUBB, so that it only is used up to trop_cloud_top_lev.
The arrays with the smaller dimension (nlev) are the same arrays that
are "upside down" to match CLUBB's assumption that levels are numbered
from the ground upward.
- SGH30 is now a pbuf field, not an argument passed from physpkg.
- Fix some minor issues, e.g. by correcting the tests for NaN in
stats_end_timestep_clubb.
M components/cam/src/physics/cam/convect_deep.F90
- Only add a heating output for gravity waves if ZM is the deep
convection scheme, since otherwise this output must come from the
currently active unified convection scheme (either UNICON, or CLUBB
with clubb_do_deep enabled).
M components/cam/src/physics/cam/diffusion_solver.F90
- Separate the setting of upper boundary conditions and molecular
diffusion data from compute_vdiff.
- Avoid recreating the pressure coordinate object for each call of
compute_vdiff, having it passed in instead.
M components/cam/src/physics/cam/eddy_diff.F90
M components/cam/src/physics/cam/hb_diff.F90
M components/cam/src/physics/cam/pbl_utils.F90
- Convert eddy_diff into mostly-portable code. The use of wv_saturation
creates a dependency on CAM, but this could easily be fixed by using
the portable water vapor saturation module. The only reason that this
was not done was because it would change answers. (The CAM-specific
routine generates a lookup table for mixed-phase calculations, and so
gives different answers.)
- Move compute_eddy_diff, associated module data, and the non-portable
parts of init_eddy_diff to the new CAM interface module.
- austausch_atm was duplicated between the eddy_diff and hb_diff
modules, so it has been moved to the pbl_utils module, which depends
only on shr_kind_mod.
- eddy_diff returns warning and error strings instead of directly
writing to iulog or calling endrun.
M components/cam/src/physics/cam/gw_drag.F90
- Minor changes related to the above; SGH is now in pbuf, it is not a
history output, and ntop_molec has been removed.
M components/cam/src/physics/cam/micro_mg1_5.F90
M components/cam/src/physics/cam/micro_mg2_0.F90
M components/cam/src/physics/cam/micro_mg_cam.F90
M components/cam/src/physics/cam/micro_mg_utils.F90
- All MG versions now output am_evp_st, the precipitation evaporation
fractional area, which is used by UNICON.
- QME is now output at all levels. It is zero above the tropospheric
cloud physics lid, trop_cloud_top_lev.
M components/cam/src/physics/cam/physpkg.F90
- Fields from the topography file are now added to pbuf instead of
residing in comsrf.
- physpkg no longer passes the fields that resided in comsrf to
individual schemes. Instead it only passes pbuf, so that those fields
can be queried by each scheme as needed.
- The vertical_diffusion module is now called even when CLUBB is
enabled. Processes that have been replaced by CLUBB should be skipped
within the vertical_diffusion module, while those which have not been
replaced should run as usual. clubb_surface is therefore redundant and
has been removed.
M components/cam/src/physics/cam/restart_physics.F90
- Because various comsrf fields have been moved to pbuf with global
scope, they no longer need to be explicitly written or read during
restart.
M components/cam/src/physics/cam/vertical_diffusion.F90
- Changes to conform to many of the above.
- vertical_diffusion now calls molecular diffusion and upper
boundary condition calculations that were being called out of
compute_vdiff.
- A significant amount of PBL-specific code is now controlled by
the "do_pbl_diags" flag, which is false in CLUBB runs.
M components/cam/test/system/config_files/f1.9c5wmclbdh
M components/cam/test/system/input_tests_master
- Add WACCM-CLUBB test.
- Add non-threaded, 10-degree CAM5 DEBUG test (useful as a very
fast NAG DEBUG test, but not present in any test list).
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
035 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Wed Sep 2 11:45:42 MDT 2015
hobart/nag:
055 bl369 TBL.sh f1.9c5clbdm outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Tue Sep 1 18:11:38 MDT 2015
hobart/pgi or jaguar/pgi:
All PASS.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
CLUBB
- what platforms/compilers:
All
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
Larger than roundoff, same climate. Pete Bogenshutz says that results are
similar, but dust levels may be mildly improved. CLUBB tuning is ongoing.
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_4_11
Originator(s): goldy
Date: 2015-08-28
One-line Summary: Fix bugs found in cesm1_4_alpha07c testing
Purpose of changes:
. Fix bug found in CESM ERP_D_Ln9.f19_f19.FSDWSF.yellowstone_pgi.cam-outfrq9s
. Restore 'gw' attribute for all fv history files
Bugs fixed (include bugzilla ID):
. bug found in CESM ERP_D_Ln9.f19_f19.FSDWSF.yellowstone_pgi.cam-outfrq9s
Describe any changes made to build system: NA
Describe any changes made to the namelist: NA
List any changes to the defaults for the boundary datasets: NA
Describe any substantial timing or memory changes: NA
Code reviewed by: goldy
List all subroutines eliminated: NA
List all subroutines added and what they do:
- cam_grid_attribute_copy: Copy an attribute from one grid to another
- physgrid_copy_attributes_d: Inform the physics grids of which grid attributes to copy
List all existing files that have been modified, and describe the changes:
M components/cam/doc/ChangeLog
- Added a bunch of stuff on top of existing document
M components/cam/src/dynamics/eul/dyn_grid.F90
M components/cam/src/dynamics/fv/dyn_grid.F90
M components/cam/src/dynamics/se/dyn_grid.F90
M components/cam/src/dynamics/sld/dyn_grid.F90
- Defined physgrid_copy_attributes_d for phys_grid_init
M components/cam/src/physics/cam/phys_grid.F90
- Call physgrid_copy_attributes_d and copy indicated attributes
M components/cam/src/utils/cam_grid_support.F90
- Defined cam_grid_attribute_copy
M components/cam/src/utils/cam_pio_utils.F90
- Fixed bug in cam_pio_get_var_2d_r8 and cam_pio_get_var_3d_r8
which was tripped by reading a variable with a time dimension
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All Pass
hobart/nag: All Pass
hobart/pgi or jaguar/pgi: All Pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_4_10
Originator(s): fvitt
Date: 21 Aug 2015
One-line Summary: Enable CAM-Chem to use CLM wild fire emissions
Purpose of changes:
. Enable CAM-Chem to use wild fire emissions produced by CLM4.5
. Let fc_type configure option accept 'cray' as a valid value
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
Passing of wild fire emissions from CLM to CAM is engaged by drv_flds_in
namelist options:
. fire_emis_specifier (array of strings) -- Each array element specifies
how CAM-Chem constituents are mapped to basic smoke compounds in
the fire emissions factors table (fire_emis_factors_file). Each
chemistry constituent name (left of '=' sign) is mapped to one or more
smoke compound (separated by + sign if more than one), which can be
proceeded by a multiplication factor (separated by '*').
Example:
fire_emis_specifier = 'bc_a1 = BC','pom_a1 = 1.4*OC','SO2 = SO2'
. fire_emis_factors_file (string) -- Input file that contains the table
of basic compounds that make up the smoke from the CLM fires. This is
used in CLM module FireEmisFactorsMod.
. fire_emis_elevated (logical) -- If true then CAM-Chem treats the fire
emission sources as 3-D vertically distributed forcings for the
corresponding chemical tracers.
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton, cacraig, santos, goldy, jedwards
List all subroutines eliminated:
List all subroutines added and what they do:
A components/cam/cimetest/testmods_dirs/cam/tssoa_fire_emis/user_nl_cam
A components/cam/cimetest/testmods_dirs/cam/tssoa_fire_emis
A components/cam/cimetest/testmods_dirs/cam/mam4_fire_emis/user_nl_clm
A components/cam/cimetest/testmods_dirs/cam/mam4_fire_emis/user_nl_cam
A components/cam/cimetest/testmods_dirs/cam/mam4_fire_emis
A components/cam/cimetest/testmods_dirs/cam/fire_emis/user_nl_cam
A components/cam/cimetest/testmods_dirs/cam/fire_emis/user_nl_clm
A components/cam/cimetest/testmods_dirs/cam/fire_emis
- added aux_fire system tests
A components/cam/src/chemistry/mozart/fire_emissions.F90
- manages mapping of CLM generated wild fire emissions to chemical constituents
List all existing files that have been modified, and describe the changes:
M components/cam/bld/config_files/definition.xml
- will accept 'cray' for fc_type
M components/cam/cimetest/testlist_cam.xml
- added aux_fire system tests
M components/cam/bld/namelist_files/namelist_definition.xml
- added namelist options for CLM produced wild fire emissions
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
- added default emissions files that do not include forest fire and
grass fire emissions
M components/cam/bld/build-namelist
- set default emissions to files that do not include forest fire and
grass fire emissions when using CLM produced wild fire emissions
M components/cam/src/cpl/atm_import_export.F90
- set cam_in%fire* arrays to fluxes from coupler
M components/cam/src/cpl/cam_cpl_indices.F90
- set indices for fire emissions
M components/cam/src/control/camsrfexch.F90
- allocates the cam_in%fire* arrays for fire emissions
M components/cam/src/physics/cam/rad_constituents.F90
- added utility function for getting the number density name
corresponding to a species name
M components/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
- pass cam_in%fire* pointers to fire_emissions_vrt for vertically distributed
fire emission forcings
M components/cam/src/chemistry/mozart/chemistry.F90
- invoke fire_emissions_init and fire_emissions_srf
- pass the cam_in%fire* pointers to chemdr for vertically distributed fire
emission forcings
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All pass
hobart/nag: All pass
hobart/pgi: All pass
CAM tag used for the baseline comparison tests if different than previous
tag: cam5_4_07
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_4_09
Originator(s): cacraig
Date: August 17, 2015
One-line Summary: Removed CESM prealpha test for FMOZ on mira
Purpose of changes:
- Currently the chemistry preprocessor fails to build on mira. This test
is being removed as it is a low priority platform for chemistry and the
primary chemistry software developer does not have access to this machine.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/cimetest/testlist_cam.xml
- removed FMOZ test from mira - this configuration is still tested
on several other platforms and passes
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: not tested
hobart/nag: not tested
hobart/pgi or jaguar/pgi: not tested
The prealpha test suite is run by CESM and is not part of the CAM standalone testing
===============================================================
===============================================================
Tag name: cam5_4_08
Originator(s): eaton
Date: Thu Aug 13 13:54:58 MDT 2015
One-line Summary: fc_type bug fix
Purpose of changes:
. recent change to have fc_type accept ibm rather than xlf was incomplete.
Add a quick fix.
Bugs fixed (include bugzilla ID): see above
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
components/cam/bld/config_files/definition.xml
. for fc_type change valid value xlf to ibm
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: not done
hobart/nag: not done
hobart/pgi or jaguar/pgi: not done
Did quick interactive check that configure does not choke when given the
arg "-fc_type ibm"
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB. No code change that is tested by
our current test platforms.
===============================================================
===============================================================
Tag name: cam5_4_07
Originator(s): bogensch, cacraig, goldy, fvitt
Date: August 13, 2015
One-line Summary: Update CLUBB tunings, NAG6.0 compiler fixes, bug fixes for serial builds, update CIME tests
Purpose of changes:
- CLUBB tunings for cam5.5 are incorporated onto the CAM trunk
- NAG6.0 uncovers non-compliant code which needed to be fixed
- cam5_4_06 had bugs when built serially for FV and SE dynamics. These bugs have been fixed
- Updates for CIME tests
- fix PORT test on SE grid
- fix FTSC4L40CCMIR1SD test
- add chemistry preprocessor tests
- add aux_chem and aux_port test types
- add pleiades tests
- CAM configure now traps if attempting to use NAG with threading turned on
Bugs fixed (include bugzilla ID): bugzilla 2198: FV dycore will not compile in serial mode
Describe any changes made to build system:
Describe any changes made to the namelist:
- Added two new namelist parameters:
zmconv_momcu - Convective momentum transport parameter (upward)
zmconv_momcd - Convective mementum transport parameter (downward)
- defaults for a number of CLUBB namelist variables have been tuned
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
A components/cam/cimetest/testmods_dirs/cam/reduced_hist3s/user_nl_cam
A components/cam/cimetest/testmods_dirs/cam/reduced_hist3s
- added for hobart 10x15 FWTC4L40CCMIR1 test
A components/cam/cimetest/testmods_dirs/cam/cam5_port_ne30/user_nl_cam
A components/cam/cimetest/testmods_dirs/cam/cam5_port_ne30
- added for testing PORT on SE grid
A components/cam/cimetest/testmods_dirs/cam/cam5_port_f45
A components/cam/cimetest/testmods_dirs/cam/cam5_port_f45/user_nl_cam
- added for testing port on hobart/nag
A components/cam/cimetest/testmods_dirs/cam/chemproc/user_nl_cam
A components/cam/cimetest/testmods_dirs/cam/chemproc/shell_commands
A components/cam/cimetest/testmods_dirs/cam/chemproc
- added for chemistry preprocessor tests
List all existing files that have been modified, and describe the changes:
M .
M components/cam
M components/cam/SVN_EXTERNAL_DIRECTORIES
- updated CLUBB external - contains NAG6.0 corrections and tuning changes
M components/cam/bld/configure
- trap for NAG compilations trying to run threaded as NAG does not support threading
M components/cam/bld/build-namelist
M components/cam/bld/namelist_files/namelist_definition.xml
- added tuning parameters: zmconv_momcu and zmconv_momcd
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
- various tunings for CLUBB
M components/cam/bld/namelist_files/use_cases/sd_1975-2010_ccmi_refc1_trop_strat_soa.xml
- corrected the date on the met_data file
MM components/cam/cimetest/testlist_cam.xml
- added chemistry preprocessor tests
- added "aux_chem" and "aux_port" tests
- added prebeta tests for pleiades
- reduced resolution of nag PORT test on hobart
- reduced resolution and history output of the FWTC4L40CCMIR1 test on hobart
- corrections for the ne30 port test
- removed the mime-type property -- allows svn diff to function normally on
local working copy without "--force" option
M components/cam/src/dynamics/fv/dyn_grid.F90
M components/cam/src/dynamics/se/dyn_grid.F90
M components/cam/src/dynamics/se/spmd_dyn.F90
- fixes for serial builds on CAM
M components/cam/src/physics/cam/clubb_intr.F90
- fix for NAG6.0 compilation error
- added outfld calls for tautmsx and tautmsy
M components/cam/src/physics/cam/zm_conv.F90
M components/cam/src/physics/cam/zm_conv_intr.F90
- momcu and momcd are now namelist parameters
- removed variable "small" which was set but never used in momtran
M components/cam/test/system/input_tests_master
M components/cam/test/system/tests_pretag_hobart_nag
- added NAG test for CLUBB without threading
M components/cam/test/system/test_driver.sh
- LAPACK_LIBDIR needs to be specified on hobart
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
035 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Wed Aug 12 21:00:49 MDT 2015
- CLUBB tunings change answers for CLUBB
hobart/nag:
055 bl369 TBL.sh f1.9c5clbdm outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Thu Aug 13 09:33:01 MDT 2015
- CLUBB tunings change answers for CLUBB
hobart/pgi or jaguar/pgi: all BFB
===============================================================
===============================================================
Tag name: cam5_4_06
Originator(s): goldy
Date: August 5, 2015
One-line Summary: New grid and history infrastructure
Purpose of changes:
- Modularize CAM grid information
- Move dycore specific information from CAM history to individual dycores
- Prepare for separate physics grid by cleaning interface between physics and dynamics
- Modularize grid start-up to allow for reconfiguration of CAM start-up and restart procedures
Bugs fixed (include bugzilla ID): 1863, 1553
Describe any changes made to build system: NA
Describe any changes made to the namelist:
- Consolidated history items into new cam_history_nl namelist
- Changed names of interpolated-output namelist variables (interp ==>
interpolate) and made them settable on a per-file basis
- Moved interpolated-output namelist variables from SE share namelist to
cam_history_nl
List any changes to the defaults for the boundary datasets: NA
Describe any substantial timing or memory changes:
Code reviewed by: eaton, cacraig, santos, jedwards
List all subroutines eliminated:
- That information is classified. Trust me, they were bad subroutines
- Almost everything in cam_pio_utils.F90 (it's a bloodbath in there)
- get_block_ldof_d (don't need specialized dof functions anymore)
- get_gcol_lat (from SE dycore)
- get_gcol_lon (from SE dycore)
List all subroutines added and what they do:
dyn_grid.F90
- define_cam_grids: in each dycore to define CAM grids
cam_pio_utils.F90
- cam_pio_closefile: interface to pio_closefile (consistency and possible
future file close on endrun)
- cam_pio_fileexists: checks to see if a file exists (logical function)
- cam_pio_def_dim: convenience function for defining NetCDF dimensions
- cam_pio_def_var: convenience function for defining NetCDF variables
- cam_pio_handle_error: prints a custom error message if error status is passed
- cam_pio_get_var: Interface to pio_get_var which replaces 'GLOBAL' infld calls
- cam_pio_var_info: Retrieve variable properties (moved from ncdio_atm and
improved)
- cam_pio_find_var: Find a variable descriptor in an open NetCDF file
- cam_pio_check_var: Make sure var exists and retrieve properties
- cam_grid_register: Create a new CAM grid
- cam_grid_attribute_register: Add an attribute to a new CAM grid
- cam_grid_write_attr: Define the dimension and coordinate attributes for
the horizontal history coordinates for a grid as well at that
grid's attributes
- cam_grid_write_var: Write the values of the dimension and coordinate
attributes for the horizontal history coordinates for a grid as
well at that grid's attributes
- cam_grid_read_dist_array: Read a distributed array defined on a CAM grid
- cam_grid_write_dist_array: Write a distributed array defined on a CAM grid
- cam_grid_dimensions: Return the dimensions of the grid For lon/lat
grids, this is (nlon, nlat) For unstructured grids, this is
(ncols, 1)
- cam_grid_num_grids: Number of currently defined grids
- cam_grid_check: Return .true. iff grid ID exists
- cam_grid_id: Grid ID (decomp) or -1 if error
- cam_grid_get_local_size: Return the size of a local array for a grid
- cam_grid_get_file_dimids: Return the NetCDF IDs for a grid's dimensions
- cam_grid_get_decomp: Given some array information, find or compute a PIO
decomposition
- cam_grid_get_gcid: Return the global index for each grid element
- cam_grid_get_array_bounds: Return lower and upper bounds for local
arrays defined on a CAM grid
- cam_grid_get_coord_names: Return the names of a grid's horizontal coordinates
- cam_grid_get_dim_names: Return the name(s) of a grid's dimensions (may
be only one dimension for unstructured grids)
- cam_grid_has_blocksize: Return .true. iff a grid's arrays are contained
in differently sized blocks (e.g., physics grid ncol)
- cam_grid_get_block_count: Return the size of a block for a given grid ID
- cam_grid_get_latvals: Return a pointer to a grid's local latitude values
- cam_grid_get_lonvals: Return a pointer to a grid's local longitude values
- cam_grid_is_unstructured: Return .true. iff grid is unstructured
- cam_grid_is_block_indexed: Return .true. iff grid is block indexed
(e.g., physics grid)
- cam_grid_compute_patch: Calculate a grid and filemap for a region. Grid
patches are currently used for regional history output
List all existing files that have been modified, and describe the changes:
M components/cam/SVN_EXTERNAL_DIRECTORIES
- Updated chem_proc to new tag which incorporates addfld
interface change
- Updated HOMME external to new tag which incorporates new
interpolated output interfaces
M components/cam/bld/build-namelist
- Moved interpolated output variables from SE/share namelist to new
cam_history_nl
- Renamed interpolated output variables to be more CAM like
- Added checks for incompatible history options
M components/cam/bld/namelist_files/namelist_definition.xml
- Organized new cam_history_nl namelist
- Changed interpolated output variable names as follows:
interpolate_analysis ==> interpolate_output
interp_nlat ==> interpolate_nlat
interp_nlon ==> interpolate_nlon
interp_type ==> interpolate_type
interp_gridtype ==> interpolate_gridtype
M components/cam/doc/ChangeLog
- Added a bunch of stuff on top of existing document
M components/cam/src/chemistry/aerosol/mo_setsox.F90
M components/cam/src/chemistry/bulk_aero/aero_model.F90
M components/cam/src/chemistry/bulk_aero/mo_aerosols.F90
M components/cam/src/chemistry/bulk_aero/mo_setsoa.F90
M components/cam/src/chemistry/bulk_aero/seasalt_model.F90
M components/cam/src/chemistry/modal_aero/aero_model.F90
M components/cam/src/chemistry/modal_aero/modal_aero_coag.F90
M components/cam/src/chemistry/modal_aero/modal_aero_convproc.F90
M components/cam/src/chemistry/modal_aero/modal_aero_gasaerexch.F90
M components/cam/src/chemistry/modal_aero/modal_aero_newnuc.F90
M components/cam/src/chemistry/modal_aero/sox_cldaero_mod.F90
M components/cam/src/chemistry/mozart/cfc11star.F90
M components/cam/src/chemistry/mozart/chemistry.F90
M components/cam/src/chemistry/mozart/exbdrift.F90
M components/cam/src/chemistry/mozart/gcr_ionization.F90
M components/cam/src/chemistry/mozart/lin_strat_chem.F90
M components/cam/src/chemistry/mozart/linoz_data.F90
M components/cam/src/chemistry/mozart/mo_aero_settling.F90
M components/cam/src/chemistry/mozart/mo_airglow.F90
M components/cam/src/chemistry/mozart/mo_aurora.F90
M components/cam/src/chemistry/mozart/mo_chm_diags.F90
M components/cam/src/chemistry/mozart/mo_cph.F90
M components/cam/src/chemistry/mozart/mo_drydep.F90
M components/cam/src/chemistry/mozart/mo_exp_sol.F90
M components/cam/src/chemistry/mozart/mo_extfrc.F90
M components/cam/src/chemistry/mozart/mo_fstrat.F90
M components/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
M components/cam/src/chemistry/mozart/mo_ghg_chem.F90
M components/cam/src/chemistry/mozart/mo_lightning.F90
M components/cam/src/chemistry/mozart/mo_neu_wetdep.F90
M components/cam/src/chemistry/mozart/mo_photo.F90
M components/cam/src/chemistry/mozart/mo_sad.F90
M components/cam/src/chemistry/mozart/mo_setext.F90
M components/cam/src/chemistry/mozart/mo_setinv.F90
M components/cam/src/chemistry/mozart/mo_sulf.F90
M components/cam/src/chemistry/mozart/mo_tgcm_ubc.F90
M components/cam/src/chemistry/mozart/mo_waccm_hrates.F90
M components/cam/src/chemistry/mozart/noy_ubc.F90
M components/cam/src/chemistry/mozart/rate_diags.F90
M components/cam/src/chemistry/mozart/short_lived_species.F90
M components/cam/src/chemistry/mozart/tracer_cnst.F90
M components/cam/src/chemistry/mozart/tracer_srcs.F90
M components/cam/src/chemistry/pp_none/mo_imp_sol.F90
M components/cam/src/chemistry/pp_super_fast_llnl/mo_imp_sol.F90
M components/cam/src/chemistry/pp_super_fast_llnl/mo_sim_dat.F90
M components/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_imp_sol.F90
M components/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_sim_dat.F90
M components/cam/src/chemistry/pp_trop_bam/mo_imp_sol.F90
M components/cam/src/chemistry/pp_trop_bam/mo_sim_dat.F90
M components/cam/src/chemistry/pp_trop_ghg/mo_imp_sol.F90
M components/cam/src/chemistry/pp_trop_ghg/mo_sim_dat.F90
M components/cam/src/chemistry/pp_trop_mam3/mo_imp_sol.F90
M components/cam/src/chemistry/pp_trop_mam4/mo_imp_sol.F90
M components/cam/src/chemistry/pp_trop_mam7/mo_imp_sol.F90
M components/cam/src/chemistry/pp_trop_mozart/mo_imp_sol.F90
M components/cam/src/chemistry/pp_trop_mozart/mo_sim_dat.F90
M components/cam/src/chemistry/pp_trop_mozart_mam3/mo_imp_sol.F90
M components/cam/src/chemistry/pp_trop_mozart_soa/mo_imp_sol.F90
M components/cam/src/chemistry/pp_trop_strat_mam3/mo_imp_sol.F90
M components/cam/src/chemistry/pp_trop_strat_mam7/mo_imp_sol.F90
M components/cam/src/chemistry/pp_trop_strat_soa/mo_imp_sol.F90
M components/cam/src/chemistry/pp_waccm_mozart/mo_imp_sol.F90
M components/cam/src/chemistry/pp_waccm_mozart_mam3/mo_imp_sol.F90
M components/cam/src/chemistry/pp_waccm_mozart_sulfur/mo_imp_sol.F90
M components/cam/src/chemistry/pp_waccm_mozart_sulfur/mo_sim_dat.F90
M components/cam/src/chemistry/pp_waccm_sc/mo_imp_sol.F90
M components/cam/src/chemistry/pp_waccm_sc/mo_sim_dat.F90
M components/cam/src/chemistry/pp_waccm_sc_mam3/mo_imp_sol.F90
M components/cam/src/chemistry/pp_waccm_tsmlt/mo_imp_sol.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_imp_sol.F90
M components/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_imp_sol.F90
M components/cam/src/chemistry/pp_waccmx_mozart/mo_imp_sol.F90
M components/cam/src/chemistry/pp_waccmx_mozart/mo_sim_dat.F90
M components/cam/src/chemistry/utils/aerodep_flx.F90
M components/cam/src/chemistry/utils/aircraft_emit.F90
M components/cam/src/chemistry/utils/modal_aero_calcsize.F90
M components/cam/src/chemistry/utils/modal_aero_wateruptake.F90
M components/cam/src/chemistry/utils/prescribed_aero.F90
M components/cam/src/chemistry/utils/prescribed_ghg.F90
M components/cam/src/chemistry/utils/prescribed_ozone.F90
M components/cam/src/chemistry/utils/prescribed_strataero.F90
M components/cam/src/chemistry/utils/prescribed_volcaero.F90
M components/cam/src/chemistry/utils/tracer_data.F90
- Modified addfld calls to match new interface
M components/cam/src/control/cam_control_mod.F90
- Removed no longer needed use_64bit_nc logical variable
M components/cam/src/control/cam_history.F90
- Reimplemented history to use new CAM grid infrastructure
- Removed dycore-specific history output and references
- Created new CAM history namelist reader
M components/cam/src/control/cam_history_buffers.F90
- Modularized use of the dimind input to the accumulators
- Fixed some typos in teh local tod code
M components/cam/src/control/cam_history_support.F90
- Modified structures and functions to implement support for
new grid infrastructure
- Modified column_info structure for new patch (regional)
output infrastructure
- Removed dycore-specific information
- Added interp_info_t for regional output
- Made use of new cam_pio_def_dim and cam_pio_def_var interfaces
M components/cam/src/control/cam_initfiles.F90
- New cam_pio_openfile interface
M components/cam/src/control/cam_restart.F90
- New interface to init_restart_physics and init_restart_dynamics
- Moved initcom call to before read_restart_dynamics so that
function can use new I/O infrastructure
M components/cam/src/control/history_defaults.F90
- Modified addfld calls to match new interface
- Removed some unused addfld calls
M components/cam/src/control/history_scam.F90
- Modified addfld calls to match new interface
M components/cam/src/control/ncdio_atm.F90
- Modified infld interfaces to use new grid infrastructure
- GLOBAL grid input calls were moved to cam_pio_get_var
- check_var was moved to cam_pio_utils (and enhanced)
- check_dim functionality was replaced with cam_pio_def_dim
M components/cam/src/control/runtime_opts.F90
- Moved history-related namelist reading to cam_history.F90
M components/cam/src/control/sat_hist.F90
- Minor changes due to variables moving between modules
M components/cam/src/cpl/atm_comp_mct.F90
- Use new cam_pio_closefile interface
M components/cam/src/dynamics/eul/dyn_comp.F90
- Add call to define_cam_grids which sets up dynamics grids
- Modified addfld calls to match new interface
M components/cam/src/dynamics/eul/dyn_grid.F90
- Moved initgrid here from initcom.F90
- Added lat_d_out and lon_d_out to get_horiz_grid_d (i.e., in degrees)
- Added subroutine, define_cam_grids, to define dynamics grids
M components/cam/src/dynamics/eul/inidat.F90
- Changed GLOBAL infld calls to cam_pio_get_var
- Modified remaining infld calls for new interface
M components/cam/src/dynamics/eul/initcom.F90
- Removed body of subroutine and moved to initgrid in dyn_grid.F90
M components/cam/src/dynamics/eul/interp_mod.F90
- Changed stub functions to use new interface (see SE)
M components/cam/src/dynamics/eul/restart_dynamics.F90
- Replace hdimids input with call to cam_grid_write_attr
M components/cam/src/dynamics/fv/ctem.F90
- Modified addfld calls to match new interface
M components/cam/src/dynamics/fv/dyn_comp.F90
- Reordered start-up calls to enable define_cam_grids to work
- Add call to define_cam_grids which sets up dynamics grids
- Modified addfld calls to match new interface
M components/cam/src/dynamics/fv/dyn_grid.F90
- Moved initgrid here from initcom.F90 (need to clean initcom?)
- Added lat_d_out and lon_d_out to get_horiz_grid_d (i.e., in degrees)
- Added subroutine, define_cam_grids, to define dynamics grids
M components/cam/src/dynamics/fv/inidat.F90
- Modified infld calls for new interface
M components/cam/src/dynamics/fv/inital.F90
- Moved call to dyn_init to make sure grids and coordinates are defined
M components/cam/src/dynamics/fv/interp_mod.F90
- Changed stub functions to use new interface (see SE)
M components/cam/src/dynamics/fv/metdata.F90
- Modified infld calls for new interface
M components/cam/src/dynamics/fv/restart_dynamics.F90
- Replace hdimids input with call to cam_grid_write_attr
M components/cam/src/dynamics/se/dyn_comp.F90
- Add call to define_cam_grids which sets up dynamics grids
- Minor cleanup including replacement of some MPI shorthand calls
M components/cam/src/dynamics/se/dyn_grid.F90
- Moved initgrid here from initcom.F90
- Added lat_d_out and lon_d_out to get_horiz_grid_d (i.e., in degrees)
- Added subroutine, define_cam_grids, to define dynamics grids
- Changed global coordinate functions to not store permanent globals
M components/cam/src/dynamics/se/gravity_waves_sources.F90
- Added elem input to support future edge exchange interface
M components/cam/src/dynamics/se/inidat.F90
- Read data into differently-shaped structures as part of new
grid support
- Modified infld calls for new interface
M components/cam/src/dynamics/se/interp_mod.F90
- Modified interface to support defining interpolation on a per
file basis (e.g., different files can have different resolutions)
- Define CAM grid for interpolated output for each file
M components/cam/src/dynamics/se/nctopo_util_mod.F90
- Modified output data structures to match new grid format
M components/cam/src/dynamics/se/restart_dynamics.F90
- Replace hdimids input with call to cam_grid_write_attr
M components/cam/src/dynamics/se/stepon.F90
- Modified addfld calls to match new interface
M components/cam/src/dynamics/sld/dyn_comp.F90
- Add call to define_cam_grids which sets up dynamics grids
- Minor cleanup including replacement of some MPI shorthand calls
M components/cam/src/dynamics/sld/dyn_grid.F90
- Moved initgrid here from initcom.F90
- Added lat_d_out and lon_d_out to get_horiz_grid_d (i.e., in degrees)
- Added subroutine, define_cam_grids, to define dynamics grids
M components/cam/src/dynamics/sld/interp_mod.F90
- Changed stub functions to use new interface (see SE)
M components/cam/src/dynamics/se/inidat.F90
- Changed GLOBAL infld calls to cam_pio_get_var
M components/cam/src/dynamics/sld/restart_dynamics.F90
- Replace hdimids input with call to cam_grid_write_attr
M components/cam/src/physics/cam/aer_rad_props.F90
M components/cam/src/physics/cam/aoa_tracers.F90
M components/cam/src/physics/cam/cam3_ozone_data.F90
M components/cam/src/physics/cam/cam_diagnostics.F90
M components/cam/src/physics/cam/check_energy.F90
M components/cam/src/physics/cam/cldwat.F90
M components/cam/src/physics/cam/cldwat2m_macro.F90
M components/cam/src/physics/cam/cloud_cover_diags.F90
M components/cam/src/physics/cam/cloud_diagnostics.F90
M components/cam/src/physics/cam/cloud_fraction.F90
M components/cam/src/physics/cam/clubb_intr.F90
M components/cam/src/physics/cam/co2_cycle.F90
M components/cam/src/physics/cam/constituent_burden.F90
M components/cam/src/physics/cam/conv_water.F90
M components/cam/src/physics/cam/convect_deep.F90
M components/cam/src/physics/cam/convect_shallow.F90
M components/cam/src/physics/cam/cospsimulator_intr.F90
M components/cam/src/physics/cam/eddy_diff.F90
M components/cam/src/physics/cam/gw_drag.F90
M components/cam/src/physics/cam/hetfrz_classnuc_cam.F90
M components/cam/src/physics/cam/hk_conv.F90
M components/cam/src/physics/cam/macrop_driver.F90
M components/cam/src/physics/cam/micro_mg_cam.F90
M components/cam/src/physics/cam/microp_aero.F90
M components/cam/src/physics/cam/modal_aer_opt.F90
M components/cam/src/physics/cam/ndrop.F90
M components/cam/src/physics/cam/ndrop_bam.F90
M components/cam/src/physics/cam/nucleate_ice_cam.F90
M components/cam/src/physics/cam/phys_debug.F90
- Modified addfld calls to match new interface
M components/cam/src/physics/cam/phys_grid.F90
- Define 'physgrid' physics CAM grid
M components/cam/src/physics/cam/physics_buffer.F90.in
- Changes to restart code to use new CAM grid interfaces
- Added interface (pbuf_dump_pbuf) useful for debugging
M components/cam/src/physics/cam/physpkg.F90
- Modified infld calls for new interface
M components/cam/src/physics/cam/rad_constituents.F90
M components/cam/src/physics/cam/radiation.F90
M components/cam/src/physics/cam/radiation_data.F90
- Modified addfld calls to match new interface
M components/cam/src/physics/cam/restart_physics.F90
- Changes to use new CAM grid interfaces
- Replaced hdimids input with grid info
M components/cam/src/physics/cam/rk_stratiform.F90
M components/cam/src/physics/cam/sslt_rebin.F90
- Modified addfld calls to match new interface
M components/cam/src/physics/cam/subcol.F90
- File should be inout in subcol_init_restart
M components/cam/src/physics/cam/subcol_utils.F90.in
- Cleaned up (probably buggy) restart code using new grid
interfaces and cleaner convenience interfaces from cam_pio_utils
M components/cam/src/physics/cam/tidal_diag.F90
M components/cam/src/physics/cam/tracers.F90
M components/cam/src/physics/cam/tropopause.F90
M components/cam/src/physics/cam/unicon_cam.F90
M components/cam/src/physics/cam/uwshcu.F90
M components/cam/src/physics/cam/vertical_diffusion.F90
M components/cam/src/physics/cam/zm_conv.F90
M components/cam/src/physics/cam/zm_conv_intr.F90
- Modified addfld calls to match new interface
M components/cam/src/physics/carma/cam/carma_getH2O.F90
M components/cam/src/physics/carma/cam/carma_getH2SO4.F90
M components/cam/src/physics/carma/cam/carma_getT.F90
M components/cam/src/physics/carma/cam/carma_intr.F90
- Cleaned up input routines to use new cam_pio_get_var interface
M components/cam/src/physics/carma/models/cirrus_dust/carma_model_mod.F90
M components/cam/src/physics/carma/models/dust/carma_model_mod.F90
M components/cam/src/physics/rrtmg/ebert_curry.F90
M components/cam/src/physics/rrtmg/radiation.F90
M components/cam/src/physics/rrtmg/slingo.F90
M components/cam/src/physics/waccm/iondrag.F90
M components/cam/src/physics/waccm/nlte_lw.F90
M components/cam/src/physics/waccm/qbo.F90
M components/cam/src/physics/waccm/radheat.F90
M components/cam/src/physics/waccm/waccm_forcing.F90
M components/cam/src/physics/waccmx/ionosphere.F90
M components/cam/src/physics/waccmx/majorsp_diffusion.F90
M components/cam/src/unit_drivers/drv_input_data.F90
M components/cam/src/utils/cam_dom/ocn_comp.F90
M components/cam/src/utils/cam_dom/sst_data.F90
- Modified addfld calls to match new interface
A + components/cam/src/utils/cam_grid_support.F90
- New module containing types and functionality for horizontal
coordinates and grids
A + components/cam/src/utils/cam_map_utils.F90
- New module containing type (cam_filemap_t) and functionalty
for maps between field arrays and NetCDF file variables
M components/cam/src/utils/cam_pio_utils.F90
- Substantially rewritten PIO interface utility layer
- Removed all dycore or physics grid specific decomposition code
- Use new CAM grid and cam_filemap_t interfaces to create and
use PIO decompositions
- Added convience interfaces which allow flexible inquiry and
definition of NetCDF dimensions and variables and also perform
error checking (with some functionality moved from ncdio_atm
functions check_var and check_dim)
- Added debug interface for comparing fields
M components/cam/src/utils/spmd_utils.F90
- Added mpi_min forward reference (needed for filemap code)
M components/cam/test/system/archive_baseline.sh
- Fixed typo related to Goldbach ==> Hobart switch
M components/cam/test/system/input_tests_master
- Added new tests to protect regional and interpolated output
A + components/cam/test/system/nl_files/ghgrmp_e8
- New regional output test
Existing ghgrmp namelist not valid with corrected region selection
A + components/cam/test/system/nl_files/ghgrmp_f4
- New regional output test
Existing ghgrmp namelist not valid with corrected region selection
A + components/cam/test/system/nl_files/ghgrmp_unstruct
- New regional output and interpolated output test for the SE dycore
M components/cam/test/system/tests_pretag_goldbach_nag
- Added new tests for regional output and interpolated output
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
004 bl132 TBL.sh e48c4dh ghgrmp_e8+1850_cam4 9s ...............................................FAIL! rc= 7 at Sat Aug 8 22:31:46 MDT 2015
024 bl332 TBL.sh f4c5dm ghgrmp_f4+1850_cam5 9s ................................................FAIL! rc= 7 at Sat Aug 8 23:23:57 MDT 2015
- There are no existing baselines for these tests
hobart/nag:
010 bl112 TBL.sh e8c5dm ghgrmp_e8 9s ..........................................................FAIL! rc= 7 at Sat Aug 8 19:37:43 MDT 2015
- There are no existing baselines for this test
hobart/pgi or jaguar/pgi:
010 bl112 TBL.sh e8c5dm ghgrmp_e8 9s ..........................................................FAIL! rc= 7 at Sat Aug 8 19:52:14 MDT 2015
- There are no existing baselines for this test
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_4_05
Originator(s): cacraig
Date: 8/5/2015
One-line Summary: Update CAM regression testing to hobart and update several externals
Purpose of changes:
- CAM regression testing will now use hobart
- cosp, rtm and homme all had compiler issues when updated to NAG6.0 which were corrected
and were non-answer changing. These externals are used with this tag
- the cime external was pointing to a private user's branch and is now pointing to a
branch on the CESM cime repository
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M .
M SVN_EXTERNAL_DIRECTORIES
M components/cam
M components/cam/SVN_EXTERNAL_DIRECTORIES
- updated to externals which compile and run with NAG6.0
M components/cam/test/system/archive_baseline.sh
M components/cam/test/system/test_driver.sh
M components/cam/doc/ChangeLog_template
- changed over to using hobart for CAM regression testing
M components/cam/test/system/tests_pretag_hobart_nag
- added new test which was added to the goldbach tests, but
had not been added to the identical hobart list
M components/cam/bld/configure
M components/cam/bld/Makefile.in
M components/cam/bld/config_files/definition.xml
- changed over to using hobart for testing
- added temporary nag53 configure/namelist option to allow compilation
of nag53 (which is what has been used on goldbach) while moving ahead
with nag6.0 on hobart
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB
goldbach/nag: all BFB
goldbach/pgi or jaguar/pgi: all BFB
* For the hobart baseline tests, updated cam5_4_04 with the minimal changes required to run on hobart in configure and Makefile.in
hobart/nag: all BFB except for cosp and homme tests which did not have the baselines pass due to the inability
of cam5_4_04 to compile with the old externals
hobart/pgi or jaguar/pgi: all BFB
===============================================================
===============================================================
Tag name: cam5_4_04
Originator(s): cacraig
Date: July 31, 2015
One-line Summary: Update to CESM1_4_beta06 externals
Purpose of changes: Update to current CESM externals
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M .
M SVN_EXTERNAL_DIRECTORIES
- update CLM, CICE and RTM to most current tags (already had updated CIME in the last CAM tag)
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
004 bl132 TBL.sh e48c4dh ghgrmp+1850_cam4 9s ..................................................FAIL! rc= 7 at Fri Jul 31 16:22:14 MDT 2015
009 bl203 TBL.sh fsd1.9c4wtsmlth outfrq3s_sd+sd_1975-2010_ccmi_refc1_waccm_tsmlt 9s ...........FAIL! rc= 7 at Fri Jul 31 16:27:36 MDT 2015
012 bl322 TBL.sh f1.9c5nonedh outfrq3s_lb0 9s .................................................FAIL! rc= 7 at Fri Jul 31 16:27:42 MDT 2015
016 bl326 TBL.sh f1.9c4wscdm outfrq3s_2005+waccm_sc_2005-2100_cam4_rcp45 9s ...................FAIL! rc= 7 at Fri Jul 31 16:27:53 MDT 2015
020 bl331 TBL.sh f4c4dh co2rmp+1850_cam4 9s ...................................................FAIL! rc= 7 at Fri Jul 31 16:27:57 MDT 2015
024 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...................................................FAIL! rc= 7 at Fri Jul 31 16:28:00 MDT 2015
029 bl336 TBL.sh f4c4wmdh outfrq3s+waccm_1850_cam4 9s .........................................FAIL! rc= 7 at Fri Jul 31 16:28:25 MDT 2015
032 bl353 TBL.sh f1.9c5.4dh outfrq3s 9s .......................................................FAIL! rc= 7 at Fri Jul 31 16:28:33 MDT 2015
035 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Fri Jul 31 16:35:51 MDT 2015
039 bl373 TBL.sh f1.9c4mozdh outfrq3s+2000_cam4_trop_chem 9s ..................................FAIL! rc= 7 at Fri Jul 31 16:36:14 MDT 2015
041 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...............................................FAIL! rc= 7 at Fri Jul 31 16:36:18 MDT 2015
043 bl376 TBL.sh f1.9c4mozdh+testmech outfrq3s 9s .............................................FAIL! rc= 7 at Fri Jul 31 16:36:29 MDT 2015
046 bl374 TBL.sh f1.9c4fastdh outfrq3s+1850-2005_cam4_super_fast_llnl 9s ......................FAIL! rc= 7 at Fri Jul 31 16:36:39 MDT 2015
049 bl390 TBL.sh f1.9c4wmxh outfrq3s+waccmx_2000_cam4 9s ......................................FAIL! rc= 7 at Fri Jul 31 16:38:21 MDT 2015
053 bl397 TBL.sh f1.9c4cm outfrq3s_cosp+1850-2005_cam4 9s .....................................FAIL! rc= 7 at Fri Jul 31 16:38:30 MDT 2015
056 bl421 TBL.sh f1.9c5mam4dh outfrq3s 9s .....................................................FAIL! rc= 7 at Fri Jul 31 16:38:38 MDT 2015
058 bl430 TBL.sh f1.9c5mzmam3dh outfrq3s_NEUwetdep 9s .........................................FAIL! rc= 7 at Fri Jul 31 16:38:52 MDT 2015
061 bl440 TBL.sh f1.9c4soadh outfrq3s+soa_chem_megan_emis 9s ..................................FAIL! rc= 7 at Fri Jul 31 16:39:23 MDT 2015
064 bl474 TBL.sh f1.9c5wtmam3h outfrq3s+1850_waccm_tsmlt_cam5 9s ..............................FAIL! rc= 7 at Fri Jul 31 16:41:30 MDT 2015
067 bl482 TBL.sh f1.9c5wscmam3dh outfrq3s_2005+waccm_sc_1955-2005_cam5 9s .....................FAIL! rc= 7 at Fri Jul 31 16:41:54 MDT 2015
071 bl387 TBL.sh f1.9c4dm outfrq3s+1850-2005_cam4 9s ..........................................FAIL! rc= 7 at Fri Jul 31 16:41:59 MDT 2015
074 bl388 TBL.sh f1.9c4m outfrq24h+1850_cam4 2d ...............................................FAIL! rc= 7 at Fri Jul 31 16:42:03 MDT 2015
077 bl379 TBL.sh f1.9c4wmh outfrq24h+waccm_2005-2100_cam4_rcp45 2d ............................FAIL! rc= 7 at Fri Jul 31 16:43:02 MDT 2015
088 bl740 TBL.sh h30c5h outfrq3s_lb0 9s .......................................................FAIL! rc= 7 at Fri Jul 31 16:43:25 MDT 2015
goldbach/nag:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s ...........................................................FAIL! rc= 7 at Fri Jul 31 10:24:51 MDT 2015
010 bl112 TBL.sh e8c5dm ghgrmp 9s .............................................................FAIL! rc= 7 at Fri Jul 31 10:32:46 MDT 2015
014 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Fri Jul 31 10:39:23 MDT 2015
018 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Fri Jul 31 10:50:17 MDT 2015
028 bl314 TBL.sh f10c4wscdm outfrq3s 9s .......................................................FAIL! rc= 7 at Fri Jul 31 11:08:53 MDT 2015
034 bl318 TBL.sh f10c4cdm sat_hist 9s .........................................................FAIL! rc= 7 at Fri Jul 31 11:25:26 MDT 2015
037 bl324 TBL.sh f4c4rrtmgdm outfrq3s 9s ......................................................FAIL! rc= 7 at Fri Jul 31 11:33:05 MDT 2015
040 bl325 TBL.sh f10c5t5mdmsc outfrq3s_subcol 9s ..............................................FAIL! rc= 7 at Fri Jul 31 11:41:10 MDT 2015
046 bl338 TBL.sh f10c5dm rad_diag_mam 9s ......................................................FAIL! rc= 7 at Fri Jul 31 11:48:26 MDT 2015
049 bl354 TBL.sh f1.9c5.4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Fri Jul 31 12:16:04 MDT 2015
052 bl368 TBL.sh f1.9c5mg2dm outfrq3s_macmic2_classnuc 9s .....................................FAIL! rc= 7 at Fri Jul 31 12:48:16 MDT 2015
058 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Fri Jul 31 13:17:29 MDT 2015
061 bl424 TBL.sh f10c4wmxdm outfrq3s 9s .......................................................FAIL! rc= 7 at Fri Jul 31 13:30:50 MDT 2015
064 bl471 TBL.sh f10c5wmam3dm outfrq3s_modalstrat 9s ..........................................FAIL! rc= 7 at Fri Jul 31 13:43:37 MDT 2015
067 bl805 TBL.sh f4c5carmtt1dm outfrq3s_carma 9s ..............................................FAIL! rc= 7 at Fri Jul 31 14:10:19 MDT 2015
goldbach/pgi or jaguar/pgi:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s ...........................................................FAIL! rc= 7 at Fri Jul 31 10:24:52 MDT 2015
010 bl112 TBL.sh e8c5dm ghgrmp 9s .............................................................FAIL! rc= 7 at Fri Jul 31 10:32:47 MDT 2015
016 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Fri Jul 31 10:40:39 MDT 2015
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Fri Jul 31 10:54:05 MDT 2015
033 bl314 TBL.sh f10c4wscdm outfrq3s 9s .......................................................FAIL! rc= 7 at Fri Jul 31 11:15:12 MDT 2015
036 bl315 TBL.sh f10c3dm outfrq3s+cam3 9s .....................................................FAIL! rc= 7 at Fri Jul 31 11:21:02 MDT 2015
041 bl317 TBL.sh f10c5cdm outfrq3s 9s .........................................................FAIL! rc= 7 at Fri Jul 31 11:35:35 MDT 2015
045 bl319 TBL.sh f10c5nonedm outfrq3s_bam 9s ..................................................FAIL! rc= 7 at Fri Jul 31 11:42:20 MDT 2015
048 bl320 TBL.sh f10c5nonedm rad_diag 9s ......................................................FAIL! rc= 7 at Fri Jul 31 11:43:57 MDT 2015
051 bl321 TBL.sh f10c5cdm atrain 9s ...........................................................FAIL! rc= 7 at Fri Jul 31 11:46:52 MDT 2015
054 bl334 TBL.sh f10c5udm outfrq3s 9s .........................................................FAIL! rc= 7 at Fri Jul 31 11:55:06 MDT 2015
057 bl354 TBL.sh f1.9c5.4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Fri Jul 31 12:31:01 MDT 2015
060 bl420 TBL.sh f4c5mam7dm outfrq3s 9s .......................................................FAIL! rc= 7 at Fri Jul 31 12:44:41 MDT 2015
063 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Fri Jul 31 12:57:09 MDT 2015
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: all runs which use cice have different answers due to changes in the cice tag
- what platforms/compilers: all
- nature of change (roundoff; larger than roundoff but same climate; new
climate): not analyzed as these are answers due to a compoonent other than CAM
===============================================================
===============================================================
Tag name: cam5_4_03
Originator(s): fvitt, eaton
Date: Thu Jul 30 13:07:08 MDT 2015
One-line Summary: Modify shr_flux_mod; change fc_type xlf to ibm; fix mam
diagnostic radiation calcs.
Purpose of changes:
. Fix problems from the update to cime1.1.15 in the cesm1_4_alpha06a
testing.
- Modify shr_flux_mod.F90 to support setting the values of certain
constants as required for aquaplanet mode. Remove the cam_aqua copy of
that file. The mods to shr_flux_mod.F90 do *not* include the previous
mods to maintain BFB results for the cam3 aquaplanet. We decided to
deprecate the cam3 version of the flux algorithm which contained a bug.
- Modify CAM's configure to accept COMPILER=ibm as the compiler "type"
for xlf compilers. This fixes the build of the chemistry proprocessor.
. Fix memory allocation errors in code to recalculate the aerosol water
uptake when doing diagnostic radiation calculations with modal aerosols.
Also add test for diagnostic radiation calc w/ MAM.
Bugs fixed (include bugzilla ID):
bugz 2194: change 86400_r8 to 86400._r8 in 3 restart_dynamics files
Describe any changes made to build system:
. valid value for the -fc_type arg to configure was changed from 'xlf' to
'ibm' for consistency with CESM scripts.
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
components/cam/src/utils/cam_aqua/shr_flux_mod.F90
. made changes to csm_share version of shr_flux_mod so that it can be used
by CAM's aquaplanet mode.
List all subroutines added and what they do:
components/cam/test/system/nl_files/rad_diag_mam
. set up a diagnostic radiation calculation for MAM3
List all existing files that have been modified, and describe the changes:
components/cam/bld/cam.buildnml
. add the -fc_type arg back to configure
components/cam/bld/config_files/definition.xml
. add parameter build_chem_proc to force the chem preprocessor to be built
components/cam/bld/configure
. add -build_chem_proc flag
. change the fc_type: xlf -> ibm
. remove env var COMPILER as a source for the fc_type info. Use fc_type
arg instead.
components/cam/bld/perl5lib/Build/ChemPreprocess.pm
. mods to recognize the build_chem_proc parameter
. add code to check that the fortran compiler is found
components/cam/src/dynamics/sld/restart_dynamics.F90
components/cam/src/dynamics/se/restart_dynamics.F90
components/cam/src/dynamics/eul/restart_dynamics.F90
. change 86400_r8 to 86400._r8
components/cam/src/chemistry/utils/modal_aero_calcsize.F90
components/cam/src/chemistry/utils/modal_aero_wateruptake.F90
. remove allocation of arrays needed for diagnostic calculations
. add check that all arrays needed for diagnostic calcs have been
associated
components/cam/src/physics/cam/modal_aer_opt.F90
. modal_aero_sw, modal_aero_lw
- add allocation/deallocation of arrays needed by diagnostic calcs
components/cam/src/utils/physconst.F90
. call shr_flux_adjust_constants after the namelist read that gets the
desired values.
components/cam/test/system/TCB.sh
components/cam/test/system/TCB_ccsm.sh
. force build of chem proc for all tests that use custom chem
components/cam/test/system/input_tests_master
. add sm338 TSM.sh f10c5dm rad_diag_mam 9s
. add bl338 TBL.sh f10c5dm rad_diag_mam 9s
components/cam/test/system/tests_pretag_goldbach_nag
. add sm338, bl338
components/cam/test/system/tests_pretag_yellowstone
. reorder test list
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS except:
009 bl203 TBL.sh fsd1.9c4wtsmlth outfrq3s_sd+sd_1975-2010_ccmi_refc1_waccm_tsmlt 9s ...........FAIL! rc= 7 at Tue Jul 28 18:41:37 MDT 2015
012 bl322 TBL.sh f1.9c5nonedh outfrq3s_lb0 9s .................................................FAIL! rc= 7 at Tue Jul 28 18:48:29 MDT 2015
016 bl326 TBL.sh f1.9c4wscdm outfrq3s_2005+waccm_sc_2005-2100_cam4_rcp45 9s ...................FAIL! rc= 7 at Tue Jul 28 19:02:12 MDT 2015
024 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...................................................FAIL! rc= 7 at Tue Jul 28 19:11:30 MDT 2015
029 bl336 TBL.sh f4c4wmdh outfrq3s+waccm_1850_cam4 9s .........................................FAIL! rc= 7 at Tue Jul 28 19:23:40 MDT 2015
032 bl353 TBL.sh f1.9c5.4dh outfrq3s 9s .......................................................FAIL! rc= 7 at Tue Jul 28 19:34:39 MDT 2015
035 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Tue Jul 28 19:47:41 MDT 2015
039 bl373 TBL.sh f1.9c4mozdh outfrq3s+2000_cam4_trop_chem 9s ..................................FAIL! rc= 7 at Tue Jul 28 20:02:58 MDT 2015
041 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...............................................FAIL! rc= 7 at Tue Jul 28 20:05:35 MDT 2015
043 bl376 TBL.sh f1.9c4mozdh+testmech outfrq3s 9s .............................................FAIL! rc= 7 at Tue Jul 28 20:11:49 MDT 2015
046 bl374 TBL.sh f1.9c4fastdh outfrq3s+1850-2005_cam4_super_fast_llnl 9s ......................FAIL! rc= 7 at Tue Jul 28 20:18:34 MDT 2015
049 bl390 TBL.sh f1.9c4wmxh outfrq3s+waccmx_2000_cam4 9s ......................................FAIL! rc= 7 at Tue Jul 28 20:27:03 MDT 2015
056 bl421 TBL.sh f1.9c5mam4dh outfrq3s 9s .....................................................FAIL! rc= 7 at Tue Jul 28 20:41:16 MDT 2015
058 bl430 TBL.sh f1.9c5mzmam3dh outfrq3s_NEUwetdep 9s .........................................FAIL! rc= 7 at Tue Jul 28 20:50:19 MDT 2015
061 bl440 TBL.sh f1.9c4soadh outfrq3s+soa_chem_megan_emis 9s ..................................FAIL! rc= 7 at Tue Jul 28 21:07:11 MDT 2015
064 bl474 TBL.sh f1.9c5wtmam3h outfrq3s+1850_waccm_tsmlt_cam5 9s ..............................FAIL! rc= 7 at Tue Jul 28 21:20:08 MDT 2015
067 bl482 TBL.sh f1.9c5wscmam3dh outfrq3s_2005+waccm_sc_1955-2005_cam5 9s .....................FAIL! rc= 7 at Tue Jul 28 21:44:03 MDT 2015
077 bl379 TBL.sh f1.9c4wmh outfrq24h+waccm_2005-2100_cam4_rcp45 2d ............................FAIL! rc= 7 at Tue Jul 28 22:09:42 MDT 2015
080 bl731 TBL.sh h16c3aqdh aqua+aquaplanet_cam3 9s ............................................FAIL! rc= 7 at Tue Jul 28 22:12:45 MDT 2015
083 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ........................................FAIL! rc= 7 at Tue Jul 28 22:24:03 MDT 2015
088 bl740 TBL.sh h30c5h outfrq3s_lb0 9s .......................................................FAIL! rc= 7 at Tue Jul 28 22:28:37 MDT 2015
All failures except bl731 are due to updating the cime external which
pulled in a change to not use the gamma and erf functions intrinsics with
the intel compiler. This is a change for performance reasons.
bl731 fails because cam3/aquaplanet changes answers due to no longer
maintaining a bug in the shr_flux_mod calculations.
goldbach/nag: All PASS except:
025 bl313 TBL.sh f10c3aqdm outfrq3s+aquaplanet_cam3 3s ........................................FAIL! rc= 7 at Tue Jul 28 15:21:37 MDT 2015
046 bl338 TBL.sh f10c5dm rad_diag_mam 9s ......................................................FAIL! rc= 7 at Tue Jul 28 16:08:43 MDT 2015
goldbach/pgi: All PASS except:
030 bl313 TBL.sh f10c3aqdm outfrq3s+aquaplanet_cam3 3s ........................................FAIL! rc= 7 at Tue Jul 28 15:29:37 MDT 2015
bl313 fails due to cam3/aquaplanet answer change as noted above.
bl338 fails due to being a new test with no existing baseline.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB except:
1) cam3/aquaplanet changes answers due to no longer maintaining a bug in
the shr_flux_mod calculations.
2) An intel compiler change to not use the gamma or erf intrinsic functions
(in shr_spfn_mod.F90).
===============================================================
===============================================================
Tag name: cam5_4_02
Originator(s): cacraig
Date: July 17, 2015
One-line Summary: Code cleanup/rearrangement and temporary fix to allow CESM to run on mira, but breaks chemistry preprocessor
Purpose of changes:
- Code reviews of the SPCAM branch demonstrated several areas where
cleanup/rearrangement of CAM code was necessary to allow other
models to use sections of CAM cleanly.
- temporary fix to allow CESM to run on mira, but chemistry preprocessor will be broken
- radiation now has an optional argument to not output its results. All of the output
is put into a structure which is returned from radiation_tend and can be used outside
radiation.
- outfld calls in radiation are now in two routines called by radiation (one for
shortwave and the other for longwave). These routines are not called if the
optional do_output variable is set to .false.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
A components/cam/src/physics/cam/radiation_utils.F90
A components/cam/src/physics/rrtmg/radiation_utils.F90
- contains the structure for outputs from radiation
List all existing files that have been modified, and describe the changes:
M components/cam/test/system/input_tests_master
- rearranged list so it is ordered consistently and corrected a couple of typos
M components/cam/bld/configure
- make the check for setting pbl_pkg more precise
M components/cam/bld/cam.buildnml
- temporary fix to allow CESM to run on mira. This will break chem preprocessor compilation
M components/cam/src/physics/cam/radiation.F90
M components/cam/src/physics/rrtmg/radiation.F90
M components/cam/src/unit_drivers/rad/unit_driver.F90
- added structure which contains all radiation output data
- added optional flag to turn off calling the outfld calls inside radiation_tend
M components/cam/src/physics/cam/cam_diagnostics.F90
M components/cam/src/physics/cam/physpkg.F90
- check status on a couple of optional pbuf fields before reading them
M components/cam/src/physics/cam/macrop_driver.F90
- fix bug where need to only process fields from top_lev:pver instead of the entire dimension as
valid data may not be set outside those bounds
M components/cam/src/utils/cam_pio_utils.F90
- temporary fix until the new cam history is implemented
M components/cam/src/chemistry/modal_aero/sox_cldaero_mod.F90
M components/cam/src/chemistry/modal_aero/aero_model.F90
M components/cam/src/chemistry/bulk_aero/aero_model.F90
M components/cam/src/chemistry/bulk_aero/sox_cldaero_mod.F90
M components/cam/src/chemistry/aerosol/mo_setsox.F90
- move addfld/outfld calls to the interface layer
M components/cam/src/chemistry/utils/modal_aero_wateruptake.F90
- allow modal_aero_kohler to be public (needed by SPCAM)
M components/cam/src/chemistry/mozart/chemistry.F90
- call sox_inti prior to aero_model_init so that the logical "has_sox" is available
when initializing the aerosols
M components/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
M components/cam/src/chemistry/mozart/mo_waccm_hrates.F90
M components/cam/src/chemistry/mozart/mo_mass_xforms.F90
- from a performance standpoint, dummy arrays are more efficient if they are declared with
their dimensions. Made all arrays be ncol long for consistency.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB
goldbach/nag: all BFB
goldbach/pgi or jaguar/pgi: all BFB
===============================================================
===============================================================
Tag name: cam5_4_01
Originator(s): santos
Date: 2015/07/13
One-line Summary: Miscellaneous minor changes.
Purpose of changes:
- Preliminary changes so that CAM can be built by CIME on hobart, and
using the new NAG 6.0 libraries.
- Minor fixes to FV and physpkg.
- Add some new diagnostic fields.
Bugs fixed (include bugzilla ID):
- #238: Remove "modcomm" namelist variables? (Actually resolved by
restoring their original functionality.)
- Occasional floating-point exceptions in CAM5 (not noted on bugzilla?).
Describe any changes made to build system:
Describe any changes made to the namelist:
- Changed modcomm_gatscat, modcomm_geopk, and modcomm_transpose so that
they are added to the correct namelist group.
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: (collection of small independent changes; mostly eaton)
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/bld/namelist_files/namelist_definition.xml
M components/cam/src/dynamics/fv/dyn_comp.F90
M components/cam/src/dynamics/fv/pmgrid.F90
M components/cam/src/dynamics/fv/spmd_dyn.F90
- Correct handling of modcomm variables so that they actually
affect communication as intended for CAM-FV.
M components/cam/cimetest/testlist_cam.xml
- Change goldbach to hobart in CIME tests.
M components/cam/src/control/cam_history_buffers.F90
M components/cam/src/dynamics/fv/dynamics_vars.F90
M components/cam/src/dynamics/fv/restart_dynamics.F90
M components/cam/src/physics/cam/cldwat.F90
- Fix incorrect uses of kind numbers that cause problems for NAG
when the "-kind=byte" flag is omitted.
M components/cam/src/physics/cam/cam_diagnostics.F90
- Add new "T400", "U500", and "V500" diagnostics.
M components/cam/src/physics/cam/physpkg.F90
- Fix bug where flux arrays that were only set up to ncol were
having every element divided, which could cause crashes due to
floating-point exceptions in DEBUG mode, from computing with
uninitialized data.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
goldbach/nag:
goldbach/pgi or jaguar/pgi:
CAM tag used for the baseline comparison tests if different than previous
tag:
- cam5_3_92 was used, because it is identical to cam5_4_00.
===============================================================
===============================================================
Tag name: cam5_4_00
Originator(s): cacraig
Date: July 8, 2015
One-line Summary: Copy of cam5_3_92 to start the cam5_4 series
Purpose of changes:
No changes, simply the start of the cam5_4 series
===============================================================
===============================================================
Tag name: cam5_3_92
Originator(s): eaton
Date: Mon Jun 29 13:31:38 MDT 2015
One-line Summary: Mods for passive tracer life cycle experiments
Purpose of changes:
Implement functionality for new simplified model "life cycle" experiments.
. Modify the test tracer functionality:
- require that the number of test tracers be specified directly via the
-nadv_tt option to configure. Disable the old optional method of
setting -nadv larger than needed by the other phys/chem schemes and
assuming that remaining constituent slots will be filled with test
tracers.
- allow user to specify names of tracers to be read from the initial file.
. Finish implementing the namelist reading for the Eulerian and SLD
dycores. Move default values that were set in the code to the namelist
defaults file. This fixes a hidden bug in the Eul dycore where a user
specified parameter could be overwritten by a hardcoded value.
. Implement the option to apply Nth order horizontal diffusion in the
Eulerian dycore. The default remains 4th order diffusion. Add namelist
specification of the diffusion order, and rename the dif4 variable to
eul_hdif_coef to reflect that it's for the Nth order diffusion.
. Allow user to specify that a topo dataset will not be used rather than
requiring that a dataset full of zeros be provided.
. Mods to inidat routines to allow unicon to initialize its constituents.
Bugs fixed (include bugzilla ID):
. fix Eul dycore overwriting user specified values of dif2 and dif4
Describe any changes made to build system: none
Describe any changes made to the namelist:
. remove tracers_flag variable - replace by test_tracer_num variable which
is meant to be set by build-namelist to the value of the configure
parameter nadv_tt
. add test_tracer_names for user specification of passive tracers to be read
from the initial file
. change names of dif2,divdampn,eps,kmxhdc to dyn_spectral_* These
names are read by both EUL and SLD dycores
. For Eulerian dycore add variables:
eul_hdif_order - sets order N of horizonal diffusion operator
(defaults to 4)
eul_hdif_coef - Nth order horizonal diffusion coefficient
. add use_topo_file. The default behavior is always to provide a default
topo file. Setting use_topo_file=.false. allows the user to specify that
PHIS, SGH, SGH30, and LANDM_COSLAT are all zero without having to
supply a topo file.
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: cam developers
List all subroutines eliminated:
List all subroutines added and what they do:
components/cam/bld/namelist_files/use_cases/adiabatic_LC1.xml
. add initial files for EUL T42, T85, T170 at 30, 60, and 120 levels
. set all topo fields to 0. via new use_topo_file=.false.
. add names of passive tracers for LC1 simulations
'strat', 'tropo', 'th320', 'th340', 'wrt'
. Set up history output for 6-hrly instantaneous values of
'U','V','T','PS','strat','tropo','th320','th340','wrt'
List all existing files that have been modified, and describe the changes:
components/cam/bld/cam.buildnml
. parse the ATM_GRID value to extract the number of levels and add the
-nlev arg to the configure commandline if it has been specified. This
supports the CESM scripts.
components/cam/bld/build-namelist
. if new use_topo_file variable is set false, then don't add default for
bnd_topo
. set new variable test_tracer_num to the value of configuration parameter
nadv_tt. Remove setting tracers_flag.
. add add_default calls for dyn_spectral_dif2, dyn_spectral_dif4,
dyn_spectral_divdampn, dyn_spectral_eps, dyn_spectral_kmxhdc
. add add_default calls for eul_hdif_order, eul_hdif_coef
components/cam/bld/namelist_files/namelist_defaults_cam.xml
. add default values for dyn_spectral_dif2, dyn_spectral_dif4,
dyn_spectral_divdampn, dyn_spectral_eps, dyn_spectral_kmxhdc,
eul_hdif_order, eul_hdif_coef. The values
had previously be hardcoded in eul_control_mod.F90 and
sld_control_mod.F90. Note that the default for eul_hdif_order is 4, and
the coefficient values which previously assigned to dyn_spectral_dif4 are
now assigned to eul_hdif_coef for the Eulerian dycore only.
components/cam/bld/namelist_files/namelist_definition.xml
. add new var test_tracer_names in group test_tracers_nl
. remove tracers_flag
. add test_tracer_num to group test_tracers_nl
. change names of dif2,divdampn,eps,kmxhdc to dyn_spectral_* These
names are read by both EUL and SLD dycores
. For Eulerian dycore add variables:
eul_hdif_order - sets order N of horizonal diffusion operator
(defaults to 4)
eul_hdif_coef - Nth order horizonal diffusion coefficient
. add use_topo_file to allow the user to specify that no topo file should
be added to the namelist specification.
. move ncdata and bnd_topo to group cam_initfiles_nl
components/cam/src/chemistry/mozart/mo_drydep.F90
. access ncdata using initial_file_get_id method of cam_initfiles rather
than from filenames module
components/cam/src/control/cam_history.F90
. access ncdata and bnd_topo from cam_initfiles rather than from filenames
module
components/cam/src/control/cam_initfiles.F90
. add public data for ncdata and bnd_topo (moved from filenames module)
. add routine cam_initfiles_readnl
. add namelist variable use_topo_file. Setting
use_topo_file=.false. implies no topo file will be read and PHIS, SGH,
SGH30, and LANDM_COSLAT will all be set to zero.
components/cam/src/control/filenames.F90
. move public data for ncdata and bnd_topo to cam_initfiles
components/cam/src/control/runtime_opts.F90
. remove ncdata and bnd_topo
. remove tracers_flag and add call to tracers_readnl
. add call to cam_initfiles_readnl
components/cam/src/control/startup_initialconds.F90
. remove pio filehandles as actual args in call to read_inidat
components/cam/src/dynamics/eul/eul_control_mod.F90
. rename kmnhd4 -> kmnhdn to reflect Nth order diffusion. Also make it
a parameter set in this module.
. change namelist variable names by prepending "dyn_spectral_" to dif2,
divdampn, eps, kmxhdc
. add new namelist variables eul_hdif_order and eul_hdif_coef. These
replace the old dif4 with variables to set both the order (N) of the
diffusion operator, and the coefficient for the Nth order diffusion.
. remove hardcoded values of namelist variables. In the old code there was
a bug where hardcoded values of dif2 and dif4 would overwrite user
specified values.
. move the log output inside a masterproc conditional
components/cam/src/dynamics/eul/grcalc.F90
. rename hdfst4 -> hdfstn to reflect implementation of general Nth order
diffusion
components/cam/src/dynamics/eul/hdinti.F90
. move setting of kmnhdn and kmxhd2 to eul_control_mod
. rename hdfst4 -> hdfstn and hdfsd4 -> hdfsdn to reflect implementation of
Nth order diffusion.
. compute coefs for Nth order diffusion. When N=4 there will be roundoff
level diffs with the coefficient values from the original 4th order calc.
components/cam/src/dynamics/eul/hordif.F90
. rename kmnhd4 -> kmnhdn, hdfst4 -> hdfstn, hdfsd4 -> hdfsdn
components/cam/src/dynamics/eul/inidat.F90
. use accessor methods to get ncid_ini and ncid_topo pointers. remove
dummy args change names to fh_ini and fh_topo.
. add check that fh_topo is not associated to the conditional that sets
phis=0.0
. fix log output for constituent initialization so that it's not written for
every latitude slice
. add calls to allow unicon to initialize its constituents
components/cam/src/dynamics/eul/spegrd.F90
. rename kmnhd4 -> kmnhdn
components/cam/src/dynamics/fv/inidat.F90
components/cam/src/dynamics/se/inidat.F90
components/cam/src/dynamics/sld/inidat.F90
. use accessor methods to get ncid_ini and ncid_topo pointers. remove
dummy args. change names to fh_ini and fh_topo.
. add check that fh_topo is not associated to the conditional that sets
phis=0.0
. add calls to allow unicon to initialize its constituents
components/cam/src/dynamics/sld/sld_control_mod.F90
. change namelist variable names by prepending "dyn_spectral_"
. remove hardcoded values of namelist variables.
. move the log output inside a masterproc conditional
components/cam/src/physics/cam/physpkg.F90
. remove calls to tracers_timestep_tend and tracers_timestep_init
. extend conditional that sets sgh=sgh30=landm=0 to also check whether
fh_topo is associated.
components/cam/src/physics/cam/tracers.F90
. add routine tracers_readnl to read namelist
. modify tracers_register and tracers_implements_cnst to allow the names to
come from either the test_tracer_names input or from the tracers_suite.
. remove unused functionality for non-passive tracers -- i.e., eliminate
routines tracers_timestep_init and tracers_timestep_tend
components/cam/src/physics/cam/tracers_suite.F90
. remove unused functionality for non-passive tracers
components/cam/test/system/nl_files/ttrac
. replace divdampn by dyn_spectral_divdampn
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all PASS except
004 bl132 TBL.sh e48c4dh ghgrmp+1850_cam4 9s ..................................................FAIL! rc= 7 at Fri Jun 26 14:08:18 MDT 2015
006 bl133 TBL.sh e48c4paqdm aqpgro+aquaplanet_cam4 3s .........................................FAIL! rc= 7 at Fri Jun 26 14:08:43 MDT 2015
goldbach/nag: all PASS except
005 bl111 TBL.sh e8c5t5mdm ttrac 9s ...........................................................FAIL! rc= 7 at Thu Jun 25 17:54:11 MDT 2015
010 bl112 TBL.sh e8c5dm ghgrmp 9s .............................................................FAIL! rc= 7 at Thu Jun 25 18:03:17 MDT 2015
014 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Thu Jun 25 18:08:48 MDT 2015
goldbach/pgi: all PASS except
005 bl111 TBL.sh e8c5t5mdm ttrac 9s ...........................................................FAIL! rc= 7 at Fri Jun 26 12:43:47 MDT 2015
010 bl112 TBL.sh e8c5dm ghgrmp 9s .............................................................FAIL! rc= 7 at Fri Jun 26 12:50:33 MDT 2015
016 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Fri Jun 26 12:58:03 MDT 2015
019 bl115 TBL.sh e8idm idphys 9s ..............................................................FAIL! rc= 7 at Fri Jun 26 13:02:10 MDT 2015
All failed baselines are due to a roundoff difference in the Eulerian 4th
order horizontal diffusion tendency which is computed using an algorithm
that has been generalized to work for Nth order diffusion rather than be
fixed at 4th order diffusion.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB except roundoff diffs in Eulerian dycore.
===============================================================
==============================================================
Tag name: cam5_3_91
Originator(s): cacraig
Date: 06/11/15
One-line Summary: Bug fix for is_convtran1 with MG1; added hobart regression testing;
limit calculations on check_energy_chng call in tphysbc to (:ncol); added more robust
subcolumn restart test
Purpose of changes:
- A bug in micro_mg_cam when calling cnst_add incorrectly with is_convtran1=.false. was discovered a number
of months ago. As it was answer changing for MG1, applying the correction was delayed until this tag.
- Added capabilities for using hobart (the next generation CGD cluster) to the CAM regression testing platforms
- A bug was discovered with a call to check_energy_chng with arithmetic calculations which were being done
over the entire array instead of the ncol subsection. In certain instances, this caused a floating point
error to occur during calculations outside the valid range.
- Discovered an error with filter and weights not being read in correctly during restart. Added a more robust
test (subcol_tstcp_restart) to test this capability. Regression test 337 has been added, but will not be added
to a pretag test list until the cam_history tag is created which corrects the bug.
Bugs fixed (include bugzilla ID): no bugzilla reports for any of the bugs listed above
Describe any changes made to build system: added hobart capability to regression testing
Describe any changes made to the namelist: added subcolumn namelist flag subcol_test_restart which turns
on a more robust restart test for weights
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do:
A components/cam/test/system/tests_pretag_hobart_nag
A components/cam/test/system/tests_pretag_hobart_pgi
- hobart pretag tests are identical to the current goldbach pretag tests
List all existing files that have been modified, and describe the changes:
M components/cam/bld/build-namelist
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
M components/cam/bld/namelist_files/namelist_definition.xml
- add subcolumn namelist variable subcol_tstcp_restart which tests restart on weights more robustly
M components/cam/src/physics/cam/micro_mg_cam.F90
- removed bug in cnst_add call which used is_convtran1=.false. instead of .true. for MG1 runs
As it was answer changing, fixing this bug was delayed until this tag
M components/cam/src/physics/cam/physpkg.F90
- bug discovered when calling check_energy_chng where an entire array is divided by cld_macmic_num_steps
In this instance a Nan was at the end of the array past the ncol which cause a floating point error
M components/cam/src/physics/cam/subcol_tstcp.F90
- introduced subcol_tstcp_restart which is designed to setup weights in such a way that if they
are incorrectly unpacked during restart, an exact restart test will fail
M components/cam/test/system/input_tests_master
- added test 337 to turn on subcol_tstcp_restart. It will be used in cam_history branch testing and
will be added to a pretag test when the updated cam_history is brought onto the trunk
M components/cam/test/system/CAM_runcmnd.sh
M components/cam/test/system/test_driver.sh
- added hobart capabilities to CAM regression testing
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
NOTE -- All of these tests were BFB with a set of regression tests that were run without the answer changing micro_mg_cam change. The
results below only failed when the micro_mg_cam bug fix was put into the code base.
yellowstone/intel:
012 bl322 TBL.sh f1.9c5nonedh outfrq3s_lb0 9s .................................................FAIL! rc= 7 at Thu Jun 11 14:04:45 MDT 2015
024 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...................................................FAIL! rc= 7 at Thu Jun 11 14:27:54 MDT 2015
034 er366 TER.sh f1.9c5clbdh outfrq3s_clubb 4+5s ..............................................FAIL! rc= 6 at Thu Jun 11 15:29:27 MDT 2015
041 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...............................................FAIL! rc= 7 at Thu Jun 11 15:50:20 MDT 2015
054 bl421 TBL.sh f1.9c5mam4dh outfrq3s 9s .....................................................FAIL! rc= 7 at Thu Jun 11 16:19:12 MDT 2015
056 bl430 TBL.sh f1.9c5mzmam3dh outfrq3s_NEUwetdep 9s .........................................FAIL! rc= 7 at Thu Jun 11 16:28:14 MDT 2015
062 bl474 TBL.sh f1.9c5wtmam3h outfrq3s+1850_waccm_tsmlt_cam5 9s ..............................FAIL! rc= 7 at Thu Jun 11 16:56:47 MDT 2015
065 bl482 TBL.sh f1.9c5wscmam3dh outfrq3s_2005+waccm_sc_1955-2005_cam5 9s .....................FAIL! rc= 7 at Thu Jun 11 17:20:48 MDT 2015
081 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ........................................FAIL! rc= 7 at Thu Jun 11 18:00:32 MDT 2015
086 bl740 TBL.sh h30c5h outfrq3s_lb0 9s .......................................................FAIL! rc= 7 at Thu Jun 11 18:05:05 MDT 2015
goldbach/nag:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s ...........................................................FAIL! rc= 7 at Thu Jun 11 11:15:52 MDT 2015
010 bl112 TBL.sh e8c5dm ghgrmp 9s .............................................................FAIL! rc= 7 at Thu Jun 11 11:24:11 MDT 2015
014 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Thu Jun 11 11:31:32 MDT 2015
018 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Thu Jun 11 11:43:00 MDT 2015
031 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .........................................FAIL! rc= 7 at Thu Jun 11 12:08:27 MDT 2015
040 bl325 TBL.sh f10c5t5mdmsc outfrq3s_subcol 9s ..............................................FAIL! rc= 7 at Thu Jun 11 12:40:19 MDT 2015
053 bl734 TBL.sh h16c5naqdm outfrq3s_bam+aquaplanet_cam5 9s ...................................FAIL! rc= 7 at Thu Jun 11 14:07:41 MDT 2015
056 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Thu Jun 11 14:19:21 MDT 2015
062 bl471 TBL.sh f10c5wmam3dm outfrq3s_modalstrat 9s ..........................................FAIL! rc= 7 at Thu Jun 11 14:47:44 MDT 2015
065 bl805 TBL.sh f4c5carmtt1dm outfrq3s_carma 9s ..............................................FAIL! rc= 7 at Thu Jun 11 15:18:01 MDT 2015
goldbach/pgi or jaguar/pgi:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s ...........................................................FAIL! rc= 7 at Thu Jun 11 11:15:16 MDT 2015
010 bl112 TBL.sh e8c5dm ghgrmp 9s .............................................................FAIL! rc= 7 at Thu Jun 11 11:23:00 MDT 2015
012 bl113 TBL.sh e8c5paqdm aqpgro+aquaplanet_cam5 3s ..........................................FAIL! rc= 7 at Thu Jun 11 11:28:10 MDT 2015
016 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Thu Jun 11 11:29:50 MDT 2015
016 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Thu Jun 11 11:29:50 MDT 2015
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Thu Jun 11 11:42:50 MDT 2015
038 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .........................................FAIL! rc= 7 at Thu Jun 11 12:17:03 MDT 2015
041 bl317 TBL.sh f10c5cdm outfrq3s 9s .........................................................FAIL! rc= 7 at Thu Jun 11 12:26:16 MDT 2015
045 bl319 TBL.sh f10c5nonedm outfrq3s_bam 9s ..................................................FAIL! rc= 7 at Thu Jun 11 12:34:10 MDT 2015
048 bl320 TBL.sh f10c5nonedm rad_diag 9s ......................................................FAIL! rc= 7 at Thu Jun 11 12:35:55 MDT 2015
051 bl321 TBL.sh f10c5cdm atrain 9s ...........................................................FAIL! rc= 7 at Thu Jun 11 12:38:26 MDT 2015
060 bl420 TBL.sh f4c5mam7dm outfrq3s 9s .......................................................FAIL! rc= 7 at Thu Jun 11 13:40:11 MDT 2015
063 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Thu Jun 11 13:53:05 MDT 2015
066 bl734 TBL.sh h16c5naqdm outfrq3s_bam+aquaplanet_cam5 9s ...................................FAIL! rc= 7 at Thu Jun 11 14:20:21 MDT 2015
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: all MG1
- what platforms/compilers: all
- nature of change (roundoff; larger than roundoff but same climate; new
climate): larger than roundoff but same climate
If bitwise differences were observed, how did you show they were no worse
than roundoff?
A 5 year run using the F_2000_CAM5 compset and MG1 was run for both a baseline as well as a
run with the is_convtran1=.true. fixed. The results were evaluated by Andrew Gettelman and
his determination was that the change was larger than roundoff, but is within the noise.
===============================================================
===============================================================
Tag name: cam5_3_90
Originator(s): cacraig, eaton
Date: Fri Jun 5 11:01:33 MDT 2015
One-line Summary: Add "-phys cam5.4" to configure
Purpose of changes:
Add "-phys cam5.4" option to configure for easy setup of the cam5.4 physics
package. This has the following effect:
. build configuration:
- trop_mam4 chemistry
- mg2 microphysics
- 32 vertical levels
- nucleate_ice.F90 contained a conditional clause that was previously turned
on by defining the CPP macro USE_XLIU_MOD. This version of the
conditional has been validated and is now the default code. Note that
this is an answer changing mod for the cam5 physics package as well as
being the cam5.4 configuration.
. runtime configuration:
- use_preexisting_ice = .true.
use_hetfrz_classnuc = .true.
cldfrc_rhminl_adj_lan = 0.00D0
modal_accum_coarse_exch = .true.
micro_mg_dcs = 250.D-6
nucleate_ice_subgrid =1.2D0
- note that modal_accum_coarse_exch=.true. triggers the tuning parameter
seasalt_emis_scale = 0.9D0
- There are currently only initial files for FV 1 and 2 deg:
cami-mam3_0000-01-01_1.9x2.5_L32_c150407.nc
cami-mam3_0000-01-01_0.9x1.25_L32_c141031.nc
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system:
. add cam5.4 as valid physics package
Describe any changes made to the namelist:
. provide correct defaults for cam5.4 physics package
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes:
. don't have performance estimates for cam5.4 at this time.
Code reviewed by: cacraig, eaton
List all subroutines eliminated:
components/cam/test/system/nl_files/outfrq3s_cam5.4
. remove this file and use outfrq3s in the cam5.4 tests. The namelist
customizations in this file for cam5.4 are no longer needed.
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
components/cam/bld/configure
. change default physics package to cam5.4
- remove cam3_5_1 as an alias for cam4
. set default cam5.4 chem package to trop_mam4
. change logic so defaults_waccm.xml is used with cam4 physics, else
use defaults_waccm5.xml
. set default cam5.4 microphys_pkg to mg2
. remove old warning about cosp/cam5 not tested
. set nlev values for non-waccm configurations
. remove old code setting ice_pkg='cice' for cam3 (cice is the default)
components/cam/bld/config_files/definition.xml
. remove phys value cam3_5_1
. add cam5.4 as valid value for phys parameter
. remove default value for nlev - set it in configure
components/cam/bld/build-namelist
. add add_default calls for use_hetfrz_classnuc and use_preexisting_ice
. add add_default call for modal_accum_coarse_exch (only for modal aerosols)
. mods to support defaults that haven't changed between cam5 and cam5.4
components/cam/bld/namelist_files/namelist_defaults_cam.xml
. add defaults for use_hetfrz_classnuc and use_preexisting_ice
. add defaults for modal_accum_coarse_exch
. add new cam5.4 default for cldfrc_rhminl_adj_land
. mods to support defaults that haven't changed between cam5 and cam5.4
. mg2 value of nucleate_ice_subgrid changed from 1.0 to 1.2
. mg2 value of micro_mg_dcs changed from 150.D-6 to 250.D-6
components/cam/src/chemistry/aerosol/wetdep.F90
components/cam/src/physics/cam/zm_conv_intr.F90
components/cam/src/physics/cam/cospsimulator_intr.F90
components/cam/src/physics/carma/cam/carma_intr.F90
. add cam_physpkg_is('cam5.4') to conditionals using cam_physpkg_is('cam5')
components/cam/src/physics/cam/nucleate_ice.F90
. use the form of the condition statement that results from defining the
CPP USE_XLIU_MOD. Make this the only option and remove the macro.
components/cam/test/system/config_files/f0.9c5.4dh
components/cam/test/system/config_files/f1.9c5.4dh
components/cam/test/system/config_files/f1.9c5.4dm
. "-phys cam5 -microphys mg2 -chem trop_mam4 -nlev 32" replaced by
"-phys cam5.4"
. remove "-cppdefs -DUSE_XLIU_MOD"
components/cam/test/system/input_tests_master
. change outfrq3s_cam5.4 to outfrq3s in all cam5.4 tests
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS except:
012 bl322 TBL.sh f1.9c5nonedh outfrq3s_lb0 9s .................................................FAIL! rc= 7 at Thu Jun 4 16:02:55 MDT 2015
024 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...................................................FAIL! rc= 7 at Thu Jun 4 16:25:34 MDT 2015
032 bl353 TBL.sh f1.9c5.4dh outfrq3s 9s .......................................................FAIL! rc= 7 at Thu Jun 4 16:48:33 MDT 2015
035 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Thu Jun 4 17:01:25 MDT 2015
041 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...............................................FAIL! rc= 7 at Thu Jun 4 17:19:26 MDT 2015
054 bl421 TBL.sh f1.9c5mam4dh outfrq3s 9s .....................................................FAIL! rc= 7 at Thu Jun 4 17:48:38 MDT 2015
056 bl430 TBL.sh f1.9c5mzmam3dh outfrq3s_NEUwetdep 9s .........................................FAIL! rc= 7 at Thu Jun 4 17:57:39 MDT 2015
062 bl474 TBL.sh f1.9c5wtmam3h outfrq3s+1850_waccm_tsmlt_cam5 9s ..............................FAIL! rc= 7 at Thu Jun 4 18:26:06 MDT 2015
065 bl482 TBL.sh f1.9c5wscmam3dh outfrq3s_2005+waccm_sc_1955-2005_cam5 9s .....................FAIL! rc= 7 at Thu Jun 4 18:50:06 MDT 2015
081 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ........................................FAIL! rc= 7 at Thu Jun 4 19:30:09 MDT 2015
086 bl740 TBL.sh h30c5h outfrq3s_lb0 9s .......................................................FAIL! rc= 7 at Thu Jun 4 19:34:41 MDT 2015
goldbach/nag: All PASS except:
018 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Thu Jun 4 12:31:29 MDT 2015
040 bl325 TBL.sh f10c5t5mdmsc outfrq3s_subcol 9s ..............................................FAIL! rc= 7 at Thu Jun 4 13:41:39 MDT 2015
047 bl354 TBL.sh f1.9c5.4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Thu Jun 4 14:29:23 MDT 2015
050 bl368 TBL.sh f1.9c5mg2dm outfrq3s_macmic2_classnuc 9s .....................................FAIL! rc= 7 at Thu Jun 4 15:10:54 MDT 2015
053 bl734 TBL.sh h16c5naqdm outfrq3s_bam+aquaplanet_cam5 9s ...................................FAIL! rc= 7 at Thu Jun 4 15:36:14 MDT 2015
056 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Thu Jun 4 15:51:26 MDT 2015
062 bl471 TBL.sh f10c5wmam3dm outfrq3s_modalstrat 9s ..........................................FAIL! rc= 7 at Thu Jun 4 16:21:59 MDT 2015
065 bl805 TBL.sh f4c5carmtt1dm outfrq3s_carma 9s ..............................................FAIL! rc= 7 at Thu Jun 4 16:56:11 MDT 2015
goldbach/pgi: All PASS except:
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Thu Jun 4 12:24:24 MDT 2015
041 bl317 TBL.sh f10c5cdm outfrq3s 9s .........................................................FAIL! rc= 7 at Thu Jun 4 13:11:57 MDT 2015
045 bl319 TBL.sh f10c5nonedm outfrq3s_bam 9s ..................................................FAIL! rc= 7 at Thu Jun 4 13:19:18 MDT 2015
048 bl320 TBL.sh f10c5nonedm rad_diag 9s ......................................................FAIL! rc= 7 at Thu Jun 4 13:21:10 MDT 2015
051 bl321 TBL.sh f10c5cdm atrain 9s ...........................................................FAIL! rc= 7 at Thu Jun 4 13:24:34 MDT 2015
054 bl334 TBL.sh f10c5udm outfrq3s 9s .........................................................FAIL! rc= 7 at Thu Jun 4 13:33:16 MDT 2015
057 bl354 TBL.sh f1.9c5.4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Thu Jun 4 14:10:00 MDT 2015
060 bl420 TBL.sh f4c5mam7dm outfrq3s 9s .......................................................FAIL! rc= 7 at Thu Jun 4 14:25:23 MDT 2015
063 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Thu Jun 4 14:38:53 MDT 2015
066 bl734 TBL.sh h16c5naqdm outfrq3s_bam+aquaplanet_cam5 9s ...................................FAIL! rc= 7 at Thu Jun 4 15:04:53 MDT 2015
All failures above are expected as explained below.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers:
BFB except cam5 and cam5.4 physics configurations:
. All cam5 configurations fail baselines due to making the USE_XLIU_MOD
form of a conditional in nucleate_ice.F90 the default. These are larger
than roundoff answer changes, but same climate. That validation is here:
http://webext.cgd.ucar.edu/FCLIMO/f.e13.F2000C5.f09_f09.cam5.3.51.ice_nucleation.005/atm/f.e13.F2000C5.f09_f09.cam5.3.51.ice_nucleation.005-f.e13.F2000C5.f09_f09.cam5.3.51.ice_nucleation.002/
. All cam5.4 configurations fail baselines due to changes in tuning
parameters. This is part of the cam5.4 development process.
===============================================================
===============================================================
Tag name: cam5_3_89
Originator(s): eaton
Date: Fri May 29 09:31:42 MDT 2015
One-line Summary: update cime and cice externals
Purpose of changes:
. Update to cime1.0.8. This tag contains an updated MCT external which
provides the mpi-serial library. This change required several
compensating changes in the CAM standalone build, and also an update of
the cice external to deal with compilation issues (NAG) stemming from the
new mpi-serial version of mpif.h.
Bugs fixed (include bugzilla ID):
. Fix problem in the dependency generation caused by an include path being
added directly to Makefile.in rather than via the Filepath file.
Describe any changes made to build system:
. Changes to support MCT's new mpi-serial build.
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
SVN_EXTERNAL_DIRECTORIES
. update to cime1.0.8
. update cice to cice/branch_tags/cice5_tags/cice5_20150402_20150528
- The new cice tag was needed to get fixes that removed global save
statements from cice/src/mpi/ice_boundary.F90 and
cice/src/mpi/ice_global_reductions.F90. These save statements
conficted with a new save statement in the mpi-serial version of
mpif.h. This is a problem for the NAG compiler only.
components/cam/bld/configure
. in mct configure section
- add copies for the mpif.h and mpi.h files
. add $cam_root/cime/share/csm_share/include to the Filepath file
- this is a bug fix required for proper dependency generation of the
shr_assert.h file (it contains a 'use shr_assert_mod' statement and
that dependency is needed to force correct build sequencing)
components/cam/bld/Makefile.in
. in rule to build mct libs
- add SRCDIR=$(ROOTDIR)/cime/externals/mct to submake commandline. This
is compensating for a bug in the MCT configure. The base directory of
the MCT source code has already been provided to MCT's configure, and
it should not have to be passed again via the make command.
. don't manually append $cam_root/cime/share/csm_share/include to
cpp_path. It's coming from the Filepath file now.
components/cam/src/utils/spmd_utils.F90
. remove the declaration of mpi_status_ignore. The new mpi-serial version
of mpif.h contains this declaration.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS
goldbach/nag: All PASS
goldbach/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB.
===============================================================
===============================================================
Tag name: cam5_3_88
Originator(s): cacraig, andrew, sean, fvitt, jenkay
Date: May 22, 2015
One-line Summary: Miscellaneous chemistry, cosp and general CAM bug fixes
and modifications
Purpose of changes:
- added output fields to the microphysics for supercooled liquid water
- new variables QTGW, CLDLIQTGW, and CLDICETGW, as well as the existing variable CT_H2O,
are added to the budget history file. This is necessary to balance the budget for
water when active chemistry and the parameterized gravity waves are enabled.
- fix bug in CESM build of chemistry preprocessor
- fix bug in chemistry preprocessor when the list of non-transported species
is long
- fix bug in interp_mod for SE dycore
- removed the need to pass dry deposition velocity fields of tagged CO
species through the model coupler -- this helps alleviate a problem which
occasionally occurs when the list of dry deposition of velocity fields
and MEGAN emission fields to transfer from CLM to CAM has gotten to
long for the model coupler to handle
- bug fix to ice_nucleate to check qi for non-zero value to prevent divide by zero
- bug fixes for cosp
- CLD_CAL_LIQ, CLD_CAL_ICE, CLD_CAL_UN,CLD_CAL_TMP, CLD_CAL_TMPLIQ,
CLD_CAL_TMPICE, CLD_CAL_TNPUN) for COSP1.4 should all be dimensioned the same
as the original CALIPSO variable CLD_CAL. CLD_CAL uses nht_cosp as a vertical
size and cosp_ht as the mdim variable. These new output fields are all being
saved on the COSP grid - not the CAM grid.
- moved the outfld calls inside of the if(llidar_sim) conditional.
- removed the adddefault calls for CLD_CAL_TMP, CLD_CAL_TMPLIQ,CLD_CAL_TMPICE,
CLD_CAL_TNPUN as they are temporary variables and unclear how they would
be used for model evaluation.
- removed the comment "fails check_accum if 'A'" where it doesn't belong. This
comment is meant to be only for COSP outputs with sub-column dimensions (e.g., TAU_ISCCP).
You cannot save COSP outputs with sub-columns as 'A' fields - they have to be 'I'. This
comment "fails check_accum if 'A'" is not applicable for CLD_CAL_LIQ, CLD_CAL_ICE,
CLD_CAL_UN,CLD_CAL_TMP, CLD_CAL_TMPLIQ,CLD_CAL_TMPICE, CLD_CAL_TNPUN.
- PGI15.1 had a compiler bug which requires associate subsections to have ranges on them
- updated goldbach PGI testing to use PGI15.1
Bugs fixed (include bugzilla ID):
- bug fixes for cosp
- bug fix for nucleate_ice which had a divide by zero when qi was zero (answer, but not climate changing for cam5.4 setup)
- components/cam/src/dynamics/se/interp_mod.F90
- bug fix in CESM build of chemistry preprocessor
- bug fix in chemistry preprocessor when the list of non-transported species is long
- bug fix in interp_mod for SE dycore
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/test/system/test_driver.sh
- update to use PGI15.1 on goldbach
M components/cam/src/physics/cam/microp_aero.F90
M components/cam/src/physics/cam/hetfrz_classnuc_cam.F90
- PGI15.1 has a compiler bug which requires associate commands to have ranges on the subsections
M components/cam/src/physics/cam/micro_mg_cam.F90
- added output fields to the microphysics for supercooled liquid water
M components/cam/src/physics/cam/gw_drag.F90
- added new variables QTGW, CLDLIQTGW, and CLDICETGW, as well as the existing variable CT_H2O,
to the budget history file. This is necessary to balance the budget for
water when active chemistry and the parameterized gravity waves are enabled.
M components/cam/src/physics/cam/nucleate_ice.F90
- bug fix to ice_nucleate to check qi for non-zero value to prevent divide by zero
M components/cam/src/physics/cam/cospsimulator_intr.F90
- bug fixes for cosp (see details above)
M components/cam/src/chemistry/mozart/chemistry.F90
- added CT_H2O, to the budget history file.
M components/cam/src/dynamics/se/interp_mod.F90
- bug fix from Francis
M SVN_EXTERNAL_DIRECTORIES
- using cime external from github (still the same version as previously)
M components/cam/SVN_EXTERNAL_DIRECTORIES
- use updated chemistry preprocessor tag which fixes a bug in not-transported species
- update cosp to a version which works with WACCM
M components/cam/bld/cam.buildnml
- invoke fc_type configure option so that chemistry preprocessor will be
built with correct compiler
M components/cam/bld/perl5lib/Build/ChemPreprocess.pm
- replaced setting of 'COMPILER' environment variable with 'USER_FC'
-- 'COMPILER' is set in CESM scripts and don't want to reset it here
M components/cam/test/system/tests_pretag_yellowstone
- moved CAM test 376 to be after CESM test ne996 -- this is so that
the chemistry preprocessor will be built with CESM scripts rather
than by stand-alone CAM configure
M components/cam/src/chemistry/mozart/mo_drydep.F90
- cleaned up tagged CO code and fixed bug in setting deposition fluxes
of the tagged CO species
M components/cam/bld/namelist_files/master_gas_drydep_list.xml
- removed tagged CO species so that these deposition velocity fields will
not be passed through the coupler which are duplicates of CO deposition
velocity fields -- only deposition velocity field of CO is needed
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB except:
032 bl353 TBL.sh f1.9c5.4dh outfrq3s_cam5.4 9s ................................................FAIL! rc= 7 at Fri May 22 14:47:08 MDT 2015
goldbach/nag: all BFB except:
047 bl354 TBL.sh f1.9c5.4dm outfrq3s_cam5.4 9s ................................................FAIL! rc= 7 at Fri May 22 13:52:16 MDT 2015
goldbach/pgi or jaguar/pgi: all BFB except:
057 bl354 TBL.sh f1.9c5.4dm outfrq3s_cam5.4 9s ................................................FAIL! rc= 7 at Fri May 22 13:56:16 MDT 2015
Summarize any changes to answers, i.e.,
- what code configurations: cam5.4
- what platforms/compilers: all
- nature of change (roundoff; larger than roundoff but same climate; new
climate): larger than roundoff, but same climate
===============================================================
===============================================================
Tag name: cam5_3_87
Originator(s): eaton
Date: Fri May 1 07:43:51 MDT 2015
One-line Summary: cleanup in atm_comp, cam_comp, time_manager
Purpose of changes:
. Move initialization code that has been duplicated in atm_comp_mct and
atm_comp_esmf down to the cam_comp layer.
. Add initializer for module data in cam_control_mod and add protected
attribute. This is the location of control info from the
driver/coupler.
. Remove restart functionality from the time_manager module. The
driver/coupler is responsible for providing this information to
components.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: cam developers
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
components/cam/SVN_EXTERNAL_DIRECTORIES
. update COSP -- fixes a missing build dependency
components/cam/bld/Makefile.in
. fix for configure's esmf link test
components/cam/bld/configure
. fix for linking with external ESMF lib (need to build driver code with
USE_ESMF_LIB defined.
. fix for configure's esmf link test
components/cam/src/chemistry/bulk_aero/mo_setsoa.F90
components/cam/src/chemistry/mozart/spedata.F90
components/cam/src/chemistry/utils/tracer_data.F90
. replace obsolete nsrest variable (or remove unused variable)
components/cam/src/control/cam_comp.F90
. remove cam specific timing code
. move cam specific initialization that had been done in atm_comp layer
into cam_init.
. data from the coupler is passed to cam_init via intent(in) args.
. Pass the Eclock object and access it's data inside cam_init rather than
passing the timestamp data as individual arguments.
. change name of module variable "dtime" to "dtime_phys" since this is not
the same dtime that cam reads from its namelist and uses to set its
internal clock timestep. dtime_phys is set by the dycore (intent(out)
arg of stepon_run1) and is used by the physics package as the timestep
for computed tendencies which are returned to the dycore.
components/cam/src/control/cam_control_mod.F90
. add protected attribute to public module variables
. add init routine for public module variables -- these are set by
information coming from the coupler
. add log output for the variables in this module (moved here from
read_namelist)
components/cam/src/control/cam_history.F90
. replace obsolete nsrest variable (or remove unused variable)
. access caseid and ctitle from cam_control_mod
components/cam/src/control/cam_restart.F90
. replace obsolete nsrest variable (or remove unused variable)
. remove call for time_manager restart
. remove obsolete reduced grid code
components/cam/src/control/filenames.F90
. caseid and brnch_retain_casename moved to cam_control_mod. These are
given to cam by the coupler.
components/cam/src/control/runtime_opts.F90
. dtime made local; version in time_manager mod made private
. single_column, scmlat, scmlon removed as module data since these are
given by the coupler and passed as intent(in) args to read_namelist
. remove check on caseid; this comes from the coupler and is handled in
cam_control_mod
. add call to timemgr_set_step_size to set dtime in time_manager
. move log output that belongs in cam_control_mod
. remove obsolete COUP_SOM code
. remove mpibcast of variables handled by cam_control_mod -- these weren't
doing anything anyway since the variable came from the coupler and were
already known in all tasks.
components/cam/src/control/scamMod.F90
. added a dycore_is check for EUL since scam_setopts is always called by
read_namelist rather than only being called when scam is on
components/cam/src/cpl/atm_comp_esmf.F90
components/cam/src/cpl/atm_comp_mct.F90
. replace the direct initialization of public module data by subroutine
calls that initialize protected module data. This is for the data that
comes from the coupler.
. pass the Eclock object to cam_init and move time_manager init there
. move broken check on co2_readflux_ocn into co2_cycle
. remove time_manager restart
. remove call to scam_use_iop_srf -- move to phys_timestep_init since this
is code to read info from the iop dataset
components/cam/src/dynamics/eul/initcom.F90
. replace obsolete nsrest variable (or remove unused variable)
components/cam/src/dynamics/eul/scanslt.F90
. scanslt_initial -- change etamid to intent(out) -- this is not the same
etamid as in the hycoef module because SCAM may overwrite it with pmid
components/cam/src/dynamics/eul/stepon.F90
. make etamid a private module variable since SCAM may overwrite it's usual
values
. remove gw and etamid from stepon_init arg list
. remove etamid from stepon_run3 arg list
components/cam/src/dynamics/fv/metdata.F90
. replace obsolete nsrest variable (or remove unused variable)
components/cam/src/dynamics/fv/stepon.F90
. replace obsolete nsrest variable (or remove unused variable)
. remove gw and etamid from stepon_init arg list
. remove etamid from stepon_run3 arg list
components/cam/src/dynamics/se/dyn_comp.F90
. replace direct use of time_manager module var dtime by get_step_size
function
components/cam/src/dynamics/se/stepon.F90
. replace direct use of time_manager module var dtime by get_step_size
function
. remove gw and etamid from stepon_init arg list
. remove etamid from stepon_run3 arg list
components/cam/src/dynamics/sld/scanslt.F90
. slt_initial - remove gw from arg list
components/cam/src/dynamics/sld/stepon.F90
. replace internal calc of etamid by use of hycoef var
. remove gw and etamid from stepon_init arg list
. remove etamid from stepon_run3 arg list
components/cam/src/physics/cam/co2_cycle.F90
. check on co2_readflux_ocn moved here from atm_comp
components/cam/src/physics/cam/physpkg.F90
. replace obsolete nsrest variable (or remove unused variable)
. add cam_in as actual arg in call to phys_timestep_init
. add cam_in as dummy arg in phys_timestep_init so can call
scam_use_iop_srf which is using data to set surface field values
components/cam/src/physics/carma/cam/carma_intr.F90
. replace obsolete nsrest variable (or remove unused variable)
components/cam/src/unit_drivers/drv_input_data.F90
. replace direct use of time_manager module var dtime by get_step_size
function
components/cam/src/utils/cam_dom/ocn_comp.F90
components/cam/src/utils/cam_dom/ocn_comp_mct.F90
. replace obsolete nsrest variable (or remove unused variable)
components/cam/src/utils/cam_dom/ocn_filenames.F90
. access caseid from cam_control_mod rather than filenames module
components/cam/src/utils/time_manager.F90
. add timemgr_set_step_size to set private dtime module variable with value
read from namelist
. timemgr_init - add current time value obtained from coupler, and extend
functionality to deal with either initial or restart cases.
. remove restart functionality -- that is driver/coupler responsibility
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS
goldbach/nag: All PASS
goldbach/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: none
===============================================================
===============================================================
Tag name: cam5_3_86
Originator(s): cacraig
Date: 4/30/2015
One-line Summary: Switch to cice5 external library
Purpose of changes:
- Use the new cice5 external library
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M .
M SVN_EXTERNAL_DIRECTORIES
M components/cam/bld/configure
- added cice5 library and directories
M components/cam/bld/build-namelist
- ice_ic file is required for branches and CAM needs to supply its name in the rpointer.ice file
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB except:
004 bl132 TBL.sh e48c4dh ghgrmp+1850_cam4 9s ..................................................FAIL! rc= 7 at Wed Apr 29 17:07:35 MDT 2015
009 bl203 TBL.sh fsd1.9c4wtsmlth outfrq3s_sd+sd_1975-2010_ccmi_refc1_waccm_tsmlt 9s ...........FAIL! rc= 7 at Wed Apr 29 17:30:40 MDT 2015
012 bl322 TBL.sh f1.9c5nonedh outfrq3s_lb0 9s .................................................FAIL! rc= 7 at Wed Apr 29 17:37:24 MDT 2015
016 bl326 TBL.sh f1.9c4wscdm outfrq3s_2005+waccm_sc_2005-2100_cam4_rcp45 9s ...................FAIL! rc= 7 at Wed Apr 29 17:51:03 MDT 2015
020 bl331 TBL.sh f4c4dh co2rmp+1850_cam4 9s ...................................................FAIL! rc= 7 at Wed Apr 29 17:54:00 MDT 2015
024 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...................................................FAIL! rc= 7 at Wed Apr 29 18:00:07 MDT 2015
029 bl336 TBL.sh f4c4wmdh outfrq3s+waccm_1850_cam4 9s .........................................FAIL! rc= 7 at Wed Apr 29 18:12:25 MDT 2015
032 bl353 TBL.sh f1.9c5.4dh outfrq3s_cam5.4 9s ................................................FAIL! rc= 7 at Wed Apr 29 18:23:13 MDT 2015
035 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Wed Apr 29 18:36:12 MDT 2015
039 bl373 TBL.sh f1.9c4mozdh outfrq3s+2000_cam4_trop_chem 9s ..................................FAIL! rc= 7 at Wed Apr 29 18:51:49 MDT 2015
041 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...............................................FAIL! rc= 7 at Wed Apr 29 18:54:23 MDT 2015
043 bl376 TBL.sh f1.9c4mozdh+testmech outfrq3s 9s .............................................FAIL! rc= 7 at Wed Apr 29 19:00:37 MDT 2015
046 bl374 TBL.sh f1.9c4fastdh outfrq3s+1850-2005_cam4_super_fast_llnl 9s ......................FAIL! rc= 7 at Wed Apr 29 19:07:20 MDT 2015
049 bl390 TBL.sh f1.9c4wmxh outfrq3s+waccmx_2000_cam4 9s ......................................FAIL! rc= 7 at Wed Apr 29 19:16:04 MDT 2015
053 bl397 TBL.sh f1.9c4cm outfrq3s_cosp+1850-2005_cam4 9s .....................................FAIL! rc= 7 at Wed Apr 29 19:19:50 MDT 2015
056 bl421 TBL.sh f1.9c5mam4dh outfrq3s 9s .....................................................FAIL! rc= 7 at Wed Apr 29 19:29:59 MDT 2015
058 bl430 TBL.sh f1.9c5mzmam3dh outfrq3s_NEUwetdep 9s .........................................FAIL! rc= 7 at Wed Apr 29 19:39:00 MDT 2015
061 bl440 TBL.sh f1.9c4soadh outfrq3s+soa_chem_megan_emis 9s ..................................FAIL! rc= 7 at Wed Apr 29 19:55:51 MDT 2015
064 bl474 TBL.sh f1.9c5wtmam3h outfrq3s+1850_waccm_tsmlt_cam5 9s ..............................FAIL! rc= 7 at Wed Apr 29 20:07:28 MDT 2015
067 bl482 TBL.sh f1.9c5wscmam3dh outfrq3s_2005+waccm_sc_1955-2005_cam5 9s .....................FAIL! rc= 7 at Wed Apr 29 20:31:17 MDT 2015
071 bl387 TBL.sh f1.9c4dm outfrq3s+1850-2005_cam4 9s ..........................................FAIL! rc= 7 at Wed Apr 29 20:37:57 MDT 2015
074 bl388 TBL.sh f1.9c4m outfrq24h+1850_cam4 2d ...............................................FAIL! rc= 7 at Wed Apr 29 20:39:27 MDT 2015
077 bl379 TBL.sh f1.9c4wmh outfrq24h+waccm_2005-2100_cam4_rcp45 2d ............................FAIL! rc= 7 at Wed Apr 29 20:56:49 MDT 2015
088 bl740 TBL.sh h30c5h outfrq3s_lb0 9s .......................................................FAIL! rc= 7 at Wed Apr 29 21:15:16 MDT 2015
goldbach/nag: all BFB except:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s ...........................................................FAIL! rc= 7 at Wed Apr 29 14:48:19 MDT 2015
010 bl112 TBL.sh e8c5dm ghgrmp 9s .............................................................FAIL! rc= 7 at Wed Apr 29 14:56:34 MDT 2015
014 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Wed Apr 29 15:02:32 MDT 2015
018 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Wed Apr 29 15:12:18 MDT 2015
028 bl314 TBL.sh f10c4wscdm outfrq3s 9s .......................................................FAIL! rc= 7 at Wed Apr 29 15:29:00 MDT 2015
034 bl318 TBL.sh f10c4cdm sat_hist 9s .........................................................FAIL! rc= 7 at Wed Apr 29 15:47:09 MDT 2015
037 bl324 TBL.sh f4c4rrtmgdm outfrq3s 9s ......................................................FAIL! rc= 7 at Wed Apr 29 15:53:49 MDT 2015
040 bl325 TBL.sh f10c5t5mdmsc outfrq3s_subcol 9s ..............................................FAIL! rc= 7 at Wed Apr 29 16:01:55 MDT 2015
047 bl354 TBL.sh f1.9c5.4dm outfrq3s_cam5.4 9s ................................................FAIL! rc= 7 at Wed Apr 29 16:36:46 MDT 2015
050 bl368 TBL.sh f1.9c5mg2dm outfrq3s_macmic2_classnuc 9s .....................................FAIL! rc= 7 at Wed Apr 29 17:06:13 MDT 2015
056 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Wed Apr 29 17:37:18 MDT 2015
059 bl424 TBL.sh f10c4wmxdm outfrq3s 9s .......................................................FAIL! rc= 7 at Wed Apr 29 17:50:20 MDT 2015
062 bl471 TBL.sh f10c5wmam3dm outfrq3s_modalstrat 9s ..........................................FAIL! rc= 7 at Wed Apr 29 18:03:05 MDT 2015
065 bl805 TBL.sh f4c5carmtt1dm outfrq3s_carma 9s ..............................................FAIL! rc= 7 at Wed Apr 29 18:27:51 MDT 2015
goldbach/pgi or jaguar/pgi: all BFB except:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s ...........................................................FAIL! rc= 7 at Wed Apr 29 14:48:27 MDT 2015
010 bl112 TBL.sh e8c5dm ghgrmp 9s .............................................................FAIL! rc= 7 at Wed Apr 29 14:54:49 MDT 2015
016 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Wed Apr 29 15:01:15 MDT 2015
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Wed Apr 29 15:13:08 MDT 2015
033 bl314 TBL.sh f10c4wscdm outfrq3s 9s .......................................................FAIL! rc= 7 at Wed Apr 29 15:33:02 MDT 2015
036 bl315 TBL.sh f10c3dm outfrq3s+cam3 9s .....................................................FAIL! rc= 7 at Wed Apr 29 15:38:39 MDT 2015
041 bl317 TBL.sh f10c5cdm outfrq3s 9s .........................................................FAIL! rc= 7 at Wed Apr 29 15:52:15 MDT 2015
045 bl319 TBL.sh f10c5nonedm outfrq3s_bam 9s ..................................................FAIL! rc= 7 at Wed Apr 29 15:58:52 MDT 2015
048 bl320 TBL.sh f10c5nonedm rad_diag 9s ......................................................FAIL! rc= 7 at Wed Apr 29 16:00:36 MDT 2015
051 bl321 TBL.sh f10c5cdm atrain 9s ...........................................................FAIL! rc= 7 at Wed Apr 29 16:02:45 MDT 2015
054 bl334 TBL.sh f10c5udm outfrq3s 9s .........................................................FAIL! rc= 7 at Wed Apr 29 16:10:46 MDT 2015
057 bl354 TBL.sh f1.9c5.4dm outfrq3s_cam5.4 9s ................................................FAIL! rc= 7 at Wed Apr 29 16:45:31 MDT 2015
060 bl420 TBL.sh f4c5mam7dm outfrq3s 9s .......................................................FAIL! rc= 7 at Wed Apr 29 16:58:40 MDT 2015
063 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Wed Apr 29 17:10:10 MDT 2015
All of the above changes are due to the change in cice and are establishing a new baseline for CAM.
===============================================================
===============================================================
Tag name: cam5_3_85
Originator(s): cbardeen
Date: 2015/04/29
One-line Summary: Limit values in RRTMG vulnerable to bad table extrapolation.
Purpose of changes:
- Some variables in RRTMG can have unrealistic values due to being extrapolated
far past the ranges provided in lookup tables. This produces problems such as
negative optical depths, so the RRTMG external is updated with limiters to
avoid extreme values.
- Fix a race condition in the COSP makefile.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: trivial, checked by santos
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/SVN_EXTERNAL_DIRECTORIES
- Update the RRTMG and COSP externals. RRTMG has the aforementioned
limiters added, while COSP has a dependency from scale_LUTs_io to
radar_simulator_types added to the Makefile.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
012 bl322 TBL.sh f1.9c5nonedh outfrq3s_lb0 9s .................................................FAIL! rc= 7 at Wed Apr 29 05:21:11 MDT 2015
024 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...................................................FAIL! rc= 7 at Wed Apr 29 05:43:53 MDT 2015
032 bl353 TBL.sh f1.9c5.4dh outfrq3s_cam5.4 9s ................................................FAIL! rc= 7 at Wed Apr 29 06:06:24 MDT 2015
035 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Wed Apr 29 06:19:06 MDT 2015
041 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...............................................FAIL! rc= 7 at Wed Apr 29 06:36:40 MDT 2015
056 bl421 TBL.sh f1.9c5mam4dh outfrq3s 9s .....................................................FAIL! rc= 7 at Wed Apr 29 07:10:29 MDT 2015
058 bl430 TBL.sh f1.9c5mzmam3dh outfrq3s_NEUwetdep 9s .........................................FAIL! rc= 7 at Wed Apr 29 07:19:28 MDT 2015
064 bl474 TBL.sh f1.9c5wtmam3h outfrq3s+1850_waccm_tsmlt_cam5 9s ..............................FAIL! rc= 7 at Wed Apr 29 07:47:21 MDT 2015
067 bl482 TBL.sh f1.9c5wscmam3dh outfrq3s_2005+waccm_sc_1955-2005_cam5 9s .....................FAIL! rc= 7 at Wed Apr 29 08:10:53 MDT 2015
083 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ........................................FAIL! rc= 7 at Wed Apr 29 08:50:11 MDT 2015
088 bl740 TBL.sh h30c5h outfrq3s_lb0 9s .......................................................FAIL! rc= 7 at Wed Apr 29 08:54:41 MDT 2015
goldbach/nag:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s ...........................................................FAIL! rc= 7 at Tue Apr 28 11:36:40 MDT 2015
010 bl112 TBL.sh e8c5dm ghgrmp 9s .............................................................FAIL! rc= 7 at Tue Apr 28 11:44:32 MDT 2015
014 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Tue Apr 28 11:50:27 MDT 2015
018 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Tue Apr 28 12:00:33 MDT 2015
031 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .........................................FAIL! rc= 7 at Tue Apr 28 12:21:04 MDT 2015
037 bl324 TBL.sh f4c4rrtmgdm outfrq3s 9s ......................................................FAIL! rc= 7 at Tue Apr 28 12:41:30 MDT 2015
040 bl325 TBL.sh f10c5t5mdmsc outfrq3s_subcol 9s ..............................................FAIL! rc= 7 at Tue Apr 28 12:48:50 MDT 2015
047 bl354 TBL.sh f1.9c5.4dm outfrq3s_cam5.4 9s ................................................FAIL! rc= 7 at Tue Apr 28 13:28:04 MDT 2015
050 bl368 TBL.sh f1.9c5mg2dm outfrq3s_macmic2_classnuc 9s .....................................FAIL! rc= 7 at Tue Apr 28 13:59:50 MDT 2015
053 bl734 TBL.sh h16c5naqdm outfrq3s_bam+aquaplanet_cam5 9s ...................................FAIL! rc= 7 at Tue Apr 28 14:22:41 MDT 2015
056 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Tue Apr 28 14:34:53 MDT 2015
062 bl471 TBL.sh f10c5wmam3dm outfrq3s_modalstrat 9s ..........................................FAIL! rc= 7 at Tue Apr 28 15:00:08 MDT 2015
065 bl805 TBL.sh f4c5carmtt1dm outfrq3s_carma 9s ..............................................FAIL! rc= 7 at Tue Apr 28 15:31:00 MDT 2015
068 bl011 TBL.sh f4c5portdm outfrq24h_port 2d .................................................FAIL! rc= 7 at Tue Apr 28 15:40:02 MDT 2015
goldbach/pgi or jaguar/pgi:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s ...........................................................FAIL! rc= 7 at Tue Apr 28 11:56:09 MDT 2015
010 bl112 TBL.sh e8c5dm ghgrmp 9s .............................................................FAIL! rc= 7 at Tue Apr 28 12:04:26 MDT 2015
012 bl113 TBL.sh e8c5paqdm aqpgro+aquaplanet_cam5 3s ..........................................FAIL! rc= 7 at Tue Apr 28 12:10:41 MDT 2015
016 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Tue Apr 28 12:12:51 MDT 2015
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Tue Apr 28 12:26:57 MDT 2015
038 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .........................................FAIL! rc= 7 at Tue Apr 28 13:02:42 MDT 2015
041 bl317 TBL.sh f10c5cdm outfrq3s 9s .........................................................FAIL! rc= 7 at Tue Apr 28 13:12:23 MDT 2015
045 bl319 TBL.sh f10c5nonedm outfrq3s_bam 9s ..................................................FAIL! rc= 7 at Tue Apr 28 13:20:30 MDT 2015
048 bl320 TBL.sh f10c5nonedm rad_diag 9s ......................................................FAIL! rc= 7 at Tue Apr 28 13:22:25 MDT 2015
051 bl321 TBL.sh f10c5cdm atrain 9s ...........................................................FAIL! rc= 7 at Tue Apr 28 13:25:43 MDT 2015
054 bl334 TBL.sh f10c5udm outfrq3s 9s .........................................................FAIL! rc= 7 at Tue Apr 28 13:35:23 MDT 2015
057 bl354 TBL.sh f1.9c5.4dm outfrq3s_cam5.4 9s ................................................FAIL! rc= 7 at Tue Apr 28 14:15:13 MDT 2015
060 bl420 TBL.sh f4c5mam7dm outfrq3s 9s .......................................................FAIL! rc= 7 at Tue Apr 28 14:31:18 MDT 2015
063 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Tue Apr 28 14:44:46 MDT 2015
066 bl734 TBL.sh h16c5naqdm outfrq3s_bam+aquaplanet_cam5 9s ...................................FAIL! rc= 7 at Tue Apr 28 15:15:08 MDT 2015
068 bl011 TBL.sh f4c5portdm outfrq24h_port 2d .................................................FAIL! rc= 7 at Tue Apr 28 15:23:23 MDT 2015
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
All RRTMG cases (mostly CAM5 and development configurations).
- what platforms/compilers:
All
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
larger than roundoff, same climate
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
Climate not changed, but note diags of FC5 and FW5 cases run on yellowstone:
http://webext.cgd.ucar.edu/FCLIMO/f.e13.FC5.f19_f19.rrtmglim.002/atm/f.e13.FC5.f19_f19.rrtmglim.002-f.e13.FC5.f19_f19.rrtmgctl.002
http://webext.cgd.ucar.edu/FCLIMO/f.e13.FW5.f19_f19.rrtmglim.002/atm/f.e13.FW5.f19_f19.rrtmglim.002-f.e13.FW5.f19_f19.rrtmgctl.002
===============================================================
===============================================================
Tag name: cam5_3_84
Originator(s): sungsup, jenkay, eaton
Date: Mon Apr 27 13:20:21 MDT 2015
One-line Summary: add UNICON kernel to trunk; COSP diagnostic fix
Purpose of changes:
. The UNICON kernel code being put on the trunk is the one used to produce
the simulations evaluated by the panel prior to the Feb 2015 AMWG
meeting. But note that those simulations were done from source code
based on cam5_3_25 with a modified version of MG1 code which is not on
CAM's trunk. Thus the simulations evaluated by the panel are not
reproducible in this tag. This is of course also a due to all the answer
changes that are being added for the candidate 5.4 physics package.
The UNICON code is enabled by setting the "-unicon" flag to configure.
. Bug fixes in the addfld calls for the following COSP diagnostic fields:
CLD_CAL_LIQ, CLD_CAL_ICE, CLD_CAL_UN,
CLD_CAL_TMP, CLD_CAL_TMPLIQ, CLD_CAL_TMPICE, CLD_CAL_TMPUN
Bugs fixed (include bugzilla ID):
. cosp diagnostics
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
components/cam/test/system/nl_files/fv2d_4tsk
components/cam/test/system/nl_files/fv2d_8tsk
. namelists for obsolete tests
List all subroutines added and what they do:
components/cam/src/physics/cam/unicon.F90
components/cam/src/physics/cam/unicon_utils.F90
. UNICON kernel
components/cam/test/system/config_files/f10c5udm
. new file for cam5/unicon config
List all existing files that have been modified, and describe the changes:
components/cam/src/physics/cam/cospsimulator_intr.F90
. fix addfld calls for CLD_CAL_LIQ, CLD_CAL_ICE, CLD_CAL_UN,
CLD_CAL_TMP, CLD_CAL_TMPLIQ, CLD_CAL_TMPICE, CLD_CAL_TMPUN
components/cam/src/physics/cam/unicon_cam.F90
. interface updates for addition of unicon_utils module
components/cam/test/system/input_tests_master
components/cam/test/system/tests_posttag_goldbach
components/cam/test/system/tests_pretag_goldbach_pgi
. add new unicon test and remove some obsolete tests
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS
goldbach/nag: All PASS
goldbach/pgi: All PASS except
054 bl334 TBL.sh f10c5udm outfrq3s 9s .........................................................FAIL! rc= 7 at Fri Apr 24 15:25:14 MDT 2015
bl334 is a new test with no previous baseline for comparison.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_3_83
Originator(s): bogensch, cacraig
Date: 04/24/15
One-line Summary: Updated to more recent CLUBB external library
Purpose of changes:
- Using next version of the CLUBB external library
- introduced several new tuning parameters for CLUBB: clubb_timestep, clubb_rnevap_effic, clubb_stabcorrect
The default values of these and other clubb tuning parameters may be changed in a future tag
- introduced microphysics logical allow_sed_supersat: allow supersaturated conditions after sedimentation loop
- Update some default namelist parameters for CLUBB in preparation for CAM5.4-CLUBB simulations.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
- added: clubb_stabcorrect (CLUBB logical - do explict diffusion with a stability correction)
clubb_timestep (CLUBB real - timestep)
clubb_rnevap_effic (CLUBB real - rain evaporation efficiency factor)
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: CAM software development team
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam
M components/cam/SVN_EXTERNAL_DIRECTORIES
- update to newer CLUBB external library
- brought in SILHS external library (needed for CLUBB to compilation even though it is not used)
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
M components/cam/bld/namelist_files/namelist_definition.xml
M components/cam/bld/build-namelist
- added clubb_timestep, clubb_rnevap_effic, clubb_stabcorrect
M components/cam/src/physics/cam/micro_mg2_0.F90
M components/cam/src/physics/cam/micro_mg_cam.F90
- introduced allow_sed_supersat which is set to false for clubb runs
M components/cam/src/physics/cam/clubb_intr.F90
- numerous changes to support changes in the CLUBB interface
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB except:
035 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Fri Apr 24 04:35:59 MDT 2015
-- CLUBB runs are expected to change answers. Runs have been validated extensively by Peter Bogenschutz
goldbach/nag: all BFB
goldbach/pgi or jaguar/pgi: all BFB
===============================================================
===============================================================
Tag name: cam5_3_82
Originator(s): cacraig, eaton
Date: 04/21/2015
One-line Summary: Revised how the new solar insolation calculation (shr_orb_avg_cosz) is called from CESM
Purpose of changes:
- After discussions with Rich Neale and Keith Lindsey, the way shr_orb_cosz is called has been revised
-- shr_orb_cosz now has an optional argument (dt_avg) which if set and is non-zero will be used as
the average time step in the new cosz calculation (there is no longer a module level variable in shr)
-- the namelist use_rad_dt_cosz sets the dt_avg parameter ONLY for the radiation calls to shr_orb_cosz (via
the CAM wrapper routine zenith)
-- Other CESM components (including chemistry) are currently defaulting to the original shr_orb_cosz calculation
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
D components/cam/src/physics/cam/zenith.F90
- This bare subroutine has been made into a module and moved into utils/orbit.F90
List all subroutines added and what they do:
A + components/cam/src/utils/orbit.F90
- This contains the zenith.F90 subroutine and has been made into a module
- Added the optional argument dt_avg which is passed along ot the shr_orb_cosz routine
List all existing files that have been modified, and describe the changes:
M .
M SVN_EXTERNAL_DIRECTORIES
- updated to cime0_3_31 which contains the revised shr_orb_cosz call
- shr_orb_cosz now has an optional parameter (dt_avg) which if it is set and is
non-zero will trigger calling the new shr_orb_avg_cosz calculation
M components/cam/bld/configure
- pio now has an additional directory layer "cime/externals/pio/pio"
M components/cam/src/physics/cam/radiation.F90
M components/cam/src/physics/rrtmg/radiation.F90
- dt_avg is now a radiation module level variable and is set via radiation_init if use_rad_dt_cosz is true,
otherwise dt_avg defaults to zero
M components/cam/src/physics/waccmx/ionosphere.F90
M components/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
M components/cam/src/chemistry/mozart/mo_waccm_hrates.F90
- added use statements for zenith which now resides in the module orbit
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB
goldbach/nag: all BFB
goldbach/pgi or jaguar/pgi: all BFB
NOTE -- The results from using the namelist use_rad_dt_cosz will be different than the results in cam5_3_80.
===============================================================
===============================================================
Tag name: cam5_3_81
Originator(s): mvertens, cacraig
Date: 04/17/2015
One-line Summary: minor changes to support pop2 being renamed to pop
Purpose of changes:
- pop2 is being renamed to pop. CAM had a few places where this needed to be changed
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/bld/configure
M components/cam/bld/config_files/definition.xml
M components/cam/bld/cam.buildnml
- pop2 is now pop
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB
goldbach/nag: all BFB
goldbach/pgi or jaguar/pgi: all BFB
===============================================================
===============================================================
Tag name: cam5_3_80
Originator(s): zhoulinjiong2010@gmail.com, minghua.zhang@stonybrook.edu, cacraig
Date: 4/14/2015
One-line Summary: Modifications to bring in new optional solar insolation calculation (shr_orb_cosz)
Purpose of changes:
- From Minghua Zhang's email:
A new routine (shr_orb_avg_cosz) has been introduced "that eliminates the spurious zonal oscillation of
daily insolation caused by discrete time sampling. The algorithm uses the mean of cosz in a time step.
The module shr_orb_mod.F90 calls this subroutine"
- Added a shr_orb_init routine which is used to set the dt to use in the above calculation. This dt is the
size of the radiation time step. If shr_orb_init is not called, the original shr_orb_cosz calculation is used.
- It is important to note that ALL CESM components will use this modified cosz calculation with the specified
dt value if it is turned on.
- This new shr functionality is controlled via the new CAM namelist logical: use_rad_dt_cosz
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
- use_rad_dt_cosz: By default it is false and the original shr_orb_cosz calculation is used. If this parameter
is set to true, then the shr_orb_avg_cosz calculation described above is used for all components in CESM which
call shr_orb_cosz.
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton, results evaluated by Cecile Hannay and Rich Neale
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M .
M SVN_EXTERNAL_DIRECTORIES
- updated to cime0_3_21 which contains the modified shr_orb_cosz
M components/cam/test/system/input_tests_master
- corrected er736 entry
M components/cam/bld/namelist_files/namelist_definition.xml
- introduced use_rad_dt_cosz logical to control the type of calculation used
when calling shr_orb_cosz
M components/cam/src/control/runtime_opts.F90
- added use_rad_dt_cosz to namelist and pass it into radiation defaultotps/setopts
M components/cam/src/physics/cam/radiation.F90
M components/cam/src/physics/rrtmg/radiation.F90
- if use_rad_dt_cosz is true, then the average radiation timestep is calculated and passed
into shr_orb_mod via the call to shr_orb_init
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB
goldbach/nag: all BFB
goldbach/pgi or jaguar/pgi: all BFB
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: when use_rad_dt_cosz is set to true
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate): larger than roundoff, but same climate
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository): https://svn-ccsm-models.cgd.ucar.edu/cime/branch_tags/sol_insolation_tags/sol_insolation_n04_cime0_3_13 (ran 1 month of cam5_3_80 and it was BFB with this version)
- platform/compilers: all
- configure commandline: all
- build-namelist command (or complete namelist): use_rad_dt_cosz=.true.
URL for AMWG diagnostics output used to validate new climate:
http://webext.cgd.ucar.edu/FCLIMO/f.e13.F2000C5.f09_f09_cam5.4_cam5_3_77.001_n03-2/atm/f.e13.F2000C5.f09_f09_cam5.4_cam5_3_77.001_n03-2-f.e13.F2000C5.f09_f09_cam5.4_cam5_3_77.001/
===============================================================
===============================================================
Tag name: cam5_3_79
Originator(s): richard.easter, po-lun.ma, balwinder.singh, eaton
Date: Fri Apr 10 11:38:10 MDT 2015
One-line Summary: Add modal aerosol convective scavenging option.
Purpose of changes:
. Unified treatment for convective transport and scavenging of modal
aerosols:
The wet removal and transport of modal aerosols by convective clouds in
CAM5 are treated in two different places (though there are linkages
between them). This new parameterization treats the two processes in a
more coherent manner in which aerosol activation within the convective
updraft is treated based on Abdul-Razzak and Ghan (2000)
parameterization. It also treats the secondary activation for aerosols
entrained from the environment. This new parameterization reduces the
aerosol transport to the upper troposphere, and provide improved aerosol
vertical distribution (in better agreement with observations).
The new parameterization is off by default, and is enabled by setting the
namelist variable convproc_do_aer=.true.
. The constituents module was extended to allow specifying that the ZM
convective transports should be turned off. This is to support the new
convective scavenging code which does the transport and scavenging
together.
. Fields that characterize the transports in the ZM deep and UW shallow
convection parameterizations have been added to the physics buffer for
use in either the ZM convtran routine or the new parameterization for
unified convection/scavenging.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist:
. Add convproc_do_aer to phys_ctl_nl group. Turns on new convective
scavenging option.
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: cam code review on 1 April 2015.
List all subroutines eliminated: none
List all subroutines added and what they do:
components/cam/src/chemistry/modal_aero/modal_aero_convproc.F90
. module for the new modal aerosol convective scavenging parameterization
List all existing files that have been modified, and describe the changes:
components/cam/bld/Makefile.in
. add missing dependency for perf_mod.o to fix occasional build failures
components/cam/bld/namelist_files/namelist_definition.xml
. Add convproc_do_aer to phys_ctl_nl group. Turns on new convective
scavenging option.
components/cam/src/chemistry/aerosol/wetdep.F90
. Add option to remove the resuspension term from bcscavt and bsscavt and
return it in rcscavt and rsscavt respectively.
components/cam/src/chemistry/modal_aero/aero_model.F90
. add diagnostics for convective scavenging
. aero_model_wetdep
. add option in loop over phases (interstitial and cloud borne) to do the
cloud borne first when convective scavenging is active
. add call to new convective scavenging run method
components/cam/src/chemistry/modal_aero/modal_aero_data.F90
. move definition of species_class values so they can be used to initialize
the array.
components/cam/src/chemistry/modal_aero/modal_aero_initialize_data.F90
. add logic to turn off the deep convective transports of the aerosols when
the new convective scavenging is active.
. add call to new convective scavenging init method
. cleanup old code in modal_aero_initialize_q that assumed lat/lon grid
components/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
. modify chem diagnostics (call to het_diags) when convective scavenging
active
components/cam/src/physics/cam/clubb_intr.F90
. access cmfmc2 from pbuf instead of dummy arg
components/cam/src/physics/cam/constituents.F90
. add option to turn off deep convective transport (ZM) of tracers after
wetdep.
components/cam/src/physics/cam/convect_deep.F90
. remove unused ref to begchunk, endchunk
components/cam/src/physics/cam/convect_shallow.F90
. add CMFMC_SH (cmfmc2) and SH_E_ED_RATIO to pbuf
. remove cmfmc2 from arg list of convect_shallow_tend
components/cam/src/physics/cam/macrop_driver.F90
. remove cmfmc2 from arg lists and access from pbuf (CMFMC_SH) instead
components/cam/src/physics/cam/ndrop.F90
. make activate_modal public, and add optional arg smax_prescribed
. add missing intents in arg declarations for both activate_modal and maxsat
components/cam/src/physics/cam/phys_control.F90
. add convproc_do_aer to namelist and accessor method
components/cam/src/physics/cam/physpkg.F90
. remove cmfmc2 from arg lists
components/cam/src/physics/cam/rk_stratiform.F90
. remove cmfmc2 from arg lists and access from pbuf (CMFMC_SH) instead
components/cam/src/physics/cam/uwshcu.F90
. add sh_e_ed_ratio to output args
components/cam/src/physics/cam/zm_conv.F90
. remove lengath from args, and add initializer to ideep to allow lengath
to be computed on the fly.
components/cam/src/physics/cam/zm_conv_intr.F90
. move the module data for the convective fluxes into the pbuf
. in zm_conv_tend_2 make use of new cnst_is_convtran2 logical array to
decide which species to transport.
components/cam/test/system/input_tests_master
. fix specification of tests er736, bl736
components/cam/test/system/tests_pretag_goldbach_nag
. remove br354, {sm,er,bl}312
components/cam/test/system/tests_pretag_goldbach_pgi
. remove br354, {sm,er,bl}312, {sm,er,bl}367
components/cam/test/system/tests_pretag_yellowstone
. remove br353
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all PASS
goldbach/nag: all PASS
goldbach/pgi: all PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: none
===============================================================
===============================================================
Tag name: cam5_3_78
Originator(s): cacraig
Date: 04/09/2015
One-line Summary: Update externals and introduce revised cam5.4 regression tests
Purpose of changes:
- Update to cesm1_4_beta02 cime external
- Update to latest cice branch for cime work (cice5 has issues with CAM and will be brought in later)
- Put in 2degree cam5.4 regression tests
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
A components/cam/test/system/config_files/f1.9c5.4dh
A components/cam/test/system/config_files/f1.9c5.4dm
- Introduced current configure options for CAM5.4 for regression testing purposes
- Use 2 degree setup since 1 degree took very long to run on yellowstone
List all existing files that have been modified, and describe the changes:
M .
M SVN_EXTERNAL_DIRECTORIES
- update to cesm1_4_beta02 cime tag
- use latest cice4 branch for cime work
- Bring in only the CLM directories which are used by CAM (more efficient checkout)
M components/cam/test/system/tests_pretag_goldbach_pgi
M components/cam/test/system/tests_pretag_goldbach_nag
- Add 2 degree CAM5.4 tests
M components/cam/test/system/input_tests_master
- Change test #353 to use 2 degree files
- Add #354 to be an MPI only test
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
- Add two degree initial condition file for 32 levels
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB except:
033 bl353 TBL.sh f1.9c5.4dh outfrq3s_cam5.4 9s ................................................FAIL! rc= 7 at Wed Apr 8 16:56:51 MDT 2015
- New test - no baseline for comparision
goldbach/nag: all BFB except:
051 bl354 TBL.sh f1.9c5.4dm outfrq3s_cam5.4 9s ................................................FAIL! rc= 7 at Wed Apr 8 12:16:56 MDT 2015
- New test - no baseline for comparision:
goldbach/pgi or jaguar/pgi: all BFB except:
058 bl354 TBL.sh f1.9c5.4dm outfrq3s_cam5.4 9s ................................................FAIL! rc= 7 at Thu Apr 9 11:27:16 MDT 2015
- New test - no baseline for comparision
===============================================================
===============================================================
Tag name: cam5_3_77
Originator(s): cacraig, eaton, hannay
Date: 04/07/2015
One-line Summary: Introduced namelist seasalt_emis_scale and changed value when modal_accum_coarse_exch is true
Purpose of changes:
- Allow emis_scale in seasalt to be set via the namelist
- Use the tuning parameter that Cecile determined was best when modal_accum_coarse_exch is set to true
- Introduced new cam regression test to test cam5.4
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
- Introduced seasalt_emis_scale. It defaults to 1.62 for mam7, 0.9 when modal_accum_coarse_exch is true and otherwise is 1.35.
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
A components/cam/test/system/config_files/f0.9c5.4dh
A components/cam/test/system/nl_files/outfrq3s_cam5.4
- setup for new regression tests
List all existing files that have been modified, and describe the changes:
M components/cam/test/system/TSM.sh
- removed debug flag which was inadvertently checked in
M components/cam/test/system/tests_pretag_yellowstone
M components/cam/test/system/input_tests_master
- added cam5.4 regression test
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
M components/cam/bld/namelist_files/namelist_definition.xml
M components/cam/bld/build-namelist
M components/cam/src/chemistry/modal_aero/aero_model.F90
M components/cam/src/chemistry/modal_aero/seasalt_model.F90
- added seasalt_emis_scale as a namelist tuning parameter
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB except:
065 bl474 TBL.sh f1.9c5wtmam3h outfrq3s+1850_waccm_tsmlt_cam5 9s ..............................FAIL! rc= 7 at Mon Apr 6 15:09:31 MDT 2015
- seasalt_emis_scale tuning parameter is changed when modal_accum_coarse_exch is set to true
goldbach/nag: all BFB except:
062 bl471 TBL.sh f10c5wmam3dm outfrq3s_modalstrat 9s ..........................................FAIL! rc= 7 at Mon Apr 6 16:48:12 MDT 2015
- seasalt_emis_scale tuning parameter is changed when modal_accum_coarse_exch is set to true
goldbach/pgi or jaguar/pgi: all BFB
Summarize any changes to answers, i.e.,
- what code configurations: answer changing when modal_accum_coarse_exch is set to true
- what platforms/compilers: all
- nature of change (roundoff; larger than roundoff but same climate; new
climate): climate
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline: -phys cam5 -microphys mg2 -chem trop_mam4 -nlev 32 -cppdefs -DUSE_XLIU_MOD
- build-namelist command (or complete namelist):
use_preexisting_ice = .true.
use_hetfrz_classnuc = .true.
cldfrc_rhminl_adj_land = 0.00D0
modal_accum_coarse_exch = .true.
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
http://webext.cgd.ucar.edu/FCLIMO/f.e13.F2000C5.f09_f09_beta16.cam5.4_mills01_cam5_3_70.011/atm/f.e13.F2000C5.f09_f09_beta16.cam5.4_mills01_cam5_3_70.011-obs/
NOTE -- Cecile plans to rerun these diagnostics with this tag and the ChangeLog will be updated to reflect these results. Also will include the details on her previous runs once she supplies it.
===============================================================
===============================================================
Tag name: cam5_3_76
Originator(s): mills, hannay, cacraig
Date: 04/03/2015
One-line Summary: limit the mode exchange between coarse accumulation in the stratosphere
Purpose of changes:
- limit the mode exchange between coarse accumulation in the stratosphere
- bug fix from Mike Mills
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/src/chemistry/modal_aero/modal_aero_rename.F90
- bug fix and limiting the mode exchange
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB except:
061 bl474 TBL.sh f1.9c5wtmam3h outfrq3s+1850_waccm_tsmlt_cam5 9s ..............................FAIL! rc= 7 at Fri Apr 3 16:36:50 MDT 2015
- Uses modal_accum_coarse_exch=.true. and answers were expected to change when this flag is set
goldbach/nag: all BFB except:
062 bl471 TBL.sh f10c5wmam3dm outfrq3s_modalstrat 9s ..........................................FAIL! rc= 7 at Fri Apr 3 13:55:28 MDT 2015
- Uses modal_accum_coarse_exch=.true. and answers were expected to change when this flag is set
goldbach/pgi or jaguar/pgi: all BFB
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: when modal_accum_coarse_exch=.true.
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate): unknown - part of the climate changing mods made in cam5_3_75, 76 and 77 and tested in aggregate by hannay
tag for the documentation of these runs will occur in upcoming tag (currently slated to be cam5_3_77)
===============================================================
===============================================================
Tag name: cam5_3_75
Originator(s): s.albani@cornell.edu, hannay, cacraig
Date: 4/2/2015
One-line Summary: Updated soil erodobility file
Purpose of changes:
- new dust erodibility file - improves the dust in the Middle East
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
- new default file for soil_erod_file
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M components/cam/bld/namelist_files/namelist_defaults_cam.xml
- new dust erodibility file from Samuel Albani (s.albani@cornell.edu)
M components/cam/doc/ReleaseNotes
- updated with COSP and dust notes
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
009 bl203 TBL.sh fsd1.9c4wtsmlth outfrq3s_sd+sd_1975-2010_ccmi_refc1_waccm_tsmlt 9s ...........FAIL! rc= 7 at Thu Apr 2 16:47:48 MDT 2015
024 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...................................................FAIL! rc= 7 at Thu Apr 2 17:21:18 MDT 2015
032 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Thu Apr 2 17:48:17 MDT 2015
036 bl373 TBL.sh f1.9c4mozdh outfrq3s+2000_cam4_trop_chem 9s ..................................FAIL! rc= 7 at Thu Apr 2 18:04:47 MDT 2015
038 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...............................................FAIL! rc= 7 at Thu Apr 2 18:07:47 MDT 2015
040 bl376 TBL.sh f1.9c4mozdh+testmech outfrq3s 9s .............................................FAIL! rc= 7 at Thu Apr 2 18:14:13 MDT 2015
053 bl421 TBL.sh f1.9c5mam4dh outfrq3s 9s .....................................................FAIL! rc= 7 at Thu Apr 2 18:46:39 MDT 2015
055 bl430 TBL.sh f1.9c5mzmam3dh outfrq3s_NEUwetdep 9s .........................................FAIL! rc= 7 at Thu Apr 2 18:56:03 MDT 2015
058 bl440 TBL.sh f1.9c4soadh outfrq3s+soa_chem_megan_emis 9s ..................................FAIL! rc= 7 at Thu Apr 2 19:13:52 MDT 2015
061 bl474 TBL.sh f1.9c5wtmam3h outfrq3s+1850_waccm_tsmlt_cam5 9s ..............................FAIL! rc= 7 at Thu Apr 2 19:26:46 MDT 2015
064 bl482 TBL.sh f1.9c5wscmam3dh outfrq3s_2005+waccm_sc_1955-2005_cam5 9s .....................FAIL! rc= 7 at Thu Apr 2 19:51:17 MDT 2015
085 bl740 TBL.sh h30c5h outfrq3s_lb0 9s .......................................................FAIL! rc= 7 at Thu Apr 2 20:39:00 MDT 2015
goldbach/nag:
010 bl112 TBL.sh e8c5dm ghgrmp 9s .............................................................FAIL! rc= 7 at Thu Apr 2 13:59:02 MDT 2015
014 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Thu Apr 2 14:07:01 MDT 2015
018 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Thu Apr 2 14:19:24 MDT 2015
026 bl312 TBL.sh f2.5c5dm ghgrmp 9s ...........................................................FAIL! rc= 7 at Thu Apr 2 14:55:58 MDT 2015
043 bl325 TBL.sh f10c5t5mdmsc outfrq3s_subcol 9s ..............................................FAIL! rc= 7 at Thu Apr 2 15:46:29 MDT 2015
050 bl368 TBL.sh f1.9c5mg2dm outfrq3s_macmic2_classnuc 9s .....................................FAIL! rc= 7 at Thu Apr 2 16:33:27 MDT 2015
056 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Thu Apr 2 17:15:28 MDT 2015
062 bl471 TBL.sh f10c5wmam3dm outfrq3s_modalstrat 9s ..........................................FAIL! rc= 7 at Thu Apr 2 17:48:31 MDT 2015
065 bl805 TBL.sh f4c5carmtt1dm outfrq3s_carma 9s ..............................................FAIL! rc= 7 at Thu Apr 2 18:29:32 MDT 2015
goldbach/pgi or jaguar/pgi:
010 bl112 TBL.sh e8c5dm ghgrmp 9s .............................................................FAIL! rc= 7 at Thu Apr 2 14:04:09 MDT 2015
016 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Thu Apr 2 14:13:51 MDT 2015
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Thu Apr 2 14:33:24 MDT 2015
031 bl312 TBL.sh f2.5c5dm ghgrmp 9s ...........................................................FAIL! rc= 7 at Thu Apr 2 15:10:31 MDT 2015
044 bl317 TBL.sh f10c5cdm outfrq3s 9s .........................................................FAIL! rc= 7 at Thu Apr 2 15:54:44 MDT 2015
054 bl321 TBL.sh f10c5cdm atrain 9s ...........................................................FAIL! rc= 7 at Thu Apr 2 16:11:08 MDT 2015
057 bl367 TBL.sh f1.9c5mg2dm outfrq3s 9s ......................................................FAIL! rc= 7 at Thu Apr 2 16:51:33 MDT 2015
060 bl420 TBL.sh f4c5mam7dm outfrq3s 9s .......................................................FAIL! rc= 7 at Thu Apr 2 17:12:15 MDT 2015
063 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Thu Apr 2 17:29:58 MDT 2015
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: all configurations which use prognostic dust
- what platforms/compilers: all
- nature of change (roundoff; larger than roundoff but same climate; new
climate): new climate
tag for the documentation of these runs (currently slated to be cam5_3_77)
===============================================================
===============================================================
Tag name: cam5_3_74
Originator(s): mvertens, cacraig
Date: March 24, 2015
One-line Summary: ** REORGANIZATION OF models/atm/cam TO components/cam ** and introducing new CESM testing framework
Purpose of changes:
- All files which were in models/atm/cam/ have been moved to components/cam/
- All externals other than main model components are now in cime/
- Introduced new CESM testing framework
- Removed janus, titan and hopper tests from cime testing and renamed FC51850 to F1850C5 to match the compset name per Jay's request
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
*** ALL FILES MOVED FROM /models/atm/cam to /components/cam ***
M .
M SVN_EXTERNAL_DIRECTORIES
- new external list matches CESM1_4_beta01
M + components/cam/test/system/TPF.sh
M + components/cam/test/system/test_driver.sh
M + components/cam/test/system/TBL_ccsm.sh
M + components/cam/test/system/TFM.sh
M + components/cam/test/system/TSM.sh
M + components/cam/test/system/gen-test-coverage
M + components/cam/test/system/TR8.sh
M + components/cam/test/system/TCB_ccsm.sh
M + components/cam/test/system/TBL.sh
M + components/cam/test/system/gen-test-table
M + components/cam/test/system/gen_test_table.sh
M + components/cam/bld/configure
M + components/cam/bld/Makefile.in
M + components/cam/bld/run-ibm.csh
M + components/cam/bld/run-yellowstone.csh
M + components/cam/bld/get_nxny.pl
M + components/cam/bld/run-lynx.csh
M + components/cam/bld/perl5lib/Build/ChemPreprocess.pm
M + components/cam/bld/run-pc.csh
M + components/cam/bld/run-scam.csh
M + components/cam/bld/run-darwin.csh
M + components/cam/bld/build-namelist
M + components/cam/bld/run-cray.csh
- changes required due to moving models/atm/cam/ to components/cam/
M + components/cam/cimetest/testlist_cam.xml
- removed janus, hopper and titan tests per Jay's request
- Corrected typo FC51850 to F1850C5 (the correct compset name)
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All BFB
goldbach/nag: All BFB
goldbach/pgi or jaguar/pgi: All BFB
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_73
Originator(s): jenkay, Po-lun, G. Cesana and H. Chepfer (LMD, Helene.Chepfer@lmd.polytechnique.fr), cacraig
Date: 03/23/15
One-line Summary: Update cosp from 1.3 to 1.4
Purpose of changes:
CESM-relevant changes from COSP1.3 to COSP1.4 taken from the CFMIP website (http://cfmip.metoffice.com/) are:
1) In radar simulator: new attenuation integration scheme and other optimsations, possibility of using a look up table, partial support of two-moment microphysics, selection of microphysics via cosp_defs.h.
2) Optimised version of cosp_change_vertical_grid in cosp_stats
3) New CALIPSO cloud phase diagnostics: large list of new cloud diagnostics developed by Gregory Cesana (Cesana and Chepfer, doi:10.1002/jgrd.50376)
4) New timing variables that allow to get a quick estimate of the performance of each simulator.
In addition - the COSP1.4 in CESM implementation has:
1) radiatively active snow in all simulators (described in Kay et al. 2012 doi: http://dx.doi.org/10.1175/JCLI-D-11-00469.1) with lidar snow treatment error fix (described in English et al. 2014 doi: http://dx.doi.org/10.1175/JCLI-D-13-00608.1)
2) bug fix identified by Po-Lun for convective ice input into COSP
Bugs fixed (include bugzilla ID):
- bug fix identified by Po-Lun for convective ice input into COSP
There are differences in the MODIS simulator outputs related to Po-Lun's bug fix.
From Robert Pincus's e-mail on 11/16/14: " The sense of the bug is to essentially
double the mixing ratio of convective water cloud but to call half of it ice. The
MODIS simulator diagnoses phase based on the relative amounts of extinction near
cloud top by cloud liquid and cloud ice. When the ratio of extinction is greater
than 70% for either ice or water we say we know the phase; if it's less we say it's
undetermined but assume water. The bug puts a bunch of convective "ice" low in the
troposphere and pairs it with equal amounts of convective liquid and possible
large-scale liquid, so it'll always be diagnosed as liquid (or undetermined and then
assumed liquid). So it makes sense that the LWP goes down, although it also implies
that there's a *lot* of convective LWP (145-130, or about 15 g/m2). The reduction in
particle size makes sense, too: with the bug, the optical properties are determined
by a mix of liquid and "ice", where the ice single scattering albedo (the quantity
relevant for the re retrieval) is consistent with large ice particles."
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M .
M models/atm/cam
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
- updated the cosp externals
M models/atm/cam/test/system/TBR.sh
M models/atm/cam/test/system/TSM.sh
- added printing of commands to better document what is being run
M models/atm/cam/test/system/tests_pretag_yellowstone
- added cosp testing (was in posttag which is no longer run)
M models/atm/cam/src/physics/cam/cospsimulator_intr.F90
- modifications to support cosp1.4
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB except:
050 bl397 TBL.sh f1.9c4cm outfrq3s_cosp+1850-2005_cam4 9s .....................................FAIL! rc= 7 at Fri Mar 20 23:20:02 MDT 2015
goldbach/nag: all BFB except:
037 bl318 TBL.sh f10c4cdm sat_hist 9s .........................................................FAIL! rc= 7 at Fri Mar 20 18:10:19 MDT 2015
goldbach/pgi or jaguar/pgi: all BFB except:
044 bl317 TBL.sh f10c5cdm outfrq3s 9s .........................................................FAIL! rc= 7 at Fri Mar 20 18:21:38 MDT 2015
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: cosp
- what platforms/compilers: all
- nature of change (roundoff; larger than roundoff but same climate; new
climate): just in cosp diagnostics
Location of cosp1.3/cosp1.4 diagnostic plots:
http://www.cgd.ucar.edu/staff/jenkay/cosp/diag/cosp_n03_cam5_3_70_intel_test-cam5_3_70_intel_control/sets.htm
===============================================================
===============================================================
Tag name: cam5_3_72
Originator(s): fvitt, dkin
Date: 13 Mar 2015
One-line Summary: CCMI chemistry updates and PORT bug fix
Purpose of changes:
- Add namelist options for BULK aerosol solubility factors.
- Replace "cldfrc_rhminp" namelist option with "rk_strat_polstrat_rhmin".
This gives the user an opportunity to use a different rhmin threshold for
dehydration in the polar stratosphere (poleward of 50 degrees on pressure
levels above 300 mbar). Previously this was applied to cloud fraction
which lead to unrealistic cloud fractions in the polar stratosphere.
- Change the vertical boundary separating the stratospheric aerosols and
tropospheric BULK aerosols in the polar region (poleward of 50 degrees)
from the tropopause to 300 mbars applied to the surface area densities used
in chemistry.
- Update stratospheric aerosol chemistry rate constants (dkin)
- Chemistry mechanism updates to packages:
trop_strat_soa
waccm_mozart
waccm_tsmlt_mam3
waccm_tsmlt
waccm_tsmlt_sulfur
- Fix a bug in PORT -- initialize physics buffer fields used
in "fixed dynamical heating" calculation
- Correct units of NEU wet deposition diagnostics
Bugs fixed (include bugzilla ID):
- initialize physics buffer fields used in "fixed dynamical heating" calculation of PORT
- correction in units of DTWR_* diagnostics NEU wet deposition module
Describe any changes made to build system:
Describe any changes made to the namelist:
Namelist options were added for BULK aerosol solubility factor lists 'aer_sol_facti'
and 'aer_sol_factb' for respectively in-cloud and below-cloud solubility factors.
For convenience the following options were added for different aerosol types which
are mapped appropriately to the 'aer_sol_facti' and 'aer_sol_factb' lists:
. SO4_sol_facti : in-cloud solubility factor for SO4
. SO4_sol_factb : below-cloud solubility factor for SO4
. NH4_sol_facti : in-cloud solubility factor for NH4
. NH4_sol_factb : below-cloud solubility factor for NH4
. NH4NO3_sol_facti : in-cloud solubility factor for NH4NO3
. NH4NO3_sol_factb : below-cloud solubility factor for NH4NO3
. CB2_sol_facti : in-cloud solubility factor for black carbon
. CB2_sol_factb : below-cloud solubility factor for black carbon
. OC2_sol_facti : in-cloud solubility factor for organic carbon
. OC2_sol_factb : below-cloud solubility factor for organic carbon
. dust_sol_facti : in-cloud solubility factor for dust
. dust_sol_factb : below-cloud solubility factor for dust
. sslt_sol_facti : in-cloud solubility factor for sea salt
. sslt_sol_factb : below-cloud solubility factor for sea salt
Replaced "cldfrc_rhminp" option with "rk_strat_polstrat_rhmin".
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton and others in the CAM code review team
List all subroutines eliminated:
D models/atm/cam/test/system/tests_titan_cb
D models/atm/cam/test/system/tests_pretag_titan_pgi_cb
- these obsolete test lists were removed
D models/atm/cam/src/physics/cam/stratiform.F90
- this has been renamed rk_stratiform
List all subroutines added and what they do:
A models/atm/cam/src/physics/cam/rk_stratiform.F90
- was previously stratiform.F90
- moved cldwat namelist to stratiform
- added rk_strat_polstrat_rhmin namelist option
A models/atm/cam/test/system/config_files/f1.9c4portdh
A models/atm/cam/test/system/config_files/f1.9c4portdm
A models/atm/cam/test/system/nl_files/port_cam4
- added for testing PORT
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/tests_pretag_yellowstone
- changed PORT test
M models/atm/cam/test/system/input_tests_master
- "cb" tests removed -- these are no longer needed
- PORT tests added
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- cldwat_* namelist options renamed to rk_strat_*
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- cldwat_* namelist options renamed to rk_strat_*
- added rk_strat_polstrat_rhmin option
- removed cldfrc_rhminp and cldfrc_rhminp_botmb options
- added bulk aerosol solubility factor lists 'aer_sol_facti' and 'aer_sol_factb'
for respectively in-cloud and below-cloud solubility factors.
For use convience the following options were added for different aerosol types:
. SO4_sol_facti : in-cloud solubility factor for SO4
. SO4_sol_factb : below-cloud solubility factor for SO4
. NH4_sol_facti : in-cloud solubility factor for NH4
. NH4_sol_factb : below-cloud solubility factor for NH4
. NH4NO3_sol_facti : in-cloud solubility factor for NH4NO3
. NH4NO3_sol_factb : below-cloud solubility factor for NH4NO3
. CB2_sol_facti : in-cloud solubility factor for black carbon
. CB2_sol_factb : below-cloud solubility factor for black carbon
. OC2_sol_facti : in-cloud solubility factor for organic carbon
. OC2_sol_factb : below-cloud solubility factor for organic carbon
. dust_sol_facti : in-cloud solubility factor for dust
. dust_sol_factb : below-cloud solubility factor for dust
. sslt_sol_facti : in-cloud solubility factor for sea salt
. sslt_sol_factb : below-cloud solubility factor for sea salt
M models/atm/cam/bld/namelist_files/use_cases/2004-2100_ccmi_refc2_rcp45_trop_strat_soa.xml
M models/atm/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_waccm_tsmlt.xml
M models/atm/cam/bld/namelist_files/use_cases/sd_1975-2010_ccmi_refc1_trop_strat_soa.xml
M models/atm/cam/bld/namelist_files/use_cases/1950-2010_ccmi_refc1_waccm_ma.xml
M models/atm/cam/bld/namelist_files/use_cases/sd_1975-2010_ccmi_refc1_waccm_tsmlt.xml
M models/atm/cam/bld/namelist_files/use_cases/sd_2007-2011_htap_trop_strat_soa.xml
M models/atm/cam/bld/namelist_files/use_cases/2000_ccmi_refc1_waccm_tsmlt.xml
M models/atm/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_waccm_tsmlt_cam5.xml
M models/atm/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_rcp85_waccm_ma.xml
M models/atm/cam/bld/namelist_files/use_cases/1979-2010_sd_trop_strat_soa.xml
M models/atm/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_waccm_ma.xml
M models/atm/cam/bld/namelist_files/use_cases/1950-2010_ccmi_refc1_waccm_tsmlt.xml
M models/atm/cam/bld/namelist_files/use_cases/2007-2011_htap_trop_strat_soa.xml
M models/atm/cam/bld/namelist_files/use_cases/1950-2010_ccmi_refc1_trop_strat_soa.xml
M models/atm/cam/bld/namelist_files/use_cases/sd_1975-2010_ccmi_refc1_waccm_ma.xml
M models/atm/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_trop_strat_soa.xml
- updates from CCMI simulations
- cldwat_* renamed to rk_strat_*
M models/atm/cam/bld/namelist_files/use_cases/aquaplanet_cam3.xml
M models/atm/cam/bld/namelist_files/use_cases/cam3.xml
- cldwat_* renamed to rk_strat_*
M models/atm/cam/bld/build-namelist
- set defaults for aer_sol_facti and aer_sol_factb lists which are returned
from ChemNameList::set_dep_list routine
- add default 'prescribed_strataero_file' rather than 'sulf_file' for trop_strat_soa chem
- cldwat_* namelist options renamed to rk_strat_*
- correction in prescribed_aero_model setting when aerosols are not prognosed
M models/atm/cam/bld/perl5lib/Build/ChemNamelist.pm
M models/atm/cam/src/control/runtime_opts.F90
M models/atm/cam/src/physics/rrtmg/radiation.F90
M models/atm/cam/src/physics/cam/radiation.F90
- pass pbuf2d to rad_data_init
M models/atm/cam/src/physics/cam/radiation_data.F90
- PORT fix
-- initialize pbuf fields used in "fixed dynamical heating" calculations
M models/atm/cam/src/physics/cam/physpkg.F90
- references to stratiform changed to rk_stratiform
- radiation_init interface change
M models/atm/cam/src/physics/cam/cloud_fraction.F90
- reverted back to cam5_3_40 (pre-CCMI version)
M models/atm/cam/src/physics/cam/cldwat.F90
- namelist moved upto rk_stratiform
- added option (rk_strat_polstrat_rhmin) to use different rhmin
poleward of 50 degrees on pressure levels above 300 mbar for
dehydration in the polar stratosphere in special cases.
Previously this was applied to cloud fraction which lead to
unrealistic cloud fractions in the polar regions.
M models/atm/cam/src/chemistry/utils/prescribed_strataero.F90
- zero out stratospheric aerosols on pressure levels below the
300 mbar level poleward of 50 degrees, elsewhere pressure levels
above the tropopause
M models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
- zero out tropospheric sulfate on pressure levels above 300 mbar level
and poleward of 50 degrees, elesewhere on levels above the tropopause
- added SULF_TROP diagnostic
- changes for ratecon_sfstrat subroutine interface change
- added diagnostics for stratospheric aerosol reaction rates
M models/atm/cam/src/chemistry/bulk_aero/aero_model.F90
- aer_sol_facti and aer_sol_factb namelist options added
- zero out tropospheric surface area densities on pressure levels above 300 mbar level
and poleward of 50 degrees, elesewhere on levels above the tropopause
M models/atm/cam/src/chemistry/modal_aero/aero_model.F90
- interface change in aero_model_surfarea subroutine to be consistent
with bulk aerosol model
M models/atm/cam/src/chemistry/mozart/mo_strato_rates.F90
- science changes provided by D Kinnison
- output diagnostics through ratecon_sfstrat subroutine interface
M models/atm/cam/src/chemistry/mozart/mo_neu_wetdep.F90
- correction in units of DTWR_* diagnostics
M models/atm/cam/src/chemistry/mozart/mo_usrrxt.F90
- interface change to pass column latitudes to aero_model_surfarea
M models/atm/cam/src/dynamics/sld/inidat.F90
M models/atm/cam/src/dynamics/eul/inidat.F90
M models/atm/cam/src/dynamics/se/inidat.F90
M models/atm/cam/src/dynamics/fv/inidat.F90
- references to stratiform changed to rk_stratiform
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_setrxt.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_rxt_rates_conv.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/m_spc_id.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_strat_soa/chem_mech.in
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/chem_mech.doc
M models/atm/cam/src/chemistry/pp_waccm_mozart/chem_mech.in
M models/atm/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt_mam3/chem_mods.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt_mam3/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt_mam3/chem_mech.doc
M models/atm/cam/src/chemistry/pp_waccm_tsmlt_mam3/chem_mech.in
M models/atm/cam/src/chemistry/pp_waccm_tsmlt/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt/mo_setrxt.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt/mo_rxt_rates_conv.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt/chem_mods.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt/m_spc_id.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt/chem_mech.doc
M models/atm/cam/src/chemistry/pp_waccm_tsmlt/chem_mech.in
M models/atm/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_setrxt.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_rxt_rates_conv.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt_sulfur/chem_mods.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt_sulfur/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt_sulfur/m_spc_id.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt_sulfur/chem_mech.doc
M models/atm/cam/src/chemistry/pp_waccm_tsmlt_sulfur/chem_mech.in
- chemistry mechanism updates
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
036 bl373 TBL.sh f1.9c4mozdh outfrq3s+2000_cam4_trop_chem 9s ..................................FAIL! rc= 7 at Fri Mar 13 13:43:10 MDT 2015
040 bl376 TBL.sh f1.9c4mozdh+testmech outfrq3s 9s .............................................FAIL! rc= 7 at Fri Mar 13 13:52:03 MDT 2015
- expected failures due to the vertical boundary between stratospheric aerosols
and tropospheric aerosols in the polar region moved from the tropopause to 300 mbars
009 bl203 TBL.sh fsd1.9c4wtsmlth outfrq3s_sd+sd_1975-2010_ccmi_refc1_waccm_tsmlt 9s ...........FAIL! rc= 7 at Fri Mar 13 12:32:35 MDT 2015
029 bl336 TBL.sh f4c4wmdh outfrq3s+waccm_1850_cam4 9s .........................................FAIL! rc= 7 at Fri Mar 13 13:14:25 MDT 2015
046 bl390 TBL.sh f1.9c4wmxh outfrq3s+waccmx_2000_cam4 9s ......................................FAIL! rc= 7 at Fri Mar 13 14:07:49 MDT 2015
054 bl440 TBL.sh f1.9c4soadh outfrq3s+soa_chem_megan_emis 9s ..................................FAIL! rc= 7 at Fri Mar 13 14:44:00 MDT 2015
057 bl474 TBL.sh f1.9c5wtmam3h outfrq3s+1850_waccm_tsmlt_cam5 9s ..............................FAIL! rc= 7 at Fri Mar 13 14:56:00 MDT 2015
070 bl379 TBL.sh f1.9c4wmh outfrq24h+waccm_2005-2100_cam4_rcp45 2d ............................FAIL! rc= 7 at Fri Mar 13 15:45:59 MDT 2015
- expected failures due to the CCMI updates:
. changes in the chemistry mechanisms
. updates in the stratospheric aerosol chemistry rate constants
. the vertical boundary between stratospheric aerosols and tropospheric aerosols
in the polar region moved from the tropopause to 300 mbars
. the dehydration threshold rhmin in the polar stratosphere in no longer applied to
cloud fraction and is applied poleward of 50 degrees (previously 60 degrees)
092 bl012 TBL.sh f1.9c4portdh port_cam4 5d ....................................................FAIL! rc= 7 at Fri Mar 13 16:28:55 MDT 2015
- expected failure -- new PORT test
goldbach/nag:
059 bl424 TBL.sh f10c4wmxdm outfrq3s 9s .......................................................FAIL! rc= 7 at Fri Mar 13 13:09:44 MDT 2015
062 bl471 TBL.sh f10c5wmam3dm outfrq3s_modalstrat 9s ..........................................FAIL! rc= 7 at Fri Mar 13 13:25:18 MDT 2015
- expected failures due to waccm chemistry mechanism update
goldbach/pgi: All Pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_71
Originator(s): fvitt, mmills
Date: 12 Mar 2015
One-line Summary: Provide CAM5 with the capability to include source
of water vapor in the stratosphere from methane oxidation
Purpose of changes:
- Add option for external forcing of H2O from CH4 oxidation when running
low-top CAM5 without chemistry. CH4 oxidation is an important source of
H2O in the stratosphere. By default this additional source of water
vapor is off. This is turned on by adding H2O and corresponding external
forcing file to the "ext_frc_specifier" namelist option.
(NOTE: The chemistry now prognoses H2O and applies a tendency to water vapor,
which gives an additional and insignificant source of water vapor from
the oxidation of H2O2 (H2O2 + OH -> H2O + HO2) that is always applied
to CAM5.)
- Add default physics properties files for stratospheric modal aerosols
- Add external forcing of SO2 by OCS oxidation when running CAM5 without
chemistry when stratospheric modal aerosols are active (off by default)
- Add OCS to chem packages:
trop_mozart_soa
trop_mozart_mam3
trop_strat_soa
trop_strat_mam3
trop_strat_mam7
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- default phys props files added for stratospheric modal aerosols
- default SO2 forcing file corresponding to OCS oxidation source
- default photolysis rsf_file the same for all chem packages
M models/atm/cam/bld/build-namelist
- when modal_strat_sulfate is true include OCS oxidation forcing file for SO2
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/m_spc_id.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_mam3/chem_mech.in
M models/atm/cam/src/chemistry/pp_trop_mam4/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_trop_mam4/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_trop_mam4/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_mam4/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_trop_mam4/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mam4/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mam4/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_trop_mam4/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_trop_mam4/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_mam4/m_spc_id.F90
M models/atm/cam/src/chemistry/pp_trop_mam4/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_mam4/chem_mech.in
M models/atm/cam/src/chemistry/pp_trop_mam7/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/m_spc_id.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_mam7/chem_mech.in
- change H2O from fixed to prognostic in the stratosphere which provides the
capability to include H2O from oxidation of methane (via external forcing)
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_setrxt.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_rxt_rates_conv.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/m_spc_id.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/chem_mech.in
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_setrxt.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_rxt_rates_conv.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/m_spc_id.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/chem_mech.in
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_setrxt.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_rxt_rates_conv.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/m_spc_id.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/chem_mech.in
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_setrxt.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_rxt_rates_conv.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/m_spc_id.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_strat_soa/chem_mech.in
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_setrxt.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_rxt_rates_conv.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/m_spc_id.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/chem_mech.in
- added OCS and associated reactions
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
024 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...................................................FAIL! rc= 7 at Wed Mar 11 14:17:49 MDT 2015
032 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Wed Mar 11 14:43:11 MDT 2015
038 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...............................................FAIL! rc= 7 at Wed Mar 11 15:01:16 MDT 2015
049 bl421 TBL.sh f1.9c5mam4dh outfrq3s 9s .....................................................FAIL! rc= 7 at Wed Mar 11 15:33:13 MDT 2015
051 bl430 TBL.sh f1.9c5mzmam3dh outfrq3s_NEUwetdep 9s .........................................FAIL! rc= 7 at Wed Mar 11 15:42:13 MDT 2015
076 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ........................................FAIL! rc= 7 at Wed Mar 11 17:14:56 MDT 2015
081 bl740 TBL.sh h30c5h outfrq3s_lb0 9s .......................................................FAIL! rc= 7 at Wed Mar 11 17:18:27 MDT 2015
- expected failures in MAM without chemistry configurations due to making H2O prognostic
in the chemistry which gives an additional small source of H2O from H2O2+OH reaction.
There is also a change in photolysis rsf_file input file.
054 bl440 TBL.sh f1.9c4soadh outfrq3s+soa_chem_megan_emis 9s ..................................FAIL! rc= 7 at Wed Mar 11 15:59:17 MDT 2015
- expected failure due to chemistry mechanism change -- additions of OCS and associated reactions
057 bl474 TBL.sh f1.9c5wtmam3h outfrq3s+1850_waccm_tsmlt_cam5 9s ..............................FAIL! rc= 7 at Wed Mar 11 16:11:16 MDT 2015
- expected failure due to change in physics property input file for stratospheric modal aerosols
goldbach/nag:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s ...........................................................FAIL! rc= 7 at Wed Mar 11 16:31:09 MDT 2015
010 bl112 TBL.sh e8c5dm ghgrmp 9s .............................................................FAIL! rc= 7 at Wed Mar 11 16:31:13 MDT 2015
014 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Wed Mar 11 16:31:18 MDT 2015
018 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Wed Mar 11 16:31:19 MDT 2015
026 bl312 TBL.sh f2.5c5dm ghgrmp 9s ...........................................................FAIL! rc= 7 at Wed Mar 11 16:31:25 MDT 2015
034 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .........................................FAIL! rc= 7 at Wed Mar 11 16:31:26 MDT 2015
043 bl325 TBL.sh f10c5t5mdmsc outfrq3s_subcol 9s ..............................................FAIL! rc= 7 at Wed Mar 11 16:31:27 MDT 2015
050 bl368 TBL.sh f1.9c5mg2dm outfrq3s_macmic2_classnuc 9s .....................................FAIL! rc= 7 at Wed Mar 11 16:31:34 MDT 2015
056 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Wed Mar 11 16:31:36 MDT 2015
- expected failures in MAM without chemistry configurations due to making H2O prognostic
in the chemistry which gives an additional small source of H2O from H2O2+OH reaction.
There is also a change in photolysis rsf_file input file.
062 bl471 TBL.sh f10c5wmam3dm outfrq3s_modalstrat 9s ..........................................FAIL! rc= 7 at Wed Mar 11 16:46:22 MDT 2015
- expected failure due to change in physics property input file for stratospheric modal aerosols
065 bl805 TBL.sh f4c5carmtt1dm outfrq3s_carma 9s ..............................................FAIL! rc= 7 at Wed Mar 11 16:46:39 MDT 2015
- expected failure due to change in photolysis rsf_file input file
goldbach/pgi:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s ...........................................................FAIL! rc= 7 at Wed Mar 11 11:30:54 MDT 2015
010 bl112 TBL.sh e8c5dm ghgrmp 9s .............................................................FAIL! rc= 7 at Wed Mar 11 11:48:58 MDT 2015
012 bl113 TBL.sh e8c5paqdm aqpgro+aquaplanet_cam5 3s ..........................................FAIL! rc= 7 at Wed Mar 11 12:02:04 MDT 2015
016 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Wed Mar 11 12:04:06 MDT 2015
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Wed Mar 11 12:33:27 MDT 2015
031 bl312 TBL.sh f2.5c5dm ghgrmp 9s ...........................................................FAIL! rc= 7 at Wed Mar 11 13:21:02 MDT 2015
041 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .........................................FAIL! rc= 7 at Wed Mar 11 14:18:17 MDT 2015
044 bl317 TBL.sh f10c5cdm outfrq3s 9s .........................................................FAIL! rc= 7 at Wed Mar 11 14:39:05 MDT 2015
054 bl321 TBL.sh f10c5cdm atrain 9s ...........................................................FAIL! rc= 7 at Wed Mar 11 15:01:14 MDT 2015
057 bl367 TBL.sh f1.9c5mg2dm outfrq3s 9s ......................................................FAIL! rc= 7 at Wed Mar 11 15:41:00 MDT 2015
060 bl420 TBL.sh f4c5mam7dm outfrq3s 9s .......................................................FAIL! rc= 7 at Wed Mar 11 16:04:02 MDT 2015
063 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Wed Mar 11 16:27:11 MDT 2015
- expected failures in MAM without chemistry configurations due to making H2O prognostic
in the chemistry which gives an additional small source of H2O from H2O2+OH reaction.
There is also a change in photolysis rsf_file input file.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: cam5
- what platforms/compilers: all
- nature of change: larger than roundoff but same climate
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag: https://svn-ccsm-models.cgd.ucar.edu/cam1/branch_tags/mills_cam5_3_67_tags/mills01_cam5_3_70
- platform/compilers: yellowstone/intel
- configure commandline:
create_newcase -mach yellowstone -compset FC5 -res f19_f19 -case /glade/p/acd/fvitt/cesm/cases/f.e13.c5370.FC5.f19_f19.test.01
- build-namelist command: default namelist
- MSS location of output:
[HSI]/home/fvitt/csm/f.e13.c5370.FC5.f19_f19.test.01
MSS location of control simulations used to validate new climate:
[HSI]/home/fvitt/csm/f.e13.c5370.FC5.f19_f19.cntl.01
URL for AMWG diagnostics output used to validate new climate:
http://webext.cgd.ucar.edu/FCLIMO/f.e13.c5370.FC5.f19_f19.test.01/atm/f.e13.c5370.FC5.f19_f19.test.01-f.e13.c5370.FC5.f19_f19.cntl.01/
===============================================================
===============================================================
Tag name: cam5_3_70
Originator(s): santos (cacraig ran testing)
Date: March 5, 2015
One-line Summary: performance improvements and unit testing for micro_mg_utils; splitting sedimentation loop; minor cleanup
Purpose of changes:
Aside from some minor diagnostic/cleanup changes, the main differences are:
1) Speeding up micro_mg_utils, especially by reducing gamma and pow calls.
2) The sedimentation loop is split up, so that constituents other than rain go through fewer iterations.
- combined #1 and #2 give about a 14% speedup in MG2 (maybe 2% in CAM, depending on configuration)
3) Basic unit tests added for micro_mg_utils.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton, cacraig
List all subroutines eliminated:
List all subroutines added and what they do:
A models/atm/cam/test/unit/micro_mg_utils
A models/atm/cam/test/unit/micro_mg_utils/test_mg_size_utils.pf
A models/atm/cam/test/unit/micro_mg_utils/test_mg_processes.pf
A models/atm/cam/test/unit/micro_mg_utils/CMakeLists.txt
- added unit tests for micro_mg_utils
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/unit/micro_mg_data/test_MGPacker.pf
- uses packer%pack_interface now
M models/atm/cam/test/unit/CMakeLists.txt
M models/atm/cam/src/physics/cam/CMakeLists.txt
- added unit testing for micro_mg_utils
M models/atm/cam/src/physics/cam/micro_mg2_0.F90
- various cleanup
M models/atm/cam/src/physics/cam/micro_mg_cam.F90
- ADRAIN and ADSNOW are now output by default to allow them to be added to AMWG diagnostics
- added QRSEDTEN and QSSEDTEN (rain and snow ratio tendencies from sedimentation) for all MG except MG1.0
M models/atm/cam/src/physics/cam/micro_mg_utils.F90
- speedup changes
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
012 bl322 TBL.sh f1.9c5nonedh outfrq3s_lb0 9s .................................................FAIL! rc= 7 at Wed Mar 4 20:06:45 MST 2015
024 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...................................................FAIL! rc= 7 at Wed Mar 4 20:29:35 MST 2015
032 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Wed Mar 4 20:54:37 MST 2015
038 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...............................................FAIL! rc= 7 at Wed Mar 4 21:12:50 MST 2015
049 bl421 TBL.sh f1.9c5mam4dh outfrq3s 9s .....................................................FAIL! rc= 7 at Wed Mar 4 21:44:24 MST 2015
051 bl430 TBL.sh f1.9c5mzmam3dh outfrq3s_NEUwetdep 9s .........................................FAIL! rc= 7 at Wed Mar 4 21:53:23 MST 2015
057 bl474 TBL.sh f1.9c5wtmam3h outfrq3s+1850_waccm_tsmlt_cam5 9s ..............................FAIL! rc= 7 at Wed Mar 4 22:21:32 MST 2015
060 bl482 TBL.sh f1.9c5wscmam3dh outfrq3s_2005+waccm_sc_1955-2005_cam5 9s .....................FAIL! rc= 7 at Wed Mar 4 22:45:17 MST 2015
076 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ........................................FAIL! rc= 7 at Wed Mar 4 23:25:12 MST 2015
081 bl740 TBL.sh h30c5h outfrq3s_lb0 9s .......................................................FAIL! rc= 7 at Wed Mar 4 23:28:42 MST 2015
goldbach/nag:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s ...........................................................FAIL! rc= 7 at Wed Mar 4 17:12:00 MST 2015
010 bl112 TBL.sh e8c5dm ghgrmp 9s .............................................................FAIL! rc= 7 at Wed Mar 4 17:18:52 MST 2015
014 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Wed Mar 4 17:24:38 MST 2015
018 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Wed Mar 4 17:33:46 MST 2015
026 bl312 TBL.sh f2.5c5dm ghgrmp 9s ...........................................................FAIL! rc= 7 at Wed Mar 4 17:58:13 MST 2015
034 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .........................................FAIL! rc= 7 at Wed Mar 4 18:11:27 MST 2015
043 bl325 TBL.sh f10c5t5mdmsc outfrq3s_subcol 9s ..............................................FAIL! rc= 7 at Wed Mar 4 18:36:09 MST 2015
050 bl368 TBL.sh f1.9c5mg2dm outfrq3s_macmic2_classnuc 9s .....................................FAIL! rc= 7 at Wed Mar 4 19:07:49 MST 2015
053 bl734 TBL.sh h16c5naqdm outfrq3s_bam+aquaplanet_cam5 9s ...................................FAIL! rc= 7 at Wed Mar 4 19:26:03 MST 2015
056 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Wed Mar 4 19:34:56 MST 2015
062 bl471 TBL.sh f10c5wmam3dm outfrq3s_modalstrat 9s ..........................................FAIL! rc= 7 at Wed Mar 4 19:59:54 MST 2015
065 bl805 TBL.sh f4c5carmtt1dm outfrq3s_carma 9s ..............................................FAIL! rc= 7 at Wed Mar 4 20:22:24 MST 2015
goldbach/pgi or jaguar/pgi:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s ...........................................................FAIL! rc= 7 at Wed Mar 4 17:12:05 MST 2015
010 bl112 TBL.sh e8c5dm ghgrmp 9s .............................................................FAIL! rc= 7 at Wed Mar 4 17:18:30 MST 2015
012 bl113 TBL.sh e8c5paqdm aqpgro+aquaplanet_cam5 3s ..........................................FAIL! rc= 7 at Wed Mar 4 17:23:24 MST 2015
016 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Wed Mar 4 17:24:56 MST 2015
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Wed Mar 4 17:36:24 MST 2015
031 bl312 TBL.sh f2.5c5dm ghgrmp 9s ...........................................................FAIL! rc= 7 at Wed Mar 4 18:01:30 MST 2015
041 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .........................................FAIL! rc= 7 at Wed Mar 4 18:23:47 MST 2015
044 bl317 TBL.sh f10c5cdm outfrq3s 9s .........................................................FAIL! rc= 7 at Wed Mar 4 18:31:13 MST 2015
048 bl319 TBL.sh f10c5nonedm outfrq3s_bam 9s ..................................................FAIL! rc= 7 at Wed Mar 4 18:37:19 MST 2015
051 bl320 TBL.sh f10c5nonedm rad_diag 9s ......................................................FAIL! rc= 7 at Wed Mar 4 18:38:54 MST 2015
054 bl321 TBL.sh f10c5cdm atrain 9s ...........................................................FAIL! rc= 7 at Wed Mar 4 18:40:57 MST 2015
057 bl367 TBL.sh f1.9c5mg2dm outfrq3s 9s ......................................................FAIL! rc= 7 at Wed Mar 4 19:10:54 MST 2015
060 bl420 TBL.sh f4c5mam7dm outfrq3s 9s .......................................................FAIL! rc= 7 at Wed Mar 4 19:24:03 MST 2015
063 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Wed Mar 4 19:35:47 MST 2015
066 bl734 TBL.sh h16c5naqdm outfrq3s_bam+aquaplanet_cam5 9s ...................................FAIL! rc= 7 at Wed Mar 4 20:01:51 MST 2015
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- all cam5
- what platforms/compilers: all
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
- Number 1(listed in Purpose of Changes) is roundoff-level
- Number 2 has a small effect on precipitation, but does not seem to significantly impact climate.
===============================================================
===============================================================
Tag name: cam5_3_69
Originator(s): bsander, santos, cacraig
Date: March 4, 2015
One-line Summary: Bug fix for exponentials in rrtmg_sw_reftra
Purpose of changes:
- The use of low precision lookup tables for exponential terms in
the clear sky reflectivity calculation was determined to be the
cause of rare occurances of extreme hot surface
temperatures (exceeding 100 C). The fix is to replace the
lookup table by double precision exponentials. We attempted to
reduce the cost of the exponential calculation by using a
single precision version of the function. This attempt failed
due to argument values outside the domain of the single
precision exp function. The initial 5 day timings indicated
maybe a 1% slowdown, but timings from the 5 year runs indicated
that any difference in time was in the run-time noise.
Bugs fixed (include bugzilla ID):
- Bug when using the exponential lookup table was removed
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton (through verbal discussions)
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
- using the updated rrtmg external library which contains the bug fix
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
012 bl322 TBL.sh f1.9c5nonedh outfrq3s_lb0 9s .................................................FAIL! rc= 7 at Tue Mar 3 19:17:02 MST 2015
024 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...................................................FAIL! rc= 7 at Tue Mar 3 19:39:27 MST 2015
032 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Tue Mar 3 20:04:44 MST 2015
038 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...............................................FAIL! rc= 7 at Tue Mar 3 20:22:25 MST 2015
049 bl421 TBL.sh f1.9c5mam4dh outfrq3s 9s .....................................................FAIL! rc= 7 at Tue Mar 3 20:53:08 MST 2015
051 bl430 TBL.sh f1.9c5mzmam3dh outfrq3s_NEUwetdep 9s .........................................FAIL! rc= 7 at Tue Mar 3 21:02:01 MST 2015
057 bl474 TBL.sh f1.9c5wtmam3h outfrq3s+1850_waccm_tsmlt_cam5 9s ..............................FAIL! rc= 7 at Tue Mar 3 21:30:11 MST 2015
060 bl482 TBL.sh f1.9c5wscmam3dh outfrq3s_2005+waccm_sc_1955-2005_cam5 9s .....................FAIL! rc= 7 at Tue Mar 3 21:53:51 MST 2015
076 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ........................................FAIL! rc= 7 at Tue Mar 3 22:33:06 MST 2015
081 bl740 TBL.sh h30c5h outfrq3s_lb0 9s .......................................................FAIL! rc= 7 at Tue Mar 3 22:36:34 MST 2015
092 bl010 TBL.sh f4c5portdh outfrq24h_port 2d .................................................FAIL! rc= 7 at Tue Mar 3 23:02:20 MST 2015
goldbach/nag:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s ...........................................................FAIL! rc= 7 at Tue Mar 3 16:36:03 MST 2015
010 bl112 TBL.sh e8c5dm ghgrmp 9s .............................................................FAIL! rc= 7 at Tue Mar 3 16:46:03 MST 2015
014 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Tue Mar 3 16:52:41 MST 2015
018 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Tue Mar 3 17:02:31 MST 2015
026 bl312 TBL.sh f2.5c5dm ghgrmp 9s ...........................................................FAIL! rc= 7 at Tue Mar 3 17:27:43 MST 2015
034 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .........................................FAIL! rc= 7 at Tue Mar 3 17:40:41 MST 2015
040 bl324 TBL.sh f4c4rrtmgdm outfrq3s 9s ......................................................FAIL! rc= 7 at Tue Mar 3 17:59:40 MST 2015
043 bl325 TBL.sh f10c5t5mdmsc outfrq3s_subcol 9s ..............................................FAIL! rc= 7 at Tue Mar 3 18:06:59 MST 2015
050 bl368 TBL.sh f1.9c5mg2dm outfrq3s_macmic2_classnuc 9s .....................................FAIL! rc= 7 at Tue Mar 3 18:39:50 MST 2015
053 bl734 TBL.sh h16c5naqdm outfrq3s_bam+aquaplanet_cam5 9s ...................................FAIL! rc= 7 at Tue Mar 3 19:01:42 MST 2015
056 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Tue Mar 3 19:12:19 MST 2015
062 bl471 TBL.sh f10c5wmam3dm outfrq3s_modalstrat 9s ..........................................FAIL! rc= 7 at Tue Mar 3 19:36:02 MST 2015
065 bl805 TBL.sh f4c5carmtt1dm outfrq3s_carma 9s ..............................................FAIL! rc= 7 at Tue Mar 3 20:00:51 MST 2015
068 bl011 TBL.sh f4c5portdm outfrq24h_port 2d .................................................FAIL! rc= 7 at Tue Mar 3 20:09:15 MST 2015
goldbach/pgi or jaguar/pgi:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s ...........................................................FAIL! rc= 7 at Tue Mar 3 16:36:53 MST 2015
010 bl112 TBL.sh e8c5dm ghgrmp 9s .............................................................FAIL! rc= 7 at Tue Mar 3 16:46:56 MST 2015
012 bl113 TBL.sh e8c5paqdm aqpgro+aquaplanet_cam5 3s ..........................................FAIL! rc= 7 at Tue Mar 3 16:52:58 MST 2015
016 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Tue Mar 3 16:54:58 MST 2015
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Tue Mar 3 17:11:08 MST 2015
031 bl312 TBL.sh f2.5c5dm ghgrmp 9s ...........................................................FAIL! rc= 7 at Tue Mar 3 17:39:11 MST 2015
041 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .........................................FAIL! rc= 7 at Tue Mar 3 18:02:51 MST 2015
044 bl317 TBL.sh f10c5cdm outfrq3s 9s .........................................................FAIL! rc= 7 at Tue Mar 3 18:10:30 MST 2015
048 bl319 TBL.sh f10c5nonedm outfrq3s_bam 9s ..................................................FAIL! rc= 7 at Tue Mar 3 18:16:54 MST 2015
051 bl320 TBL.sh f10c5nonedm rad_diag 9s ......................................................FAIL! rc= 7 at Tue Mar 3 18:18:34 MST 2015
054 bl321 TBL.sh f10c5cdm atrain 9s ...........................................................FAIL! rc= 7 at Tue Mar 3 18:21:08 MST 2015
057 bl367 TBL.sh f1.9c5mg2dm outfrq3s 9s ......................................................FAIL! rc= 7 at Tue Mar 3 18:52:29 MST 2015
060 bl420 TBL.sh f4c5mam7dm outfrq3s 9s .......................................................FAIL! rc= 7 at Tue Mar 3 19:06:36 MST 2015
063 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Tue Mar 3 19:18:40 MST 2015
066 bl734 TBL.sh h16c5naqdm outfrq3s_bam+aquaplanet_cam5 9s ...................................FAIL! rc= 7 at Tue Mar 3 19:42:51 MST 2015
068 bl011 TBL.sh f4c5portdm outfrq24h_port 2d .................................................FAIL! rc= 7 at Tue Mar 3 19:49:41 MST 2015
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: rrtmg radiation (cam5)
- what platforms/compilers: all
- nature of change (roundoff; larger than roundoff but same climate; new
climate): Changes all implementations which use the rrtmg implementation of radiation (cam5). According to Rich Neale who inspected the 5 year comparison runs, the "simulations look pretty near identical"
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
* https://svn-ccsm-models.cgd.ucar.edu/cam1/branch_tags/reftra_lookup_fix_tags/reftra_lookup_fix_n03_cam5_3_66
- platform/compilers: yellowstone / intel
- configure commandline: create_newcase -case /glade/scratch/cacraig/cam_dev/cases/cam5_3_66_reftra_5years -mach yellowstone -res f09_f09 -compset FC5 -project P93300606
- build-namelist command (or complete namelist): default namelist aside from setting stop option to 1 year and resubmitting 4 times
- MSS location of output:
/home/cacraig/csm/cam5_3_66_reftra_5years
MSS location of control simulations used to validate new climate:
/home/cacraig/csm/cam5_3_66_test_5years
URL for AMWG diagnostics output used to validate new climate:
http://webext.cgd.ucar.edu/FCLIMO/cam5_3_66_reftra_5years-cam5_3_66_test_5years/
===============================================================
===============================================================
Tag name: cam5_3_68
Originator(s): Natalie Mahowald, Steve Ghan, Xiaohong Liu, Po-Lun Ma, Cecile Hannay
Date: Mar 2, 2015
One-line Summary: new soil erodibility map; new optical properties for MAM3 and MAM4; changes ice nucleation for MAM4; tuning for cam5/mg2 (cldfrc_rhminl)
Purpose of changes:
- Soil erodibility map, tuned following Albani et al., 2014 to best match observations.
- New optical properties with less absorbing optics for MAM3 and MAM4 (use aeronet dust optics and dust in the aitken mode 2).
- Changes to make the ice nucleation compatible with MAM4
- Tuning for cam5/mg2 (cldfrc_rhminl)
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton, cacraig
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
M models/atm/cam/bld/namelist_files/use_cases/1850-2005_cam5.xml
M models/atm/cam/bld/namelist_files/use_cases/1850_cam5.xml
M models/atm/cam/bld/build-namelist
M models/atm/cam/src/physics/cam/hetfrz_classnuc_cam.F90
M models/atm/cam/src/chemistry/modal_aero/dust_model.F90
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
009 bl203 TBL.sh fsd1.9c4wtsmlth outfrq3s_sd+sd_1975-2010_ccmi_refc1_waccm_tsmlt 9s ...........FAIL! rc= 7 at Mon Mar 2 10:38:06 MST 2015
012 bl322 TBL.sh f1.9c5nonedh outfrq3s_lb0 9s .................................................FAIL! rc= 7 at Mon Mar 2 10:38:13 MST 2015
024 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...................................................FAIL! rc= 7 at Mon Mar 2 10:38:17 MST 2015
032 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Mon Mar 2 10:38:28 MST 2015
036 bl373 TBL.sh f1.9c4mozdh outfrq3s+2000_cam4_trop_chem 9s ..................................FAIL! rc= 7 at Mon Mar 2 10:38:41 MST 2015
038 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...............................................FAIL! rc= 7 at Mon Mar 2 10:41:14 MST 2015
040 bl376 TBL.sh f1.9c4mozdh+testmech outfrq3s 9s .............................................FAIL! rc= 7 at Mon Mar 2 10:41:25 MST 2015
049 bl421 TBL.sh f1.9c5mam4dh outfrq3s 9s .....................................................FAIL! rc= 7 at Mon Mar 2 10:41:33 MST 2015
051 bl430 TBL.sh f1.9c5mzmam3dh outfrq3s_NEUwetdep 9s .........................................FAIL! rc= 7 at Mon Mar 2 10:41:48 MST 2015
054 bl440 TBL.sh f1.9c4soadh outfrq3s+soa_chem_megan_emis 9s ..................................FAIL! rc= 7 at Mon Mar 2 10:42:16 MST 2015
057 bl474 TBL.sh f1.9c5wtmam3h outfrq3s+1850_waccm_tsmlt_cam5 9s ..............................FAIL! rc= 7 at Mon Mar 2 10:44:26 MST 2015
060 bl482 TBL.sh f1.9c5wscmam3dh outfrq3s_2005+waccm_sc_1955-2005_cam5 9s .....................FAIL! rc= 7 at Mon Mar 2 10:44:51 MST 2015
076 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ........................................FAIL! rc= 7 at Mon Mar 2 10:44:58 MST 2015
081 bl740 TBL.sh h30c5h outfrq3s_lb0 9s .......................................................FAIL! rc= 7 at Mon Mar 2 10:45:21 MST 2015
092 bl010 TBL.sh f4c5portdh outfrq24h_port 2d .................................................FAIL! rc= 7 at Mon Mar 2 10:45:22 MST 2015
goldbach/nag:
001 fm001 TFM.sh ..............................................................................FAIL! rc= 2 at Fri Feb 27 14:28:21 MST 2015
- This was expected to fail as this tag was checked out on yellowstone and the subversion version is not compatible with goldbach
005 bl111 TBL.sh e8c5t5mdm ttrac 9s ...........................................................FAIL! rc= 7 at Fri Feb 27 14:34:00 MST 2015
010 bl112 TBL.sh e8c5dm ghgrmp 9s .............................................................FAIL! rc= 7 at Fri Feb 27 14:37:36 MST 2015
014 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Fri Feb 27 14:41:19 MST 2015
018 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Fri Feb 27 14:48:18 MST 2015
026 bl312 TBL.sh f2.5c5dm ghgrmp 9s ...........................................................FAIL! rc= 7 at Fri Feb 27 15:12:17 MST 2015
043 bl325 TBL.sh f10c5t5mdmsc outfrq3s_subcol 9s ..............................................FAIL! rc= 7 at Fri Feb 27 15:44:18 MST 2015
050 bl368 TBL.sh f1.9c5mg2dm outfrq3s_macmic2_classnuc 9s .....................................FAIL! rc= 7 at Fri Feb 27 16:19:07 MST 2015
056 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Fri Feb 27 16:52:26 MST 2015
062 bl471 TBL.sh f10c5wmam3dm outfrq3s_modalstrat 9s ..........................................FAIL! rc= 7 at Fri Feb 27 17:11:25 MST 2015
065 bl805 TBL.sh f4c5carmtt1dm outfrq3s_carma 9s ..............................................FAIL! rc= 7 at Fri Feb 27 18:12:05 MST 2015
068 bl011 TBL.sh f4c5portdm outfrq24h_port 2d .................................................FAIL! rc= 7 at Fri Feb 27 18:20:30 MST 2015
goldbach/pgi or jaguar/pgi:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s ...........................................................FAIL! rc= 7 at Fri Feb 27 14:38:24 MST 2015
010 bl112 TBL.sh e8c5dm ghgrmp 9s .............................................................FAIL! rc= 7 at Fri Feb 27 14:45:17 MST 2015
016 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Fri Feb 27 14:52:43 MST 2015
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Fri Feb 27 15:04:51 MST 2015
031 bl312 TBL.sh f2.5c5dm ghgrmp 9s ...........................................................FAIL! rc= 7 at Fri Feb 27 15:34:52 MST 2015
044 bl317 TBL.sh f10c5cdm outfrq3s 9s .........................................................FAIL! rc= 7 at Fri Feb 27 16:05:07 MST 2015
054 bl321 TBL.sh f10c5cdm atrain 9s ...........................................................FAIL! rc= 7 at Fri Feb 27 16:16:23 MST 2015
057 bl367 TBL.sh f1.9c5mg2dm outfrq3s 9s ......................................................FAIL! rc= 7 at Fri Feb 27 16:51:38 MST 2015
060 bl420 TBL.sh f4c5mam7dm outfrq3s 9s .......................................................FAIL! rc= 7 at Fri Feb 27 17:06:42 MST 2015
063 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Fri Feb 27 17:20:14 MST 2015
068 bl011 TBL.sh f4c5portdm outfrq24h_port 2d .................................................FAIL! rc= 7 at Fri Feb 27 18:21:20 MST 2015
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
- new climate
AMWG diagnostics for the new climate:
http://webext.cgd.ucar.edu/FCLIMO/f.e13.F2000C5.f09_f09_cam5.3.68.001/atm/f.e13.F2000C5.f09_f09_cam5.3.68.001-obs/
http://webext.cgd.ucar.edu/FCLIMO/f.e13.F2000C5.f09_f09_cam5.3.68.001/atm/f.e13.F2000C5.f09_f09_cam5.3.68.001-f.e13.F2000C5.f09_f09.cam5.3.release.001/
===============================================================
===============================================================
Tag name: cam5_3_67
Originator(s): olson, cacraig
Date: Feb 25, 2015
One-line Summary: Reformulate total energy as applied in physics_update and the energy fixers for CAM-FV and CAM-SE dycores.
Purpose of changes:
- To "correct" the total energy definition in scattered regions of the model
Bugs fixed (include bugzilla ID):
- 2148: lengthened character variable "tcase" to handle longer filenames
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Eaton, Santos
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/src/advection/slt/engy_te.F90
- modify total energy calculation
M models/atm/cam/src/control/cam_restart.F90
- fixed bug #2148 lengthened character variable "tcase"
M models/atm/cam/src/dynamics/eul/dp_coupling.F90
- check for neg tracers
M models/atm/cam/src/dynamics/eul/linemsdyn.F90
M models/atm/cam/src/dynamics/eul/spegrd.F90
M models/atm/cam/src/dynamics/fv/dp_coupling.F90
- check for neg tracers
M models/atm/cam/src/dynamics/se/dp_coupling.F90
- check for neg tracers
M models/atm/cam/src/dynamics/sld/dp_coupling.F90
- check for neg tracers
M models/atm/cam/src/dynamics/sld/linemsdyn.F90
M models/atm/cam/src/dynamics/sld/spegrd.F90
M models/atm/cam/src/physics/cam/cam_diagnostics.F90
- modify some diagnostics
M models/atm/cam/src/physics/cam/check_energy.F90
- modify total energy for energy conservation check
M models/atm/cam/src/physics/cam/cloud_diagnostics.F90
- fix diagnostics bug
M models/atm/cam/src/physics/cam/modal_aer_opt.F90
- clerical corrections
M models/atm/cam/src/physics/cam/physics_types.F90
- update temperature based on new energy definition
M models/atm/cam/src/physics/cam/physpkg.F90
- remove Boville heating kluge for FV
- fix DTCORE diagnostic
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all TBL tests *FAILED* due to answer changes by energy fix
goldbach/nag: all TBL tests *FAILED* due to answer changes by energy fix
goldbach/pgi or jaguar/pgi: all TBL tests *FAILED* due to answer changes by energy fix
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: all
- what platforms/compilers: all
- nature of change (roundoff; larger than roundoff but same climate; new
climate): larger than roundoff but same climate
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
http://www.cesm.ucar.edu/working_groups/Atmosphere/development/cam6/eb/
===============================================================
===============================================================
Tag name: cam5_3_66
Originator(s): cacraig, eaton, bogensch
Date: Feb 13, 2015
One-line Summary: Various updates/fixes and update externals to alpha17d
Purpose of changes:
- update to alpha17d externals
- time_period_freq attribute is now added to history files
- add history_budget changes to CLUBB (per Pete's email)
- fixed bug when running SCAM with history_budget=true and ndens=1(default)
- timing mods now that GPTL is its own directory (eaton)
- allow CLUBB to be run threaded when using clubb_history
Bugs fixed (include bugzilla ID):
- 1727: CLUBB_history is now threadsafe so remove check prohibiting this in configure
- fixed bug when running SCAM with history_budget=true and the default setting of ndens=1
Describe any changes made to build system:
Describe any changes made to the namelist:
- clubb_history and threading is now permitted
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
A models/atm/cam/test/system/nl_files/outfrq1s_hist
A models/atm/cam/test/system/nl_files/outfrq1s_carma
A models/atm/cam/test/system/nl_files/outfrq1s_clubb
- namelists turning on histories and forcing conversion to trop any fields not set, but are written out (cacraig)
List all existing files that have been modified, and describe the changes:
M .
M SVN_EXTERNAL_DIRECTORIES
- update to beta17 externals
M models/atm/cam/test/system/tests_pretag_goldbach_pgi
M models/atm/cam/test/system/tests_pretag_yellowstone
M models/atm/cam/test/system/tests_pretag_goldbach_nag
M models/atm/cam/test/system/nl_files/scmarm
M models/atm/cam/test/system/input_tests_master
- added "hi" tests to turn on the history flags and force conversion to trap any values not set but are output (cacraig)
M models/atm/cam/test/system/TSM_ccsm.sh
- "SUCCESSFUL TERMINATION" is no longer echoed to standard out, so now grep for the success string in atm.log (cacraig)
M models/atm/cam/bld/configure
M models/atm/cam/bld/Makefile.in
- mods to how timing is built since GPTL is its own directory (eaton)
M models/atm/cam/bld/build-namelist
- timing mods since GPTL is its own directory (eaton)
- clubb_history is now threadsafe. Removed check preventing threading with clubb_history=true (cacraig)
M models/atm/cam/src/control/cam_history.F90
- added time_per_freq attribute (cacraig)
M models/atm/cam/src/physics/cam/clubb_intr.F90
- added history_budget to CLUBB (bogensch)
M models/atm/cam/src/physics/cam/phys_control.F90
- fixed typo (bogensch)
M models/atm/cam/src/dynamics/eul/dyn_comp.F90
- fixed history_budget when running single column mode (cacraig)
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB
goldbach/nag: all BFB
goldbach/pgi or jaguar/pgi: all BFB
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_65
Originator(s): bogensch, cacraig
Date: February 4, 2015
One-line Summary: Updated release of CLUBB
Purpose of changes:
- Provide updates to CAM-CLUBB to perform scientifically similar to the candidate CAM-CLUBB simulations
(NOTE: no tuning has yet been performed to account for recent CAM5 answer changes committed to the
trunk). Includes scientific support for CAM-CLUBB to run with prognostic precipitation (MG2), which is
now the default microphysics for CAM-CLUBB simulations. The option has been added to provide the
capability for CLUBB's turbulent moments to be advected by dynamics. In addition, updates have been
made to allow CAM-CLUBB to run as a deep convection scheme, as an option. Outside of CLUBB, other
options have been made available. The first being the option to allow the Zhang McFarlane (ZM) deep
convection scheme to run with the organization parameterization based on Mapes and Neale (2010). The
second being the option for an ice-supersaturation adjustment closure.
- fixed missing character in waccmx use case (fix from Francis Vitt)
- additional error messages in physics_buffer to help give more information if buffer_field is not allocated
- mo_neu_wetdep was causing a floating point exception when RCA = 0 - added a check to prevent this
Bugs fixed (include bugzilla ID):
- bugzilla 2127 (format error in waccmxi_2000_cam4.xml)
Describe any changes made to build system:
- Added clubb_opts: clubb_do_adv (Advect CLUBB moments), clubb_do_deep(CLUBB does the deep convection)
- Added zmconv_org: Include parameterization for sub-grid scale convective organization for the
ZM deep convective scheme based on Mapes and Neale (2011)
Describe any changes made to the namelist:
Added:
- cldfrc2m_rhmaxi(rhi at which ice cloud fraction = 1)
- micro_mg_dcs (microphysics autoconversion size threshold)
- micro_mg_precip_frac_method (micorphysics type of precipitation fraction - max_overlap or in_cloud)
- micro_mg_berg_eff_factor (microphysics efficiency factor for berg)
- micro_do_icesupersat (apply ice supersaturation adjustment code)
- zmconv_ke_lnd (Tunable evaporation efficiency for land in ZM deep convection scheme)
- zmconv_org (organization parameterization in ZM)
- deep_scheme (CLUBB_SGS is now a valid option for this flag)
- clubb_cloudtop_cooling (CLUBB - turns on/off cloud top radiative cooling parameterization in CLUBB)
- clubb_rainevap_turb (CLUBB - turns on/off including effects of precip evaporation on turbulent moments
- clubb_expldiff (CLUBB - turns on/off explicit diffusion on temperature and moisture when CLUBB is on
- clubb_do_adv (CLUBB - on/off switch for advecting CLUBB moments)
- clubb_do_deep (CLUBB - on/off switch for having CLUBB calculate the deep convection)
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: CAM software engineers at code reviews
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M .
M models/atm/cam
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
- updated to new CLUBB external library and added SILHS library
M models/atm/cam/bld/configure
M models/atm/cam/bld/config_files/definition.xml
- changes to support new options: clubb_opts(clubb_do_deep and clubb_do_adv) and zmconv_org
- set default microphys_pkg to MG2 when CLUBB is turned on
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
M models/atm/cam/bld/namelist_files/namelist_definition.xml
M models/atm/cam/bld/build-namelist
- changes to support new namelist variables:
clubb_rainevap_turb, clubb_cloudtop_cooling, clubb_expldiff,
micro_mg_dcs, micro_mg_precip_frac_method, micro_mg_berg_eff_factor, micro_do_icesupersat,
cldfrc2m_rhmaxi, zmconv_org
- new settings for CLUBB of nucleate_ice_subgrid, sol_facti_cloud_borne, deep_scheme, zmconv_c0_lnd,
zmconv_ke_lnd, cld_macmic_num_steps, micro_mg_num_steps
M models/atm/cam/bld/namelist_files/use_cases/waccmxi_2000_cam4.xml
- Added missing character (at Francis Vitt request)
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
- updated CLUBB and added SILHS external libraries
M models/atm/cam/src/control/runtime_opts.F90
- clubb now has a namelist read
- zmconv namelist read is now in zm_conv_intr
M models/atm/cam/src/physics/cam/micro_mg2_0.F90
- prer_evap is added as a new field to micro_mg_tend
- uses namelist fields micro_mg_dcs, micro_mg_precip_frac_method and micro_mg_berg_eff_factor
- bug with negative temperature causing a core dump was solved by moving a section of code below conservation checks
M models/atm/cam/src/physics/cam/microp_aero.F90
- WP2 is now more correctly named WP2_nadv
- use micro_do_icesupersat flag
M models/atm/cam/src/physics/cam/micro_mg_cam.F90
M models/atm/cam/src/physics/cam/micro_mg1_0.F90
M models/atm/cam/src/physics/cam/micro_mg1_5.F90
- uses namelist fields micro_mg_dcs, micro_mg_precip_frac_method and micro_mg_berg_eff_factor
- prer_evap is added as a new field
M models/atm/cam/src/physics/cam/radsw.F90
M models/atm/cam/src/physics/cam/physics_types.F90
- whitespace changes only
M models/atm/cam/src/physics/cam/cldfrc2m.F90
- added cldfrc2m_rhmaxi to namelist
- ice water number is now passed into the aist calculations
- added code for new iceopt=6
M models/atm/cam/src/physics/cam/clubb_intr.F90
- all of the changes to support the details listed in the purpose section above
M models/atm/cam/src/physics/cam/convect_shallow.F90
- added new variable landfracdum
M models/atm/cam/src/physics/cam/pbl_utils.F90
- new compute_radf routine which was originally a section of code in eddy_diff
M models/atm/cam/src/physics/cam/zm_conv.F90
M models/atm/cam/src/physics/cam/zm_conv_intr.F90
- added logic for organization parameterization detailed above in the purpose section
- moved the namelist out from zm_conv into the interface (zm_conv_intr) module
- passed namelist variables into the zm init routine (zm_convi)
M models/atm/cam/src/physics/cam/eddy_diff.F90
- uses new compute_radf routine
M models/atm/cam/src/physics/cam/phys_control.F90
- added micro_do_icesupersat control variable
- added support for macrop_scheme being CLUBB
M models/atm/cam/src/physics/cam/physpkg.F90
- added code based on using micro_d_icesupersat
- pass in additional parameters to clubb_tend_cam to support using macmic logic
M models/atm/cam/src/physics/cam/macrop_driver.F90
- changes to support micro_do_icesupersat
M models/atm/cam/src/physics/cam/cldwat2m_macro.F90
- rhmaxi and ni1 are now passed to lower level routines
M models/atm/cam/src/physics/cam/convect_deep.F90
- support for CLUBB_SGS being a valid deep_scheme
M models/atm/cam/src/physics/cam/physics_buffer.F90.in
- additional error messages to help give more information if buffer_field is not allocated
M models/atm/cam/src/chemistry/mozart/mo_neu_wetdep.F90
- added check for RCA > zero as this was causing a floating point exception
M models/atm/cam/src/dynamics/sld/inidat.F90
M models/atm/cam/src/dynamics/eul/inidat.F90
M models/atm/cam/src/dynamics/se/inidat.F90
M models/atm/cam/src/dynamics/fv/inidat.F90
- call clubb_init_cnst for clubb constituents
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
- updated Machines branch to update Intel compiler to 15.0.1 (15.0.0 was giving CLUBB/MG2 issues)
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All BFB except:
032 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Tue Feb 3 21:31:15 MST 2015
This CLUBB baseline was expected to fail as updates to CLUBB are answer changing
goldbach/nag: All BFB except:
049 bl368 TBL.sh f1.9c5mg2dm outfrq3s_macmic2_classnuc 9s .....................................FAIL! rc= 7 at Tue Feb 3 19:16:12 MST 2015
Change to MG2 to fix negative temperature bug is expected to changes answers for MG2
goldbach/pgi or jaguar/pgi: All BFB except:
056 bl367 TBL.sh f1.9c5mg2dm outfrq3s 9s ......................................................FAIL! rc= 7 at Tue Feb 3 20:01:28 MST 2015
Change to MG2 to fix negative temperature bug is expected to changes answers for MG2
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
Changes for CLUBB - improved physics produces different climate just for CLUBB
Changes for MG2 - larger than roundoff but same climate - confirmed by Andrew Gettelman
If bitwise differences were observed, how did you show they were no worse:
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
CLUBB answers not strictly controlled yet, but was evaluated extensivly by Pete Bogenschutz
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_64
Originator(s): sungsup, eaton
Date: Tue Jan 27 09:21:25 MST 2015
One-line Summary: Update UNICON interfaces; add park macrophysics option.
Purpose of changes:
. Update the UNICON interface to put convective detrainment fields in the
pbuf.
. Add option to Park macrophysics to compute variations in the rhminl,
rhminh, rhmini parameters using information about convective detrainment
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist:
. add cldfrc2m_rhmini used by routines computing cloud fractions for ice
clouds
. add logical variable (offline_driver) to indicate model is configured to
use offline driver. This is set automatically by build-namelist based on
the model configuration.
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: cam developers
List all subroutines eliminated: none
List all subroutines added and what they do:
models/atm/cam/src/physics/cam/cldfrc2m.F90
. Move the cloud fraction subroutines out of cldwat2m.F90 and into this new
module. The parameter rhmini is maintained here, and is set by the
namelist variable cldfrc2m_rhmini.
. These routines have all been modified to accept optional arguments
allowing rhminl, rhminh, rhmini to vary spatially
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/build-namelist
. set default for cldfrc2m_rhmini
. set new variable, offline_driver, to true when CAM has been configured to
use an offline driver
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. add default for cldfrc2m_rhmini
models/atm/cam/bld/namelist_files/namelist_definition.xml
. add cldfrc2m_rhmini
. add logical variable (offline_driver) to indicate model is configured to
use offline driver
models/atm/cam/src/control/runtime_opts.F90
. add calls for cldfrc2m and unicon namelist readers
models/atm/cam/src/physics/cam/cldwat2m_macro.F90
. add new options to compute variation in rhminl, rhminh, rhmini using
information on convective detrainment
. move the routines that compute cloud fraction to new module, cldfrc2m.
models/atm/cam/src/physics/cam/clubb_intr.F90
models/atm/cam/src/physics/cam/micro_mg_cam.F90
models/atm/cam/src/physics/carma/models/cirrus_dust/carma_cloudfraction.F90
models/atm/cam/src/physics/carma/models/cirrus/carma_cloudfraction.F90
. change use association of cldwat2m_macro to cldfrc2m
models/atm/cam/src/physics/cam/conv_water.F90
. add diagnostic fields
models/atm/cam/src/physics/cam/macrop_driver.F90
. call Park macrophysics init routine (ini_macro) from macrop_driver_init
. add logic to access an pass new pbuf fields to cldwat2m_macro routines
when unicon is enabled
models/atm/cam/src/physics/cam/microp_driver.F90
. remove call to ini_macro (it was moved to macrop_driver_init)
models/atm/cam/src/physics/cam/phys_control.F90
. add offline_driver to phys_ctl_nl
models/atm/cam/src/physics/cam/physpkg.F90
. add call to cldfrc2m_init
models/atm/cam/src/physics/cam/qneg4.F90
. fix format statement
models/atm/cam/src/physics/cam/unicon_cam.F90
. add convective detrainment quantities to pbuf
models/atm/cam/src/physics/cam/vertical_diffusion.F90
. add fields to the pbuf for use in new Park macrophysics options (only
enabled when UNICON is on)
models/atm/cam/src/unit_drivers/drv_input_data.F90
. remove use of subroutine getfil. This subroutine looks for files in the
working directory before using the file specified by the namelist with an
absolute pathname. This may be not be what you want.
. add better error messages when reading input file.
models/atm/cam/test/system/test_driver.sh
. Set ACCOUNT environment variable for yellowstone testing.
. update intel/15.0.0 to intel/15.0.1 on yellowstone
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS
goldbach/nag: All PASS
goldbach/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_3_63
Originator(s): cacraig, santos, eaton, jedwards
Date: January 14, 2015
One-line Summary: longer casenames, MG2 change to avoid FPE, turn off use_contiguous for gfortran, added missing & which caused issues on mira
Purpose of changes:
- Need to allow longer casenames for CESM runs
- micro_mg_cam needed to set entire arrays to avoid a floating point exception in later code
- when using gfortran 4.8.3, got a segmentation fault in decomp_left_div when USE_CONTIGUOUS is turned on
- added missing & on a continuation line which caused issues on mira
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/Makefile.in
- turn off USE_CONTIGUOUS for gfortran
M models/atm/cam/src/control/filenames.F90
- use the longer casename lengths
M models/atm/cam/src/physics/cam/micro_mg_cam.F90
- set entire arrays so don't get a segmentation fault in later code
M models/atm/cam/src/physics/cam/chem_surfvals.F90
- added missing & which caused issues on mira
M .
M SVN_EXTERNAL_DIRECTORIES
- update to csm_share version which supports the longer casename setting
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB except which differ due to the updated csm_share external:
072 bl731 TBL.sh h16c3aqdh aqua+aquaplanet_cam3 9s ............................................FAIL! rc= 7 at Tue Jan 13 10:16:33 MST 2015
075 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ........................................FAIL! rc= 7 at Tue Jan 13 10:16:40 MST 2015
080 bl740 TBL.sh h30c5h outfrq3s_lb0 9s .......................................................FAIL! rc= 7 at Tue Jan 13 10:17:03 MST 2015
084 bl751 TBL.sh h30c4aqdm outfrq3s+aquaplanet_cam4 9s ........................................FAIL! rc= 7 at Tue Jan 13 10:17:15 MST 2015
Created baseline files with the same csm_share external as used in cam5_3_63 and these 4 tests are BFB
goldbach/nag: all BFB
goldbach/pgi or jaguar/pgi: all BFB
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate): roundoff due to changes with DNO_SHR_VMATH
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_62
Originator(s): joemci
Date: 10 Jan 2015
One-line Summary: Fix for cam5_3_61
Purpose of changes:
Fix error in cam5_3_61 commit in short_lived_species.F90
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton, CAM code reviewers
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/src/chemistry/mozart/short_lived_species.F90
- Remove extraneous endif
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
045 bl390 TBL.sh f1.9c4wmxh outfrq3s+waccmx_2000_cam4 9s ......................................FAIL! rc= 7 at Fri Jan 9 19:20:48 MST 2015
- expected failure -- added a third airglow heating term in mo_aiglow.F90 and
changed from constant to variable mean mass in charge_neutrality.F90 which
both change answers for WACCM-X with neutral option
goldbach/nag:
003 bl424 TBL.sh f10c4wmxdm outfrq3s 9s .......................................................FAIL! rc= 7 at Sat Jan 10 12:59:58 MST 2015
- expected failure -- added a third airglow heating term in mo_aiglow.F90 and
changed from constant to variable mean mass in charge_neutrality.F90 which
both change answers for WACCM-X with neutral option
goldbach/pgi: All BFB
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_61
Originator(s): joemci, liuh
Date: 5 Jan 2015
One-line Summary: Update WACCM-X with extended ionosphere
Purpose of changes:
Add capability to switch on vertical ion transport ambipolar diffusion and
electron and ion temperature calculations in WACCM-X. This will give a more
realistic simulation of the ionosphere in the model and is the next step in
adding full thermosphere/ionosphere physics and dynamics to CESM.
Also, added is the infrastructure for the more realistic setting of ions
and electrons to short-lived non-transported species in the thermosphere/
ionosphere but no species are set to short-lived..
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton, CAM code reviewers
List all subroutines eliminated:
List all subroutines added and what they do:
models/atm/cam/bld/namelist_files/use_cases/waccmxi_2000_cam4.xml
- new WACCM-X extended ionosphere use case
models/atm/cam/src/chemistry/pp_waccmx_mozart/chem_mech.doc
models/atm/cam/src/chemistry/pp_waccmx_mozart/chem_mech.in
models/atm/cam/src/chemistry/pp_waccmx_mozart/chem_mods.F90
models/atm/cam/src/chemistry/pp_waccmx_mozart/m_rxt_id.F90
models/atm/cam/src/chemistry/pp_waccmx_mozart/m_spc_id.F90
models/atm/cam/src/chemistry/pp_waccmx_mozart/mo_adjrxt.F90
models/atm/cam/src/chemistry/pp_waccmx_mozart/mo_imp_sol.F90
models/atm/cam/src/chemistry/pp_waccmx_mozart/mo_indprd.F90
models/atm/cam/src/chemistry/pp_waccmx_mozart/mo_lin_matrix.F90
models/atm/cam/src/chemistry/pp_waccmx_mozart/mo_lu_factor.F90
models/atm/cam/src/chemistry/pp_waccmx_mozart/mo_lu_solve.F90
models/atm/cam/src/chemistry/pp_waccmx_mozart/mo_nln_matrix.F90
models/atm/cam/src/chemistry/pp_waccmx_mozart/mo_phtadj.F90
models/atm/cam/src/chemistry/pp_waccmx_mozart/mo_prod_loss.F90
models/atm/cam/src/chemistry/pp_waccmx_mozart/mo_rxt_rates_conv.F90
models/atm/cam/src/chemistry/pp_waccmx_mozart/mo_setrxt.F90
models/atm/cam/src/chemistry/pp_waccmx_mozart/mo_sim_dat.F90
- new WACCM-X MOZART chem package currently identical to WACCM MOZART
List all existing files that have been modified, and describe the changes:
M models/atm/cam/src/physics/cam/physpkg.F90
- changed name of interface routine to extended ionosphere and added
charge neutrality check to update electron number density after
extended ionosphere calculation of O+
M models/atm/cam/src/physics/cam/ionosphere.F90
- changed name of interface routine to extended ionosphere
M models/atm/cam/src/physics/waccmx/ionosphere.F90
- updated algorithms for electron and ion temperature and ion
transport ambipolar diffusion
M models/atm/cam/src/chemistry/mozart/mo_airglow.F90
M models/atm/cam/src/chemistry/mozart/mo_waccm_hrates.F90
- added 630 nm O1D airglow heating rate to total rate and clarified
output field names
M models/atm/cam/src/chemistry/mozart/charge_neutrality.F90
- now ensures charge neutrality for short-lived species in pbuf
- changed mean mass from constant to variable for WACCM-X
M models/atm/cam/src/chemistry/mozart/chemistry.F90
- modified conditionals for chemistry package to include new WACCM-X
MOZART package
M models/atm/cam/src/chemistry/mozart/short_lived_species.F90
- changed from three to two dimensional the temporary pointer which
is used for setting short-lived species in pbuf
M models/atm/cam/src/chemistry/mozart/exbdrift.F90
- uncommented code to calculate vertical ExB drift and add to pbuf
M models/atm/cam/bld/build-namelist
- set namelist defaults for new waccmx_mozart chemistry package
M models/atm/cam/bld/config_files/definition.xml
- updated the chemistry option definition for new waccmx_mozart chemistry
package
M models/atm/cam/bld/configure
- updated the chem option description for new waccmx_mozart chemistry
package
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- set default emission inputs for new waccmx_mozart chemistry package
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- added waccmx_mozart to valid values for cam_chempkg
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
045 bl390 TBL.sh f1.9c4wmxh outfrq3s+waccmx_2000_cam4 9s ......................................FAIL! rc= 7 at Tue Jan 6 22:27:50 MST 2015
- expected failure -- added a third airglow heating term in mo_aiglow.F90 and
changed from constant to variable mean mass in charge_neutrality.F90 which
both change answers for WACCM-X with neutral option
goldbach/nag:
003 bl424 TBL.sh f10c4wmxdm outfrq3s 9s .......................................................FAIL! rc= 7 at Thu Jan 8 13:41:50 MST 2015
- expected failure -- added a third airglow heating term in mo_aiglow.F90 and
changed from constant to variable mean mass in charge_neutrality.F90 which
both change answers for WACCM-X with neutral option
goldbach/pgi: All BFB
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_60
Originator(s): fvitt, mmills
Date: 5 Jan 2015
One-line Summary: Bring in prognostic stratospheric modal aerosols and SC-WACCM5
Purpose of changes:
Provide the capability to prognose modal aerosols in the stratosphere. This
gives CAM5 and WACCM5 the ability to simulate aerosols in the stratosphere
which originate from volcanic eruptions. To this end, accumulation to coarse
mode exchange is allowed and the widths and edges of the modes are modified
Provide new capabilities for specifying emissions (surface and elevated):
- ability to specify emissions from multiple input files for any given species
- optional global attribute 'input_method' (set to: 'SERIAL', 'CYCLICAL',
or 'INTERP_MISSING_MONTHS') in the emissions input file which overrides the
corresponding *type namelist option on a file-by-file basis
- optional multiplier proceeding the emissions filepath, e.g.:
'NAME -> 0.5*/path.../filename.nc'
Provide chemistry package for SC-WACCM5 which has prognostic modal aerosols
(this mechanism lacks OCS which is important for the non-volcanic stratospheric sulfate)
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton, CAM code reviewers
List all subroutines eliminated:
D models/atm/cam/src/chemistry/pp_waccm_ghg
D models/atm/cam/src/chemistry/pp_waccm_ghg/mo_lu_solve.F90
D models/atm/cam/src/chemistry/pp_waccm_ghg/mo_setrxt.F90
D models/atm/cam/src/chemistry/pp_waccm_ghg/mo_adjrxt.F90
D models/atm/cam/src/chemistry/pp_waccm_ghg/mo_rxt_rates_conv.F90
D models/atm/cam/src/chemistry/pp_waccm_ghg/mo_imp_sol.F90
D models/atm/cam/src/chemistry/pp_waccm_ghg/chem_mods.F90
D models/atm/cam/src/chemistry/pp_waccm_ghg/mo_prod_loss.F90
D models/atm/cam/src/chemistry/pp_waccm_ghg/mo_lin_matrix.F90
D models/atm/cam/src/chemistry/pp_waccm_ghg/m_rxt_id.F90
D models/atm/cam/src/chemistry/pp_waccm_ghg/mo_phtadj.F90
D models/atm/cam/src/chemistry/pp_waccm_ghg/mo_nln_matrix.F90
D models/atm/cam/src/chemistry/pp_waccm_ghg/mo_lu_factor.F90
D models/atm/cam/src/chemistry/pp_waccm_ghg/mo_sim_dat.F90
D models/atm/cam/src/chemistry/pp_waccm_ghg/mo_indprd.F90
D models/atm/cam/src/chemistry/pp_waccm_ghg/m_spc_id.F90
D models/atm/cam/src/chemistry/pp_waccm_ghg/chem_mech.doc
D models/atm/cam/src/chemistry/pp_waccm_ghg/chem_mech.in
- renamed 'waccm_ghg' to 'waccm_sc'
D models/atm/cam/test/system/config_files/f1.9c4wghgdm
D models/atm/cam/test/system/config_files/f4c4wghgdm
D models/atm/cam/test/system/config_files/f10c4wghgdm
D models/atm/cam/test/system/config_files/h30c4wgdm
- changed 'wghg' to 'wsc'
D models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/chem_mech.dat
- preprocessor by-product -- not needed in CAM
List all subroutines added and what they do:
A models/atm/cam/test/system/config_files/f1.9c4wscdm
A models/atm/cam/test/system/config_files/f4c4wscdm
A models/atm/cam/test/system/config_files/f10c4wscdm
A models/atm/cam/test/system/config_files/h30c4wscdm
- changed 'wghg' to 'wsc'
A models/atm/cam/test/system/config_files/f1.9c4carmwtsuldh
A models/atm/cam/test/system/config_files/f10c5wtmam3dm
A models/atm/cam/test/system/config_files/f10c5wscmam3dm
A models/atm/cam/test/system/config_files/f1.9c5wmam3h
A models/atm/cam/test/system/config_files/f1.9c5wscmam3dh
A models/atm/cam/test/system/config_files/f1.9c5wscmam3dm
A models/atm/cam/test/system/config_files/f1.9c5wtmam3dh
A models/atm/cam/test/system/config_files/f1.9c4wscdh
A models/atm/cam/test/system/config_files/f10c4wmxdm
A models/atm/cam/test/system/config_files/f10c4carmwtsuldm
A models/atm/cam/test/system/config_files/f10c5wmam3dm
A models/atm/cam/test/system/config_files/f1.9c5wtmam3h
- new test configurations
A models/atm/cam/test/system/tests_waccm_mpi
A models/atm/cam/test/system/tests_waccm_hybrid
- tests many flavors/configurations of waccm
A models/atm/cam/test/system/nl_files/outfrq3s_modalstrat
- for prognostic stratospheric modal aerosols
A models/atm/cam/test/system/nl_files/outfrq3s_carma2000
- for carma-waccm-suflur tests
A models/atm/cam/bld/namelist_files/use_cases/waccm_1955-2005_cam5.xml
A models/atm/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_waccm_tsmlt_cam5.xml
A models/atm/cam/bld/namelist_files/use_cases/1850-2005_waccm_tsmlt_cam5.xml
A models/atm/cam/bld/namelist_files/use_cases/1850_waccm_tsmlt_cam5.xml
- new waccm5 use cases
A models/atm/cam/bld/namelist_files/use_cases/waccm_sc_1955-2005_cam5.xml
A models/atm/cam/bld/namelist_files/use_cases/sd_waccm5_geos5.xml
A models/atm/cam/bld/namelist_files/use_cases/waccm_sc_2000_cam5.xml
- new SC-waccm5 use cases
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_lu_solve.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_setrxt.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_adjrxt.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_rxt_rates_conv.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_imp_sol.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_mam3/chem_mods.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_prod_loss.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_lin_matrix.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_mam3/m_rxt_id.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_phtadj.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_nln_matrix.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_lu_factor.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_sim_dat.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_mam3/mo_indprd.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_mam3/m_spc_id.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_mam3/chem_mech.doc
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_mam3/chem_mech.in
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_mam3
- new chem package
A models/atm/cam/src/chemistry/pp_waccm_sc_mam3/mo_lu_solve.F90
A models/atm/cam/src/chemistry/pp_waccm_sc_mam3/mo_setrxt.F90
A models/atm/cam/src/chemistry/pp_waccm_sc_mam3/mo_adjrxt.F90
A models/atm/cam/src/chemistry/pp_waccm_sc_mam3/mo_rxt_rates_conv.F90
A models/atm/cam/src/chemistry/pp_waccm_sc_mam3/mo_imp_sol.F90
A models/atm/cam/src/chemistry/pp_waccm_sc_mam3/chem_mods.F90
A models/atm/cam/src/chemistry/pp_waccm_sc_mam3/mo_prod_loss.F90
A models/atm/cam/src/chemistry/pp_waccm_sc_mam3/mo_lin_matrix.F90
A models/atm/cam/src/chemistry/pp_waccm_sc_mam3/m_rxt_id.F90
A models/atm/cam/src/chemistry/pp_waccm_sc_mam3/mo_phtadj.F90
A models/atm/cam/src/chemistry/pp_waccm_sc_mam3/mo_nln_matrix.F90
A models/atm/cam/src/chemistry/pp_waccm_sc_mam3/mo_lu_factor.F90
A models/atm/cam/src/chemistry/pp_waccm_sc_mam3/mo_sim_dat.F90
A models/atm/cam/src/chemistry/pp_waccm_sc_mam3/mo_indprd.F90
A models/atm/cam/src/chemistry/pp_waccm_sc_mam3/m_spc_id.F90
A models/atm/cam/src/chemistry/pp_waccm_sc_mam3/chem_mech.doc
A models/atm/cam/src/chemistry/pp_waccm_sc_mam3/chem_mech.in
A models/atm/cam/src/chemistry/pp_waccm_sc_mam3
- new chem package (this mechanism lacks OCS which is important for the
non-volcanic stratospheric sulfate)
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_lu_solve.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_setrxt.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_adjrxt.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_rxt_rates_conv.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_imp_sol.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_sulfur/chem_mods.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_prod_loss.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_lin_matrix.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_sulfur/m_rxt_id.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_phtadj.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_nln_matrix.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_lu_factor.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_sim_dat.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_sulfur/mo_indprd.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_sulfur/m_spc_id.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_sulfur/chem_mech.doc
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_sulfur/chem_mech.in
A models/atm/cam/src/chemistry/pp_waccm_tsmlt_sulfur
- new chem package
A models/atm/cam/src/chemistry/pp_waccm_sc/mo_lu_solve.F90
A models/atm/cam/src/chemistry/pp_waccm_sc/mo_setrxt.F90
A models/atm/cam/src/chemistry/pp_waccm_sc/mo_adjrxt.F90
A models/atm/cam/src/chemistry/pp_waccm_sc/mo_rxt_rates_conv.F90
A models/atm/cam/src/chemistry/pp_waccm_sc/mo_imp_sol.F90
A models/atm/cam/src/chemistry/pp_waccm_sc/chem_mods.F90
A models/atm/cam/src/chemistry/pp_waccm_sc/mo_prod_loss.F90
A models/atm/cam/src/chemistry/pp_waccm_sc/mo_lin_matrix.F90
A models/atm/cam/src/chemistry/pp_waccm_sc/m_rxt_id.F90
A models/atm/cam/src/chemistry/pp_waccm_sc/mo_phtadj.F90
A models/atm/cam/src/chemistry/pp_waccm_sc/mo_nln_matrix.F90
A models/atm/cam/src/chemistry/pp_waccm_sc/mo_lu_factor.F90
A models/atm/cam/src/chemistry/pp_waccm_sc/mo_sim_dat.F90
A models/atm/cam/src/chemistry/pp_waccm_sc/mo_indprd.F90
A models/atm/cam/src/chemistry/pp_waccm_sc/m_spc_id.F90
A models/atm/cam/src/chemistry/pp_waccm_sc/chem_mech.doc
A models/atm/cam/src/chemistry/pp_waccm_sc/chem_mech.in
A models/atm/cam/src/chemistry/pp_waccm_sc
- renamed 'waccm_ghg' to 'waccm_sc'
List all existing files that have been modified, and describe the changes:
M models/atm/cam/src/control/cam_history.F90
- add F10.7, Kp and Ap data to cam history if they are input into the model
regardless of which chem package is used
M models/atm/cam/src/physics/cam/carma_flags_mod.F90
- added 'carm_rad_feedback' and 'carm_do_partialinit'
- changed 'carma_to_hetchem' to 'carma_hetchem_feedback'
M models/atm/cam/src/physics/cam/tropopause.F90
- added check for troplev > 0
M models/atm/cam/src/physics/cam/micro_mg_cam.F90
- a workaround for pgi compiler bug on goldbach
M models/atm/cam/src/physics/cam/modal_aer_opt.F90
M models/atm/cam/src/physics/cam/aer_rad_props.F90
- added diagnostics for stratospheric aerosols
M models/atm/cam/src/physics/cam/phys_control.F90
- prog_modal_aero set by checking for '_mam' in chem package name
M models/atm/cam/src/physics/cam/cam_diagnostics.F90
- removed the chem package name check for tidal diagnostics
M models/atm/cam/src/physics/cam/nucleate_ice.F90
- initialize variables 'deles' and 'esi' to zero
M models/atm/cam/src/physics/waccmx/ionosphere.F90
- 'get_solar_parms' changed to 'solar_parms_get'
M models/atm/cam/src/physics/carma/cam/carma_intr.F90
M models/atm/cam/src/physics/carma/models/sulfate/carma_model_mod.F90
- carma updates for rad feedback and partial init
M models/atm/cam/src/chemistry/utils/prescribed_strataero.F90
- correction to comment
M models/atm/cam/src/chemistry/utils/tracer_data.F90
- increased size of buffer used to read in units attribute
M models/atm/cam/src/chemistry/utils/solar_data.F90
- masterproc read namelist
- added check to prevent users from setting solar_data_ymd when
solar_data_type is 'SERIAL'
M models/atm/cam/src/chemistry/utils/mo_solar_parms.F90
- added readnl subroutine
- added "solar_parm_on" indicator
M models/atm/cam/src/chemistry/utils/modal_aero_wateruptake.F90
- mods for stratospheric sulfate
M models/atm/cam/src/chemistry/modal_aero/aero_model.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_rename.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_gasaerexch.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_initialize_data.F90
- mods for stratospheric sulfate and accumulation to coarse mode exchanges
M models/atm/cam/src/chemistry/modal_aero/modal_aero_newnuc.F90
- removed MODAL_AERO ifdef
M models/atm/cam/src/chemistry/bulk_aero/aero_model.F90
- interface change in gasaerexch routine
- added stub strat_surfarea subroutine
M models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
- call aero_modal_strat_surfarea
M models/atm/cam/src/chemistry/mozart/photo_bkgrnd.F90
- 'get_solar_parms' changed to 'solar_parms_get'
M models/atm/cam/src/chemistry/mozart/spedata.F90
- correction in log message
M models/atm/cam/src/chemistry/mozart/upper_bc.F90
- use pressure of top of model to determine of upper boundary conditions are applied
M models/atm/cam/src/chemistry/mozart/mo_chemini.F90
- removed is_waccm checks
M models/atm/cam/src/chemistry/mozart/mo_extfrc.F90
- new capability added to allow individual species to have forcings from
multiple files
- global attribute 'input_method' in the input file which overrides the
ext_frc_type namelist setting on a file-by-file basis
M models/atm/cam/src/chemistry/mozart/euvac.F90
- added euvac_on indicator
M models/atm/cam/src/chemistry/mozart/chemistry.F90
- remove chem_is check used to set constituent mixtype -- now all
chemical constituents are registered with mixtype='dry'.
If do_molec_diff is true cnst_type is set to wet in chem_init.
- water vapor has fixed upper boundary if top of model pressure is
comparable to waccm's top
- removed chem_is checks and WACCM ifdefs -- waccm parameterizations
know if they are active
M models/atm/cam/src/chemistry/mozart/mo_photo.F90
- removed waccm checks -- use top of model pressure
M models/atm/cam/src/chemistry/mozart/mag_parms.F90
M models/atm/cam/src/chemistry/mozart/efield.F90
- 'get_solar_parms' changed to 'solar_parms_get'
M models/atm/cam/src/chemistry/mozart/mo_usrrxt.F90
- 'carma_do_hetchem' changed to carm_hetchem_feedback'
M models/atm/cam/src/chemistry/mozart/mo_jeuv.F90
- added check for active jeuv reactions
M models/atm/cam/src/chemistry/mozart/mo_srf_emissions.F90
- new capability added to allow individual species to have emissions from
multiple files
- global attribute 'input_method' in the input file which overrides the
ext_frc_type namelist setting on a file-by-file basis
M models/atm/cam/src/dynamics/fv/trac2d.F90
M models/atm/cam/src/dynamics/fv/cd_core.F90
- replaced WACCM cppdefs with top of model pressure checks
M models/atm/cam/bld/configure
- updated the chem option description:
. waccm_ghg renamed to waccm_sc
. new chem pckgs -- waccm_sc_mam3, waccm_tsmlt_mam3, waccm_tsmlt_sulfur
- removed the numerous chem_pkg conditionals which set chem_nadv (hard wired) --
now let ChemPreprocess perl module determine the number of advected
chemical tracers
- removed the WACCM cpp defines
M models/atm/cam/bld/config_files/definition.xml
- updated the chem option definition:
. waccm_ghg renamed to waccm_sc
. new chem pckgs -- waccm_sc_mam3, waccm_tsmlt_mam3, waccm_tsmlt_sulfur
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- new default IC files for
. 10x15 waccm5
. 10x15 waccmx
. waccm/carma/sulfur
- new mode definition files for prognostic stratospheric aerosols
- set default inputs for waccm_sc_mam3
- update default long-wavelength photolysis cross section input file
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- changes to carma options:
. added carma_do_partialinit
. added carma_rad_feedback
. carma_do_hetchem renamed to carma_hetchem_feedback
- changed valid values for cam_chempkg
- modal aersol options
. modal_accum_coarse_exch -- allows accumulation to coarse mode exchange to take place
-- needed for prognostic stratospheric aerosols
. modal_strat_sulfate -- turns on prognostic stratospheric sulfate
- several places where 'waccm_ghg' changed to 'waccm_sc'
M models/atm/cam/bld/build-namelist
- 'waccm_ghg' to 'waccm_sc' changes
- set appropriate mode definition files if 'modal_accum_coarse_exch' is set to TRUE
- set rad_climate appropriately when 'carma_rad_feedback' is set to TRUE
- determine if prescribed stratospheric aerosols are needed --
when there are no prognostic stratospheric aerosols
- set namelist defaults for waccm_sc_mam3 and waccm_tsmlt_mam3
- don't override cam_chempkg setting for waccm
- in check_input_files look for multiplier coefficient proceeding emissions filepath
M models/atm/cam/bld/perl5lib/Build/ChemPreprocess.pm
- separated out the code which determines the number of advected chemical tracers
M models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_setrxt.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_rxt_rates_conv.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/chem_mods.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_phtadj.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/m_spc_id.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/chem_mech.doc
M models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/chem_mech.in
- chemistry updates
M models/atm/cam/bld/namelist_files/use_cases/sd_2007-2011_htap_trop_strat_soa.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp26.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1955-2005_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1974-2005_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_carma_bc_2013_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1850_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccmx_1996_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/sd_waccm_sulfur.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_sc_2005-2100_cam4_rcp26.xml
M models/atm/cam/bld/namelist_files/use_cases/waccmx_2000_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/2000_ccmi_refc1_waccm_tsmlt.xml
M models/atm/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_rcp85_waccm_ma.xml
M models/atm/cam/bld/namelist_files/use_cases/sd_waccm_tsmlt_geos5.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_refb1.xml
M models/atm/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_waccm_ma.xml
M models/atm/cam/bld/namelist_files/use_cases/1950-2010_ccmi_refc1_waccm_tsmlt.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1850-2005_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2000_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2000_cam5.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_sc_1850_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp45.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp85.xml
M models/atm/cam/bld/namelist_files/use_cases/2007-2011_htap_trop_strat_soa.xml
M models/atm/cam/bld/namelist_files/use_cases/1950-2010_ccmi_refc1_trop_strat_soa.xml
M models/atm/cam/bld/namelist_files/use_cases/sd_1975-2010_ccmi_refc1_waccm_ma.xml
M models/atm/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_trop_strat_soa.xml
M models/atm/cam/bld/namelist_files/use_cases/2004-2100_ccmi_refc2_rcp45_trop_strat_soa.xml
M models/atm/cam/bld/namelist_files/use_cases/sd_waccm_geos5.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_sc_2005-2100_cam4_rcp45.xml
M models/atm/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_waccm_tsmlt.xml
M models/atm/cam/bld/namelist_files/use_cases/sd_waccm_tslt_geos5.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_sc_2005-2100_cam4_rcp85.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_sc_1955-2005_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/sd_1975-2010_ccmi_refc1_trop_strat_soa.xml
M models/atm/cam/bld/namelist_files/use_cases/1950-2010_ccmi_refc1_waccm_ma.xml
M models/atm/cam/bld/namelist_files/use_cases/sd_1975-2010_ccmi_refc1_waccm_tsmlt.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_sc_2000_cam4.xml
- removed namelist settings that were set to default values or should not be set here
such as phys_loadbalance and dtime
M models/atm/cam/bld/namelist_files/use_cases/waccm_1850_cam5.xml
- removed namelist settings that were set to default values or should not be set here
- update LBC file
- IC file for 4x5
M models/atm/cam/test/system/tests_chem_hybrid
M models/atm/cam/test/system/tests_chem_mpi
- new chemistry tests added
M models/atm/cam/test/system/tests_pretag_yellowstone
- added waccm5_tsmlt and waccm5_sc tests
M models/atm/cam/test/system/tests_pretag_goldbach_nag
- added 10-degree waccmx and waccm5 tests
M models/atm/cam/test/system/nl_files/outfrq1m
- set frequency of all history streams to 1-month
M models/atm/cam/test/system/input_tests_master
- changed 'wghg' to 'wsc'
- new tests added
- corrected specification of sd-waccm test 501
- for CCSM tests 993 use 'ne30_ne30' resolution rather than 'ne30_g16'
-- this is consistent with the resolution specified in eq993
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
- updated carma model
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
016 bl326 TBL.sh f1.9c4wscdm outfrq3s_2005+waccm_sc_2005-2100_cam4_rcp45 9s ...................FAIL! rc= 7 at Wed Dec 31 17:43:05 MST 2014
- expected failure -- fixed upper bound of water vapor for waccm_ghg chem package
042 bl374 TBL.sh f1.9c4fastdh outfrq3s+1850-2005_cam4_super_fast_llnl 9s ......................FAIL! rc= 7 at Wed Dec 31 18:43:44 MST 2014
- expected failure -- changed constituent mixing type from 'wet' to 'dry' for super_fast_llnl chem package
056 bl474 TBL.sh f1.9c5wtmam3h outfrq3s+1850_waccm_tsmlt_cam5 9s ..............................FAIL! rc= 7 at Wed Dec 31 19:36:57 MST 2014
- expected failure -- waccm_tsmlt_mam3 is a new chem package -- no baseline to compare against
059 bl482 TBL.sh f1.9c5wscmam3dh outfrq3s_2005+waccm_sc_1955-2005_cam5 9s .....................FAIL! rc= 7 at Wed Dec 31 19:59:39 MST 2014
- expected failure -- waccm_sc_mam3 is a new chem package -- no baseline to compare against
goldbach/nag:
031 bl314 TBL.sh f10c4wscdm outfrq3s 9s .......................................................FAIL! rc= 7 at Wed Dec 31 16:11:05 MST 2014
- expected failure -- fixed upper bound of water vapor for waccm_ghg chem package
058 bl424 TBL.sh f10c4wmxdm outfrq3s 9s .......................................................FAIL! rc= 7 at Wed Dec 31 18:16:11 MST 2014
- expected failure -- new waccmx test -- no baseline to compare against
061 bl471 TBL.sh f10c5wmam3dm outfrq3s_modalstrat 9s ..........................................FAIL! rc= 7 at Wed Dec 31 18:32:04 MST 2014
- expected failure -- chemistry mechanism changes
goldbach/pgi:
036 bl314 TBL.sh f10c4wscdm outfrq3s 9s .......................................................FAIL! rc= 7 at Wed Dec 31 16:12:31 MST 2014
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_59
Originator(s): santos (contributions from others, see below)
Date: 2014/12/24
One-line Summary: New gravity wave and vertical diffusion code, unit tests.
Purpose of changes:
- Fix bug regarding molecular diffusion of heat in WACCM.
- Add inertial gravity waves triggered by frontogenesis.
- Adapted from the WACCM code used for CCMI, designed by X.H. Xue, with
modifications from liuh and aksmith.
- Add new finite volume diffusion solver code to simplify processes that
perform vertical diffusion in the column physics. The method is largely
identical to old code, but the interface is intended to provide more
flexibility and reduce the likelihood of errors when making changes in
the middle-layer science code, e.g. when changing boundary conditions.
- Add first unit tests for CAM, for new diffusion code.
- Allow different gravity wave sources to produce different spectra.
- Add gravity waves sourced from shallow convection (disabled by default).
- Adapted from code from Tzu-Ling Lai (aka Leslie).
- Add new "consistent" topography files for FV 1 and 2 degree.
- Turn on diffusion for orographic gravity waves.
- Gravity wave efficiencies are now applied to the generated stress profile
(tau), and not just the wind tendencies.
- Fix a number of bugs and tuning issues in gravity waves.
- Fix bug where orographic waves were erroneously adding in tendencies
with contributions from other gravity wave types (fixing this yields a
somewhat hotter mesosphere in WACCM).
- Remove problematic limit on effective diffusivity due to gravity waves.
- Increase minimum value of Brunt-Vaisalla frequency to address numerical
inaccuracies.
- Fix issue where heating from breaking waves used the wrong wave speed
("c" instead of "u-c").
- Replace "alpha0" (Newtonian cooling profile) with a version considered
to be better-founded, and use this for both CAM and WACCM.
- Scale sgh up using landfrac before calculating source sizes, then scale
the efficiency down.
- Several minor bugs.
- Resolve minor discrepancies between CAM and WACCM gravity wave treatment,
especially limiters that are redundant.
- Add USE_CONTIGUOUS flag to the CAM Makefile, allowing use of the
"contiguous" attribute to assist optimization on compilers that support
it.
Bugs fixed (include bugzilla ID):
- #1799: gw_drag: Wrong expression for converting kinetic energy to
thermal in wave dissipation.
- #1800: gw_drag: Orographic waves produce spurious cooling upon
dissipation.
- #1801: gw_drag: gw_diffusion functions are called with kbotbg instead of
kbot
- #1812: gw_drag: Beres scheme uses cshift on a table when it only needs a
normal shift.
- #1878: Molecular diffusion upper boundary condition has an extra term in
WACCM.
- #1980: WACCM: output of "ap" to history file actually overwrites "kp"
instead
- #2016: SC-WACCM: uninitialized flags cause a crash on XLF with DEBUG.
- #2069: Molecular diffusion of heat disabled in WACCM.
- #2103: gw_drag: Wrong value of tau used in momentum fixer.
Describe any changes made to build system:
Describe any changes made to the namelist:
- Add shallow convection gravity wave source configuration.
- Pre-existing variables now have the suffix "_dp", except for
gw_drag_file, which may provide a file usable for a combined
shallow and deep convection scheme in the future.
- New variables now have the suffix "_sh".
List any changes to the defaults for the boundary datasets:
- CAM-FV 1 and 2 degree configurations now use topography files generated
with the same consistent method used to create the CAM-FV topography
files.
Describe any substantial timing or memory changes:
Code reviewed by: eaton, CAM code reviewers
List all subroutines eliminated:
List all subroutines added and what they do:
M models/atm/cam/test/unit/CMakeLists.txt
M models/atm/cam/src/physics/cam/CMakeLists.txt
A models/atm/cam/src/utils/CMakeLists.txt
A models/atm/cam/test/unit/coords_1d
A models/atm/cam/test/unit/coords_1d/CMakeLists.txt
A models/atm/cam/test/unit/coords_1d/test_coords_1d.pf
A models/atm/cam/test/unit/linear_1d_operators
A models/atm/cam/test/unit/linear_1d_operators/CMakeLists.txt
A models/atm/cam/test/unit/linear_1d_operators/test_diagonal.pf
A models/atm/cam/test/unit/linear_1d_operators/test_derivatives.pf
A models/atm/cam/test/unit/linear_1d_operators/test_arithmetic.pf
A models/atm/cam/test/unit/vdiff_lu_solver
A models/atm/cam/test/unit/vdiff_lu_solver/CMakeLists.txt
A models/atm/cam/test/unit/vdiff_lu_solver/test_fd_solver.pf
A models/atm/cam/test/unit/vdiff_lu_solver/test_fv_solver.pf
- Add tests (directory names correspond to the names of the modules
they test).
A models/atm/cam/src/physics/cam/coords_1d.F90
- New vertical coordinate type (introduced to cache derived values
of the pressure coordinates for performance reasons).
A models/atm/cam/src/physics/cam/linear_1d_operators.F90
- New operator type for linear operations on 1D grids that can be
represented by tridiagonal matrices (plus boundary conditions).
List all existing files that have been modified, and describe the changes:
M SVN_EXTERNAL_DIRECTORIES
- Add unit_testing external.
M models/atm/cam/bld/Makefile.in
- Add USE_CONTIGUOUS CPP macro.
M models/atm/cam/bld/build-namelist
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- Add second gravity wave spectrum for inertial waves.
- Add shallow convection variables to namelist.
- Add inertial gravity wave variables to namelist.
- gw_polar_taper now controls the polar taper for non-inertial
waves from frontogenesis (taper is on by default for FV only).
- The "60 level model" Beres efficiency now applies to all
non-WACCM cases.
- Change to new topography files for f19 and f09 resolutions.
M models/atm/cam/bld/namelist_files/use_cases/1950-2010_ccmi_refc1_trop_strat_soa.xml
M models/atm/cam/bld/namelist_files/use_cases/1950-2010_ccmi_refc1_waccm_ma.xml
M models/atm/cam/bld/namelist_files/use_cases/1950-2010_ccmi_refc1_waccm_tsmlt.xml
M models/atm/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_rcp85_waccm_ma.xml
M models/atm/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_trop_strat_soa.xml
M models/atm/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_waccm_ma.xml
M models/atm/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_waccm_tsmlt.xml
M models/atm/cam/bld/namelist_files/use_cases/1979-2010_sd_trop_strat_soa.xml
M models/atm/cam/bld/namelist_files/use_cases/2000_ccmi_refc1_waccm_tsmlt.xml
M models/atm/cam/bld/namelist_files/use_cases/2004-2100_ccmi_refc2_rcp45_trop_strat_soa.xml
M models/atm/cam/bld/namelist_files/use_cases/2007-2011_htap_trop_strat_soa.xml
M models/atm/cam/bld/namelist_files/use_cases/sd_1975-2010_ccmi_refc1_trop_strat_soa.xml
M models/atm/cam/bld/namelist_files/use_cases/sd_1975-2010_ccmi_refc1_waccm_ma.xml
M models/atm/cam/bld/namelist_files/use_cases/sd_1975-2010_ccmi_refc1_waccm_tsmlt.xml
M models/atm/cam/bld/namelist_files/use_cases/sd_2007-2011_htap_trop_strat_soa.xml
M models/atm/cam/bld/namelist_files/use_cases/sd_waccm_geos5.xml
M models/atm/cam/bld/namelist_files/use_cases/sd_waccm_sulfur.xml
M models/atm/cam/bld/namelist_files/use_cases/sd_waccm_tslt_geos5.xml
M models/atm/cam/bld/namelist_files/use_cases/sd_waccm_tsmlt_geos5.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1850-2005_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1850_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1850_cam5.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1955-2005_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1974-2005_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2000_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2000_cam5.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp26.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp45.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp85.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_carma_bc_2013_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_refb1.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_sc_1850_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_sc_1955-2005_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_sc_2000_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_sc_2005-2100_cam4_rcp26.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_sc_2005-2100_cam4_rcp45.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_sc_2005-2100_cam4_rcp85.xml
M models/atm/cam/bld/namelist_files/use_cases/waccmx_1996_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccmx_2000_cam4.xml
- Rename "EKGWSPEC" to "EKGW", since it now includes contributions
from orographic waves (which do not cover a spectrum).
M models/atm/cam/src/dynamics/fv/dp_coupling.F90
M models/atm/cam/src/dynamics/fv/dyn_comp.F90
M models/atm/cam/src/dynamics/se/dp_coupling.F90
M models/atm/cam/src/dynamics/se/dyn_comp.F90
M models/atm/cam/src/dynamics/se/stepon.F90
- Frontogenesis now has to be calculated if either inertial or
non-inertial gravity waves are triggered by it.
M models/atm/cam/src/physics/cam/boundarydata.F90
- Initialize some flags that were causing crashes on platforms
where module data is not initialized to 0 (or in this case,
to ".false.").
M models/atm/cam/src/physics/cam/convect_deep.F90
M models/atm/cam/src/physics/cam/convect_shallow.F90
M models/atm/cam/src/physics/cam/phys_control.F90
- Changes to provide separate data/flags for deep and shallow
convective gravity waves.
- Add use_gw_front_igw flag.
M models/atm/cam/src/physics/cam/diffusion_solver.F90
M models/atm/cam/src/physics/cam/eddy_diff.F90
M models/atm/cam/src/physics/cam/molec_diff.F90
M models/atm/cam/src/physics/cam/vertical_diffusion.F90
- Fix bug where molecular diffusion of heat was disabled in WACCM
(#2069).
- Incorporate new diffusion code. The new code handles some details
such as pressure level thicknesses, which therefore do not need
to be accounted for in higher-level code (e.g. we no longer pass
tmpi2).
- Fully calculate WACCM-X diffusion coefficient terms down to the
lowest internal interface in the molecular diffusion region, not
the second-lowest, as was done before.
M models/atm/cam/src/physics/cam/gw_common.F90
M models/atm/cam/src/physics/cam/gw_drag.F90
M models/atm/cam/src/physics/cam/physpkg.F90
- Allow different gravity wave source parameterizations to emit
waves in different spectra, defined by a new type called
GWSpectrum.
- Add new inertial gravity waves triggered by frontogenesis.
- Add new shallow convection source, which has its own efficiency
and input file.
- The "tndmax" limiter (for numerical stability) is now used on the
total tendency for each grid point, rather than the contribution
from each phase speed, as was done before.
- Add file and line numbers to error messages.
- Use PIO to read in input files, not a direct call to netCDF.
- Read in Beres lookup table dimensions, since we can no longer
make assumptions about some of their values or sizes.
- Remove "orographic_only" and "do_spectral_waves" flags, and stop
using "do_molec_diff" except in the CAM interface layer, since
these represent ultimately unnecessary differences between WACCM
and CAM gravity waves.
- Refactor momentum/energy conservation code into four very
short routines, and report energy excess from the orographic
and shallow convection waves as a heating rate to
check_energy_chng.
- Fix bug where orographic waves added extra wind tendencies.
- Use a higher minimum value for B-V frequency.
- Fix bug where a wrong speed was used to calculate kinetic energy
differences.
- Use per-column gravity wave efficiencies, so that CAM can tweak
these values. This allows local scaling factors, such as the
polar taper, to be implemented by simply adjusting the
efficiency.
- Gravity wave efficiencies are now applied to tau as well as to
the wind tendencies. This makes the limiters on tau less
restrictive, and also fixes an issue where the momentum fixer
failed to correctly account for the efficiency.
- Remove CAM-specific code affecting the diffusivity that damps
stresses, since this is now thought to be an poor method of
accounting for self-damping.
- Remove several limiters on various quantities, as most of these
are either numerically unnecessary or redundant with other
checks.
- Improve handling of landfrac by using it to scale sgh as well as
as the efficiency.
- Enable diffusion from orographic gravity waves.
- Output "EKGWSPEC" is now renamed to "EKGW" and includes
orographic diffusion.
- Output temperature tendency for each gravity wave source.
- Remove redundant code which was left in during a previous
refactoring purely in order to preserve answers.
- Fix bug where all diffusion was limited to occur above the bottom
level for waves from frontogenesis, instead of a source-specific
bottom level.
- Index arrays defined on level interfaces from 1, not 0.
M models/atm/cam/src/physics/cam/gw_convect.F90
- Modifications to make the code more generic, so that it can be
used for sourcing waves from shallow convection.
- Add BeresSourceSpec type, which holds data read in from the input
file, and other differences between deep and shallow convective
sources.
- Remove assumption that heating depth dimension of the input file
contains only integers.
- Fix bug where storm speed was accounted for with a circular shift
(cshift) instead of a regular shift (eoshift).
- Removed unnecessary casts of storm cell speed to an integer and
back to double precision, which lost precision during table
lookup.
- Change handling of pole points to reduce the chance of
mishandling due to inexact floating point arithmetic.
M models/atm/cam/src/physics/cam/gw_diffusion.F90
- Modifications to use new diffusion code.
- Remove limiter on egwdffi, which turned out to be too low and
ultimately unnecessary.
M models/atm/cam/src/physics/cam/gw_front.F90
- Instead of using the trapezoidal method to integrate a gaussian,
use erfc.
- Use the type "CMSourceSpec" to hold information that was used as
module data, to improve flexibility enough to handle inertial and
non-inertial waves with the same routine.
M models/atm/cam/src/physics/cam/gw_oro.F90
- Changes for compatibility with gw_common, especially using the
new spectra.
M models/atm/cam/src/physics/cam/gw_utils.F90
- Convert "elemental" subroutines to vector subroutines.
M models/atm/cam/src/physics/cam/vdiff_lu_solver.F90
- Now uses linear_1d_operators to construct operators before
decomposing them.
- The definition and application of boundary conditions is more
generic.
M models/atm/cam/src/utils/cam_pio_utils.F90
- Remove dead code and make error messages more specific.
M models/atm/cam/test/system/input_tests_master
A models/atm/cam/test/system/nl_files/outfrq3s_gw_igw
A models/atm/cam/test/system/nl_files/outfrq3s_gw_sh
- Add test for inertial gravity waves.
- Add test for gravity waves from shallow convection.
M models/atm/cam/test/system/test_driver.sh
- Add ability to override CAM_TASKS and CAM_RESTART_TASKS using the
environment (this is necessary to run tests with very low
resolution or high overall cost, when the machine defaults will
not work).
Finally, the following files were moved/changed as part of a general move
of utility code to src/utils rather than src/control or src/physics/cam.
The changed (rather than added/deleted) files were mostly changed to add
use statements for subroutines that were placed inside modules as well as
being moved to utils.
D models/atm/cam/src/control/mpishorthand.F
D models/atm/cam/src/control/physconst.F90
D models/atm/cam/src/control/intp_util.F90
D models/atm/cam/src/control/bnddyi.F90
D models/atm/cam/src/control/interpolate_data.F90
D models/atm/cam/src/control/wrap_nf.F90
D models/atm/cam/src/control/datetime.F90
D models/atm/cam/src/control/srchutil.F90
D models/atm/cam/src/control/units.F90
D models/atm/cam/src/control/gauaw_mod.F90
D models/atm/cam/src/control/ioFileMod.F90
D models/atm/cam/src/control/vrtmap.F90
D models/atm/cam/src/control/infnan.F90
D models/atm/cam/src/control/wrap_mpi.F90
D models/atm/cam/src/control/error_messages.F90
M models/atm/cam/src/control/cam_history.F90
A models/atm/cam/src/utils/mpishorthand.F
A models/atm/cam/src/utils/physconst.F90
A models/atm/cam/src/utils/intp_util.F90
A models/atm/cam/src/utils/bnddyi.F90
A models/atm/cam/src/utils/coords_1d.F90
A models/atm/cam/src/utils/interpolate_data.F90
A models/atm/cam/src/utils/wrap_nf.F90
A models/atm/cam/src/utils/datetime.F90
A models/atm/cam/src/utils/linear_1d_operators.F90
A models/atm/cam/src/utils/srchutil.F90
M models/atm/cam/src/utils/cam_dom/sst_data.F90
A models/atm/cam/src/utils/units.F90
A models/atm/cam/src/utils/gauaw_mod.F90
A models/atm/cam/src/utils/ioFileMod.F90
A models/atm/cam/src/utils/infnan.F90
A models/atm/cam/src/utils/vrtmap.F90
A models/atm/cam/src/utils/wrap_mpi.F90
A models/atm/cam/src/utils/error_messages.F90
D models/atm/cam/src/physics/cam/linear_1d_operators.F90
D models/atm/cam/src/physics/cam/coords_1d.F90
M models/atm/cam/src/physics/cam/co2_data_flux.F90
M models/atm/cam/src/physics/waccm/qbo.F90
M models/atm/cam/src/chemistry/utils/mo_flbc.F90
M models/atm/cam/src/chemistry/utils/tracer_data.F90
M models/atm/cam/src/dynamics/sld/scanslt.F90
M models/atm/cam/src/dynamics/eul/scanslt.F90
M models/atm/cam/src/dynamics/se/native_mapping.F90
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
004 bl132 TBL.sh e48c4dh ghgrmp+1850_cam4 9s ..................................................FAIL! rc= 7 at Tue Dec 23 17:38:45 MST 2014
006 bl133 TBL.sh e48c4paqdm aqpgro+aquaplanet_cam4 3s .........................................FAIL! rc= 7 at Tue Dec 23 17:39:09 MST 2014
009 bl203 TBL.sh fsd1.9c4wtsmlth outfrq3s_sd+sd_1975-2010_ccmi_refc1_waccm_tsmlt 9s ...........FAIL! rc= 7 at Tue Dec 23 17:59:52 MST 2014
012 bl322 TBL.sh f1.9c5nonedh outfrq3s_lb0 9s .................................................FAIL! rc= 7 at Tue Dec 23 18:06:47 MST 2014
016 bl326 TBL.sh f1.9c4wghgdm outfrq3s_2005+waccm_sc_2005-2100_cam4_rcp45 9s ..................FAIL! rc= 7 at Tue Dec 23 18:20:14 MST 2014
020 bl331 TBL.sh f4c4dh co2rmp+1850_cam4 9s ...................................................FAIL! rc= 7 at Tue Dec 23 18:23:21 MST 2014
024 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...................................................FAIL! rc= 7 at Tue Dec 23 18:29:43 MST 2014
026 bl333 TBL.sh f4c4paqdh aqpgro+aquaplanet_cam4 3s ..........................................FAIL! rc= 7 at Tue Dec 23 18:29:59 MST 2014
029 bl336 TBL.sh f4c4wmdh outfrq3s+waccm_1850_cam4 9s .........................................FAIL! rc= 7 at Tue Dec 23 18:41:51 MST 2014
032 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Tue Dec 23 18:52:12 MST 2014
035 bl373 TBL.sh f1.9c4mozdh outfrq3s+2000_cam4_trop_chem 9s ..................................FAIL! rc= 7 at Tue Dec 23 19:05:31 MST 2014
037 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...............................................FAIL! rc= 7 at Tue Dec 23 19:08:01 MST 2014
039 bl376 TBL.sh f1.9c4mozdh+testmech outfrq3s 9s .............................................FAIL! rc= 7 at Tue Dec 23 19:14:03 MST 2014
042 bl374 TBL.sh f1.9c4fastdh outfrq3s+1850-2005_cam4_super_fast_llnl 9s ......................FAIL! rc= 7 at Tue Dec 23 19:20:53 MST 2014
045 bl390 TBL.sh f1.9c4wmxh outfrq3s+waccmx_2000_cam4 9s ......................................FAIL! rc= 7 at Tue Dec 23 19:30:06 MST 2014
048 bl421 TBL.sh f1.9c5mam4dh outfrq3s 9s .....................................................FAIL! rc= 7 at Tue Dec 23 19:40:06 MST 2014
050 bl430 TBL.sh f1.9c5mzmam3dh outfrq3s_NEUwetdep 9s .........................................FAIL! rc= 7 at Tue Dec 23 19:48:54 MST 2014
053 bl440 TBL.sh f1.9c4soadh outfrq3s+soa_chem_megan_emis 9s ..................................FAIL! rc= 7 at Tue Dec 23 20:05:34 MST 2014
057 bl387 TBL.sh f1.9c4dm outfrq3s+1850-2005_cam4 9s ..........................................FAIL! rc= 7 at Tue Dec 23 20:11:54 MST 2014
060 bl388 TBL.sh f1.9c4m outfrq24h+1850_cam4 2d ...............................................FAIL! rc= 7 at Tue Dec 23 20:13:32 MST 2014
063 bl379 TBL.sh f1.9c4wmh outfrq24h+waccm_2005-2100_cam4_rcp45 2d ............................FAIL! rc= 7 at Tue Dec 23 20:30:50 MST 2014
066 bl731 TBL.sh h16c3aqdh aqua+aquaplanet_cam3 9s ............................................FAIL! rc= 7 at Tue Dec 23 20:33:41 MST 2014
069 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ........................................FAIL! rc= 7 at Tue Dec 23 20:44:32 MST 2014
074 bl740 TBL.sh h30c5h outfrq3s_lb0 9s .......................................................FAIL! rc= 7 at Tue Dec 23 20:48:01 MST 2014
078 bl751 TBL.sh h30c4aqdm outfrq3s+aquaplanet_cam4 9s ........................................FAIL! rc= 7 at Tue Dec 23 21:02:54 MST 2014
goldbach/nag:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s ...........................................................FAIL! rc= 7 at Tue Dec 23 19:40:47 MST 2014
010 bl112 TBL.sh e8c5dm ghgrmp 9s .............................................................FAIL! rc= 7 at Tue Dec 23 19:40:49 MST 2014
014 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Tue Dec 23 19:40:52 MST 2014
018 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Tue Dec 23 19:40:54 MST 2014
026 bl312 TBL.sh f2.5c5dm ghgrmp 9s ...........................................................FAIL! rc= 7 at Tue Dec 23 19:41:04 MST 2014
028 bl313 TBL.sh f10c3aqdm outfrq3s+aquaplanet_cam3 3s ........................................FAIL! rc= 7 at Tue Dec 23 19:41:05 MST 2014
031 bl314 TBL.sh f10c4wghgdm outfrq3s 9s ......................................................FAIL! rc= 7 at Tue Dec 23 19:41:07 MST 2014
033 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .........................................FAIL! rc= 7 at Tue Dec 23 19:41:08 MST 2014
036 bl318 TBL.sh f10c4cdm sat_hist 9s .........................................................FAIL! rc= 7 at Tue Dec 23 19:41:10 MST 2014
039 bl324 TBL.sh f4c4rrtmgdm outfrq3s 9s ......................................................FAIL! rc= 7 at Tue Dec 23 19:41:13 MST 2014
042 bl325 TBL.sh f10c5t5mdmsc outfrq3s_subcol 9s ..............................................FAIL! rc= 7 at Tue Dec 23 19:41:14 MST 2014
049 bl368 TBL.sh f1.9c5mg2dm outfrq3s_macmic2_classnuc 9s .....................................FAIL! rc= 7 at Tue Dec 23 20:11:59 MST 2014
052 bl734 TBL.sh h16c5naqdm outfrq3s_bam+aquaplanet_cam5 9s ...................................FAIL! rc= 7 at Tue Dec 23 20:12:12 MST 2014
055 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Tue Dec 23 20:12:17 MST 2014
058 bl805 TBL.sh f4c5carmtt1dm outfrq3s_carma 9s ..............................................FAIL! rc= 7 at Tue Dec 23 20:12:51 MST 2014
goldbach/pgi or jaguar/pgi:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s ...........................................................FAIL! rc= 7 at Tue Dec 23 20:17:40 MST 2014
010 bl112 TBL.sh e8c5dm ghgrmp 9s .............................................................FAIL! rc= 7 at Tue Dec 23 20:17:42 MST 2014
012 bl113 TBL.sh e8c5paqdm aqpgro+aquaplanet_cam5 3s ..........................................FAIL! rc= 7 at Tue Dec 23 20:17:43 MST 2014
016 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Tue Dec 23 20:17:46 MST 2014
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Tue Dec 23 20:17:48 MST 2014
031 bl312 TBL.sh f2.5c5dm ghgrmp 9s ...........................................................FAIL! rc= 7 at Tue Dec 23 20:17:58 MST 2014
033 bl313 TBL.sh f10c3aqdm outfrq3s+aquaplanet_cam3 3s ........................................FAIL! rc= 7 at Tue Dec 23 20:17:59 MST 2014
036 bl314 TBL.sh f10c4wghgdm outfrq3s 9s ......................................................FAIL! rc= 7 at Tue Dec 23 20:18:01 MST 2014
039 bl315 TBL.sh f10c3dm outfrq3s+cam3 9s .....................................................FAIL! rc= 7 at Tue Dec 23 20:18:02 MST 2014
041 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .........................................FAIL! rc= 7 at Tue Dec 23 20:18:03 MST 2014
044 bl317 TBL.sh f10c5cdm outfrq3s 9s .........................................................FAIL! rc= 7 at Tue Dec 23 20:18:05 MST 2014
047 bl319 TBL.sh f10c5nonedm outfrq3s_bam 9s ..................................................FAIL! rc= 7 at Tue Dec 23 20:18:06 MST 2014
050 bl320 TBL.sh f10c5nonedm rad_diag 9s ......................................................FAIL! rc= 7 at Tue Dec 23 20:18:08 MST 2014
053 bl321 TBL.sh f10c5cdm atrain 9s ...........................................................FAIL! rc= 7 at Tue Dec 23 20:18:09 MST 2014
056 bl367 TBL.sh f1.9c5mg2dm outfrq3s 9s ......................................................FAIL! rc= 7 at Tue Dec 23 20:44:22 MST 2014
059 bl420 TBL.sh f4c5mam7dm outfrq3s 9s .......................................................FAIL! rc= 7 at Tue Dec 23 20:44:27 MST 2014
062 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Tue Dec 23 20:44:32 MST 2014
065 bl734 TBL.sh h16c5naqdm outfrq3s_bam+aquaplanet_cam5 9s ...................................FAIL! rc= 7 at Tue Dec 23 20:44:45 MST 2014
ALL CAM configurations except ideal and adiabatic physics (and PORT) are
expected to change answers due to diffusion, gravity wave, and topography
file changes.
CAM tag used for the baseline comparison tests if different than previous
tag:Summarize any changes to answers, i.e.,
- what code configurations: CAM 3/4/5
- what platforms/compilers: All
- nature of change (roundoff; larger than roundoff but same climate; new
climate): New climate
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
https://svn-ccsm-models.cgd.ucar.edu/cam1/branch_tags/gw_fixes_tags/gw_fixes_n14_cam5_3_55
- platform/compilers: yellowstone/intel
- configure commandline: (CESM script defaults for FC5)
- build-namelist command (or complete namelist): Sole change was bnd_topo
bnd_topo = '/glade/p/cesmdata/cseg/inputdata/atm/cam/topo/consistent-topo-fv1.9x2.5_c130424.nc'
- MSS location of output:
/CCSM/csm/f.e13.FC5.f19_f19.GwFxTCam.002
(Evaluation run is #1894 in the experiment database.)
MSS location of control simulations used to validate new climate:
/CCSM/csm/f.e13.FC5.f19_f19.GwCtlCam.001
(Control is #1895 in the experiment database.)
URL for AMWG diagnostics output used to validate new climate:
http://webext.cgd.ucar.edu/FCLIMO/f.e13.FC5.f19_f19.GwFxTCam.002/atm/f.e13.FC5.f19_f19.GwFxTCam.002-f.e13.FC5.f19_f19.GwCtlCam.001/
===============================================================
===============================================================
Tag name: cam5_3_58
Originator(s): jedwards, santos, cacraig
Date: Dec 23, 2014
One-line Summary: pio changes needed for pio2, change to MG2 to allow PGI to compile/run properly
Purpose of changes:
- Changes required in preparation for pio2
- fix to MG2 to allow PGI to compile and run properly
- update PGI compiler on goldbach to 14.10
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: goldy
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/test_driver.sh
- update to PGI 14.10 on goldbach
M models/atm/cam/bld/mkDepends
M models/atm/cam/src/control/ncdio_atm.F90
M models/atm/cam/src/control/sat_hist.F90
M models/atm/cam/src/control/cam_restart.F90
M models/atm/cam/src/control/cam_history.F90
M models/atm/cam/src/physics/cam/subcol_utils.F90.in
M models/atm/cam/src/utils/cam_pio_utils.F90
M models/atm/cam/src/dynamics/sld/restart_dynamics.F90
M models/atm/cam/src/dynamics/sld/dyn_grid.F90
M models/atm/cam/src/dynamics/eul/restart_dynamics.F90
M models/atm/cam/src/dynamics/eul/dyn_grid.F90
M models/atm/cam/src/dynamics/eul/inidat.F90
M models/atm/cam/src/dynamics/se/dyn_grid.F90
M models/atm/cam/src/dynamics/se/inidat.F90
M models/atm/cam/src/dynamics/se/restart_dynamics.F90
M models/atm/cam/src/dynamics/fv/restart_dynamics.F90
M models/atm/cam/src/dynamics/fv/metdata.F90
M models/atm/cam/src/dynamics/fv/dyn_grid.F90
M models/atm/cam/src/dynamics/fv/inidat.F90
M SVN_EXTERNAL_DIRECTORIES
- changes made by Jim to prepare for pio2
M models/atm/cam/src/physics/cam/micro_mg_data.F90
M models/atm/cam/src/physics/cam/micro_mg_cam.F90
- changes to allow PGI to compile/run properly
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB
goldbach/nag: all BFB
goldbach/pgi or jaguar/pgi: all BFB
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_57
Originator(s): santos, cchen
Date: 2014/12/15
One-line Summary: Change FV vertical remapping from energy to temperature
Purpose of changes:
- In the FV dycore, vertical remapping of total energy can cause
significant numerical errors in WACCM due to the high model top. By
remapping temperature instead, this error is reduced.
- Fix a bug in MG2, and work around an Intel compiler bug that also affects
MG2.
Bugs fixed (include bugzilla ID):
- #2098: Double division error in sub-saturation check in MG2
- #2104: MG2 produces garbage output in some diagnostics using Intel 15
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
- Short changes to numerical science code only, not formally reviewed.
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/src/dynamics/fv/te_map.F90
- Replace total energy remapping with temperature remapping over
log(p).
- Remove some dead/unneeded code.
M models/atm/cam/src/physics/cam/micro_mg2_0.F90
- Initialize many variables at the top of the tend subroutine (they
should be overwritten in any case, but this initialization is
still necessary to workaround an apparent Intel bug).
- Remove extra factor of deltat that should not be applied to
mnuccd and vap_dep.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
009 bl203 TBL.sh fsd1.9c4wtsmlth outfrq3s_sd+sd_1975-2010_ccmi_refc1_waccm_tsmlt 9s ...........FAIL! rc= 7 at Fri Dec 12 17:52:24 MST 2014
012 bl322 TBL.sh f1.9c5nonedh outfrq3s_lb0 9s .................................................FAIL! rc= 7 at Fri Dec 12 17:59:04 MST 2014
016 bl326 TBL.sh f1.9c4wghgdm outfrq3s_2005+waccm_sc_2005-2100_cam4_rcp45 9s ..................FAIL! rc= 7 at Fri Dec 12 18:12:19 MST 2014
020 bl331 TBL.sh f4c4dh co2rmp+1850_cam4 9s ...................................................FAIL! rc= 7 at Fri Dec 12 18:15:26 MST 2014
024 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...................................................FAIL! rc= 7 at Fri Dec 12 18:21:48 MST 2014
026 bl333 TBL.sh f4c4paqdh aqpgro+aquaplanet_cam4 3s ..........................................FAIL! rc= 7 at Fri Dec 12 18:22:03 MST 2014
029 bl336 TBL.sh f4c4wmdh outfrq3s+waccm_1850_cam4 9s .........................................FAIL! rc= 7 at Fri Dec 12 18:33:53 MST 2014
032 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Fri Dec 12 18:44:23 MST 2014
035 bl373 TBL.sh f1.9c4mozdh outfrq3s+2000_cam4_trop_chem 9s ..................................FAIL! rc= 7 at Fri Dec 12 18:57:40 MST 2014
037 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...............................................FAIL! rc= 7 at Fri Dec 12 19:00:24 MST 2014
039 bl376 TBL.sh f1.9c4mozdh+testmech outfrq3s 9s .............................................FAIL! rc= 7 at Fri Dec 12 19:06:31 MST 2014
042 bl374 TBL.sh f1.9c4fastdh outfrq3s+1850-2005_cam4_super_fast_llnl 9s ......................FAIL! rc= 7 at Fri Dec 12 19:13:30 MST 2014
045 bl390 TBL.sh f1.9c4wmxh outfrq3s+waccmx_2000_cam4 9s ......................................FAIL! rc= 7 at Fri Dec 12 19:22:26 MST 2014
048 bl421 TBL.sh f1.9c5mam4dh outfrq3s 9s .....................................................FAIL! rc= 7 at Fri Dec 12 19:32:30 MST 2014
050 bl430 TBL.sh f1.9c5mzmam3dh outfrq3s_NEUwetdep 9s .........................................FAIL! rc= 7 at Fri Dec 12 19:41:19 MST 2014
053 bl440 TBL.sh f1.9c4soadh outfrq3s+soa_chem_megan_emis 9s ..................................FAIL! rc= 7 at Fri Dec 12 19:57:51 MST 2014
057 bl387 TBL.sh f1.9c4dm outfrq3s+1850-2005_cam4 9s ..........................................FAIL! rc= 7 at Fri Dec 12 20:05:21 MST 2014
060 bl388 TBL.sh f1.9c4m outfrq24h+1850_cam4 2d ...............................................FAIL! rc= 7 at Fri Dec 12 20:06:59 MST 2014
063 bl379 TBL.sh f1.9c4wmh outfrq24h+waccm_2005-2100_cam4_rcp45 2d ............................FAIL! rc= 7 at Fri Dec 12 20:24:27 MST 2014
goldbach/nag:
018 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Fri Dec 12 15:58:47 MST 2014
026 bl312 TBL.sh f2.5c5dm ghgrmp 9s ...........................................................FAIL! rc= 7 at Fri Dec 12 16:30:20 MST 2014
028 bl313 TBL.sh f10c3aqdm outfrq3s+aquaplanet_cam3 3s ........................................FAIL! rc= 7 at Fri Dec 12 16:32:35 MST 2014
031 bl314 TBL.sh f10c4wghgdm outfrq3s 9s ......................................................FAIL! rc= 7 at Fri Dec 12 16:40:36 MST 2014
033 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .........................................FAIL! rc= 7 at Fri Dec 12 16:42:49 MST 2014
036 bl318 TBL.sh f10c4cdm sat_hist 9s .........................................................FAIL! rc= 7 at Fri Dec 12 16:58:17 MST 2014
039 bl324 TBL.sh f4c4rrtmgdm outfrq3s 9s ......................................................FAIL! rc= 7 at Fri Dec 12 17:06:53 MST 2014
042 bl325 TBL.sh f10c5t5mdmsc outfrq3s_subcol 9s ..............................................FAIL! rc= 7 at Fri Dec 12 17:15:16 MST 2014
046 bl335 TBL.sh f10idm idphys 9s .............................................................FAIL! rc= 7 at Fri Dec 12 17:20:56 MST 2014
049 bl368 TBL.sh f1.9c5mg2dm outfrq3s_macmic2_classnuc 9s .....................................FAIL! rc= 7 at Fri Dec 12 17:52:50 MST 2014
055 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Fri Dec 12 18:25:18 MST 2014
058 bl805 TBL.sh f4c5carmtt1dm outfrq3s_carma 9s ..............................................FAIL! rc= 7 at Fri Dec 12 18:56:34 MST 2014
goldbach/pgi or jaguar/pgi:
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Fri Dec 12 16:03:18 MST 2014
031 bl312 TBL.sh f2.5c5dm ghgrmp 9s ...........................................................FAIL! rc= 7 at Fri Dec 12 16:37:26 MST 2014
033 bl313 TBL.sh f10c3aqdm outfrq3s+aquaplanet_cam3 3s ........................................FAIL! rc= 7 at Fri Dec 12 16:43:20 MST 2014
036 bl314 TBL.sh f10c4wghgdm outfrq3s 9s ......................................................FAIL! rc= 7 at Fri Dec 12 16:52:39 MST 2014
039 bl315 TBL.sh f10c3dm outfrq3s+cam3 9s .....................................................FAIL! rc= 7 at Fri Dec 12 16:59:55 MST 2014
041 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .........................................FAIL! rc= 7 at Fri Dec 12 17:07:07 MST 2014
044 bl317 TBL.sh f10c5cdm outfrq3s 9s .........................................................FAIL! rc= 7 at Fri Dec 12 17:17:39 MST 2014
047 bl319 TBL.sh f10c5nonedm outfrq3s_bam 9s ..................................................FAIL! rc= 7 at Fri Dec 12 17:25:40 MST 2014
050 bl320 TBL.sh f10c5nonedm rad_diag 9s ......................................................FAIL! rc= 7 at Fri Dec 12 17:28:20 MST 2014
053 bl321 TBL.sh f10c5cdm atrain 9s ...........................................................FAIL! rc= 7 at Fri Dec 12 17:32:18 MST 2014
056 bl367 TBL.sh f1.9c5mg2dm outfrq3s 9s ......................................................FAIL! rc= 7 at Fri Dec 12 18:03:10 MST 2014
059 bl420 TBL.sh f4c5mam7dm outfrq3s 9s .......................................................FAIL! rc= 7 at Fri Dec 12 18:19:48 MST 2014
062 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Fri Dec 12 18:35:28 MST 2014
All failures due to expected answer changes in the FV dycore and MG2.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
All FV and MG2.
- what platforms/compilers:
All.
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
Larger than roundoff but same climate. The MG2 bug fix may have a small
effect on climate, but no significant change in climate is observed in
standard CAM5 or WACCM4.
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_56
Originator(s): cacraig
Date: 12/09/15
One-line Summary: Update to CESM1_3_beta14 externals
Purpose of changes:
- Update to CESM1_3_beta14 externals
- Remove -DNO_SHR_VMATH compiler flag to match CESM settings (non-CAM bug which required this flag was fixed)
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: None (minor changes)
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/Makefile.in
- removed -DNO_SHR_VMATH to match the CESM settings
M SVN_EXTERNAL_DIRECTORIES
- updated to match CESM1_3_beta14
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All BFB except:
066 bl731 TBL.sh h16c3aqdh aqua+aquaplanet_cam3 9s ............................................FAIL! rc= 7 at Mon Dec 8 16:20:51 MST 2014
069 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ........................................FAIL! rc= 7 at Mon Dec 8 16:31:52 MST 2014
074 bl740 TBL.sh h30c5h outfrq3s_lb0 9s .......................................................FAIL! rc= 7 at Mon Dec 8 16:35:24 MST 2014
078 bl751 TBL.sh h30c4aqdm outfrq3s+aquaplanet_cam4 9s ........................................FAIL! rc= 7 at Mon Dec 8 16:50:21 MST 2014
These tests fail due to the change with removing the -DNO_SHR_VMATH compilation flag. They passed when this flag was used.
goldbach/nag: All BFB
goldbach/pgi or jaguar/pgi: All BFB
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_55
Originator(s): santos, cacraig
Date: 12/02/15
One-line Summary: Fix compiler issue on mira XLF
Purpose of changes:
- issues with default initialization using mira XLF compiler - one line workaround.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: None (minor changes)
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/src/physics/cam/micro_mg_data.F90
- one line fix for mira XLF compiler initialization issue
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All BFB
goldbach/nag: All BFB
goldbach/pgi or jaguar/pgi: All BFB
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_54
Originator(s): santos
Date: 2014/11/20
One-line Summary: Fix memory leak with the PGI compiler.
Purpose of changes:
- Fix memory leak for MG on PGI.
- Allow detection of the LAPACK library to find .so and .dylib files.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: None (minor changes)
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/configure
- Check for liblapack.so and liblapack.dylib as well as
liblapack.a.
M models/atm/cam/src/physics/cam/micro_mg_data.F90
- Do explicit deallocation on PGI.
- Suppress intent warning.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS.
goldbach/nag: All PASS.
goldbach/pgi or jaguar/pgi: All PASS.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_53
Originator(s): cacraig, santos
Date: 2014/11/18
One-line Summary: Update externals to at least cesm1_3_beta13
Purpose of changes:
- Update externals to cesm1_3_beta13 (with a couple beyond, as needed for
previous tags and Intel compiler bug workarounds).
- Work around compiler bugs.
Bugs fixed (include bugzilla ID): None
Describe any changes made to build system: Intel compiler flag changes.
Describe any changes made to the namelist: None
List any changes to the defaults for the boundary datasets: None
Describe any substantial timing or memory changes: None
Code reviewed by: none (trival changes).
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/test_driver.sh
- Updated yellowstone intel to 15.0.0 to match CESM compiler.
M models/atm/cam/bld/configure
- CLM rearranged their directory structure.
M models/atm/cam/bld/build-namelist
- CICE has changed their CAM flag from cam5 to cam4.
M models/atm/cam/src/physics/cam/micro_mg_cam.F90
- Avoid unpacking am_evp_st in MG1.5 or MG2, since it is only set
by MG1.0.
M models/atm/cam/src/physics/cam/micro_mg_utils.F90
- Add workaround for XLF compiler bug.
M models/atm/cam/src/dynamics/eul/inidat.F90
M models/atm/cam/src/dynamics/eul/stepon.F90
- Removed array subsections in subroutine argument lists, since
Intel's bounds checker complains when these arrays are
allocated with size 0.
M SVN_EXTERNAL_DIRECTORIES
- Updated to cesm1_3_beta13 externals, except Machines and CMake
are more recent because CAM needs them.
- CICE is more recent to fix Intel 14 bug.
M models/atm/cam/bld/Makefile.in
- Modify Intel settings to match CESM flags for yellowstone
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
009 bl203 TBL.sh fsd1.9c4wtsmlth outfrq3s_sd+sd_1975-2010_ccmi_refc1_waccm_tsmlt 9s ...........FAIL! rc= 7 at Mon Nov 17 17:10:13 MST 2014
037 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...............................................FAIL! rc= 7 at Mon Nov 17 18:17:55 MST 2014
045 bl390 TBL.sh f1.9c4wmxh outfrq3s+waccmx_2000_cam4 9s ......................................FAIL! rc= 7 at Mon Nov 17 18:39:07 MST 2014
060 bl388 TBL.sh f1.9c4m outfrq24h+1850_cam4 2d ...............................................FAIL! rc= 7 at Mon Nov 17 19:22:39 MST 2014
063 bl379 TBL.sh f1.9c4wmh outfrq24h+waccm_2005-2100_cam4_rcp45 2d ............................FAIL! rc= 7 at Mon Nov 17 19:39:54 MST 2014
066 bl731 TBL.sh h16c3aqdh aqua+aquaplanet_cam3 9s ............................................FAIL! rc= 7 at Mon Nov 17 19:42:41 MST 2014
069 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ........................................FAIL! rc= 7 at Mon Nov 17 19:53:31 MST 2014
074 bl740 TBL.sh h30c5h outfrq3s_lb0 9s .......................................................FAIL! rc= 7 at Mon Nov 17 19:57:02 MST 2014
078 bl751 TBL.sh h30c4aqdm outfrq3s+aquaplanet_cam4 9s ........................................FAIL! rc= 7 at Mon Nov 17 20:11:51 MST 2014
All changes are due to Intel update to version 15 and changes in compiler
flags. (If these changes are backed out, including the Machines update, all
tests pass.)
goldbach/nag: All BFB
goldbach/pgi or jaguar/pgi: All BFB
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: None for most compilers.
- what platforms/compilers: Intel only.
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_52
Originator(s): santos
Date: 2014/10/17
One-line Summary: Remove NO_SHR_VMATH from Makefile
Purpose of changes:
- Allow use of compiler-specific vector math functions in shr_vmath_mod for
CAM standalone, so that it remains bit-for-bit with CESM script runs.
- Use default physics load balancing settings for CAM-SE, since doing so
now performs better than using phys_loadbalance = 0 (h/t jedwards).
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
- No longer using -DNO_SHR_VMATH.
Describe any changes made to the namelist:
- CAM-SE, like CAM-FV, defaults to phys_loadbalance = 2.
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: none (trival changes).
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/Makefile.in
- Remove -DNO_SHR_VMATH.
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
M models/atm/cam/bld/build-namelist
- Remove setting of phys_loadbalance to 0 for CAM-SE.
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- Correct mistake in documentation of the phys_loadbalance default.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
066 bl731 TBL.sh h16c3aqdh aqua+aquaplanet_cam3 9s ............................................FAIL! rc= 7 at Thu Oct 16 22:33:08 MDT 2014
069 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ........................................FAIL! rc= 7 at Thu Oct 16 22:44:03 MDT 2014
074 bl740 TBL.sh h30c5h outfrq3s_lb0 9s .......................................................FAIL! rc= 7 at Thu Oct 16 22:47:35 MDT 2014
078 bl751 TBL.sh h30c4aqdm outfrq3s+aquaplanet_cam4 9s ........................................FAIL! rc= 7 at Thu Oct 16 23:02:41 MDT 2014
All CAM-SE runs with non-trival physics (i.e. not adiabatic or ideal) have
answer changes due to the NO_SHR_VMATH change. Note that the outfrq3s_lb0
baseline failed, but test eq740 (TEQ of different phys_loadbalance
settings) passed. This shows that the answer changes are due to use of
shr_vmath_mod in CAM-SE's dp_coupling layer, and not due to the
phys_loadbalance change.
goldbach/nag: All PASS.
goldbach/pgi or jaguar/pgi: All PASS.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- CAM-SE, except adiabatic/ideal, when using CAM Makefile.
- what platforms/compilers:
- Only those with special shr_vmath_mod implementations defined (in
practice, the Intel compiler is the only one left still in common use).
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
- Roundoff.
If bitwise differences were observed, how did you show they were no worse
than roundoff?
- Monitor outputs from first (instantaneous) timestep; no fields in any of
the history outputs had a relative difference much greater than machine
epsilon.
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_51
Originator(s): santos andrew morrison bogensch caldwell
Date: 2014/10/16
One-line Summary: Add new MG2 microphysics scheme.
Purpose of changes:
- Add MG2, a new version of the MG microphysics with prognostic
precipitation.
- Add new macro/micro and microphysics-only substepping features.
- Add first unit tests to CAM.
Bugs fixed (include bugzilla ID):
- Bug 1603: microphysics: BAM aerosols has hardcoded assumption that MG2 is
substepped twice
- Bug 2046: Fv dycore spmd_dyn.F90 and intel 14.x compiler
- Bug 2054: bug in cosp ice diagnostic
Describe any changes made to build system:
- Add a csm_share include directory.
- PGI builds now use the "-traceback" flag in DEBUG mode.
Describe any changes made to the namelist:
- Add cld_macmic_num_steps, which controls joint substepping of
macrophysics and microphysics within a physics timestep.
- Add micro_mg_num_steps, which controls microphysics-only substepping
within a macro/micro substep.
- Add nucleate_ice_subgrid, a new tuning parameter for ice nucleation.
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton, cacraig, CAM reviewers
List all subroutines eliminated:
List all subroutines added and what they do:
A models/atm/cam/src/physics/cam/CMakeLists.txt
- CMakeLists used to add files to be built for unit tests.
A models/atm/cam/src/physics/cam/micro_mg2_0.F90
- Portable science module for new MG2 code.
A models/atm/cam/src/physics/cam/micro_mg_data.F90
- Utility types for dealing with output data.
A models/atm/cam/test/unit/CMakeLists.txt
A models/atm/cam/test/unit/micro_mg_data/
A models/atm/cam/test/unit/micro_mg_data/CMakeLists.txt
A models/atm/cam/test/unit/micro_mg_data/test_MGFieldPostProc.pf
A models/atm/cam/test/unit/micro_mg_data/test_MGPacker.pf
A models/atm/cam/test/unit/micro_mg_data/test_MGPostProc.pf
A models/atm/cam/test/unit/mock/
A models/atm/cam/test/unit/mock/CMakeLists.txt
A models/atm/cam/test/unit/mock/shr_sys_mod.nompi_abortthrows.F90
- Unit tests for micro_mg_data module.
List all existing files that have been modified, and describe the changes:
M SVN_EXTERNAL_DIRECTORIES
- Update csm_share and Machines externals for utility code used by
the MG interface.
- Add unit_testing external.
M models/atm/cam/bld/Makefile.in
- Add new csm_share include directory.
- Add "-traceback" option for PGI debug runs.
M models/atm/cam/bld/build-namelist
M models/atm/cam/bld/config_files/definition.xml
M models/atm/cam/bld/configure
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- Add MG2 microphysics as an option.
- Add new namelist options.
- Set tuning parameters for MG2.
M models/atm/cam/src/dynamics/fv/spmd_dyn.F90
- Workaround for an Intel 14 and 15 regression (h/t jedwards).
M models/atm/cam/src/physics/cam/check_energy.F90
M models/atm/cam/src/physics/cam/physics_types.F90
- Add rain and snow to lists of advected water constituents.
- Add physics_ptend_scale to scale an entire physics tendency by
some scalar factor.
M models/atm/cam/src/physics/cam/cospsimulator_intr.F90
- Fix bug where COSP used liquid instead of ice for an ice
diagnostic (via Po-lun Ma, hannay, cacraig).
M models/atm/cam/src/physics/cam/micro_mg1_0.F90
M models/atm/cam/src/physics/cam/micro_mg1_5.F90
- Change MG optional input handling.
- Give MG1.0 a "get_cols" routine.
- Significant MG1.5 changes from tracking parts of MG2 development.
M models/atm/cam/src/physics/cam/micro_mg_cam.F90
- Add MG2 and new inputs/outputs for prognostic precipitation and
new diagnostics.
- Add microphysics-only substepping at this layer.
- Match MG1.0 interface changes.
- Use micro_mg_data types to handle packing and output averages.
M models/atm/cam/src/physics/cam/micro_mg_utils.F90
- Change value of hard-coded eii parameter.
- Minor changes (e.g. comments).
M models/atm/cam/src/physics/cam/microp_aero.F90
- Remove (incorrect) assumption in BAM that npccn should be tweaked
to account for MG substepping. This was not even correct for MG1,
and hard-codes a value that can now vary using the new
microphysics substepping.
M models/atm/cam/src/physics/cam/nucleate_ice.F90
M models/atm/cam/src/physics/cam/nucleate_ice_cam.F90
- "subgrid" is now a namelist option.
M models/atm/cam/src/physics/cam/phys_control.F90
M models/atm/cam/src/physics/cam/physpkg.F90
- Add macro/micro sub-stepping.
D models/atm/cam/test/system/config_files/f1.9c5mg1_5m
A models/atm/cam/test/system/config_files/f1.9c5mg1_5dm
A models/atm/cam/test/system/config_files/f1.9c5mg2dm
M models/atm/cam/test/system/input_tests_master
A models/atm/cam/test/system/nl_files/outfrq3s_macmic2_classnuc
M models/atm/cam/test/system/tests_pretag_goldbach_nag
M models/atm/cam/test/system/tests_pretag_goldbach_pgi
- MG1.5 test is now a DEBUG test.
- Add MG2 test.
- Add macro/micro substepping test with use_hetfrz_classnuc
enabled.
- Add microphysics tests to NAG and PGI pretag lists on goldbach.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
082 eq993 TEQ_ccsm.sh ne30_ne30 FC5 h30c5h fcase 9s ...........................................FAIL! rc= 7 at Wed Oct 15 23:26:35 MDT 2014
- Test fails due to csm_share update that uses NO_SHR_VMATH, which CAM
defines, but CESM does not for the Intel compiler. Reconciling the two is
deferred to a near-future tag.
goldbach/nag:
049 bl368 TBL.sh f1.9c5mg2dm outfrq3s_macmic2_classnuc 9s .....................................FAIL! rc= 7 at Wed Oct 15 18:48:13 MDT 2014
052 bl734 TBL.sh h16c5naqdm outfrq3s_bam+aquaplanet_cam5 9s ...................................FAIL! rc= 7 at Wed Oct 15 19:09:32 MDT 2014
- bl368 is a new test with no existing baseline.
- bl734 is affected by the fix for bug 1603, which affects all cases with
BAM and MG microphysics.
goldbach/pgi or jaguar/pgi:
044 bl317 TBL.sh f10c5cdm outfrq3s 9s .........................................................FAIL! rc= 7 at Wed Oct 15 18:12:45 MDT 2014
047 bl319 TBL.sh f10c5nonedm outfrq3s_bam 9s ..................................................FAIL! rc= 7 at Wed Oct 15 18:21:36 MDT 2014
050 bl320 TBL.sh f10c5nonedm rad_diag 9s ......................................................FAIL! rc= 7 at Wed Oct 15 18:24:13 MDT 2014
056 bl367 TBL.sh f1.9c5mg2dm outfrq3s 9s ......................................................FAIL! rc= 7 at Wed Oct 15 19:01:53 MDT 2014
065 bl734 TBL.sh h16c5naqdm outfrq3s_bam+aquaplanet_cam5 9s ...................................FAIL! rc= 7 at Wed Oct 15 20:01:59 MDT 2014
- bl317 changes answers for COSP diagnostics only, due to bug 2054 being
fixed.
- bl367 is a new test with no existing baseline.
- bl319, bl320, and bl734 are (prescribed) bulk modal aerosol cases with
MG, so are affected by the fix for bug 1603.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- Fix for bug 1603 affects all BAM+MG cases.
- Fix for bug 2054 affects all COSP cases.
- what platforms/compilers: All.
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
- BAM+MG is larger than roundoff. This configuration is not
scientifically supported at this time.
- COSP changes diagnostics only, larger than roundoff
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_50
Originator(s): eaton
Date: Mon Sep 22 14:22:41 MDT 2014
One-line Summary: updates to heterogeneous freezing code
Purpose of changes:
. Updates to heterogeneous freezing code.
- Finish support for MAM7.
- Change two parameter values to match published values.
- Change name of modules and associated method names from hoose_hetfrz to
hetfrz_classnuc.
. Add support for low resolution SE grid: ne5np4
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist:
. use_hoose_hetfrz -> use_hetfrz_classnuc
. hist_hoose_hetfrz -> hist_hetfrz_classnuc
List any changes to the defaults for the boundary datasets:
. added some defaults for low res SE grids
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
models/atm/cam/src/physics/cam/hoose_hetfrz.F90
models/atm/cam/src/physics/cam/hoose_hetfrz_cam.F90
. Rename hoose_hetfrz to hetfrz_classnuc
List all subroutines added and what they do:
models/atm/cam/src/physics/cam/hetfrz_classnuc.F90
models/atm/cam/src/physics/cam/hetfrz_classnuc_cam.F90
. Rename hoose_hetfrz to hetfrz_classnuc
. Finish support of MAM7.
. Minor changes for the values of two variables: "theta_imm_bc" and
"dga_imm_bc" to be consistent with published paper.
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/config_files/horiz_grid.xml
. add parameter values for low res SE grids
models/atm/cam/bld/configure
. remove the CAM ldflags from the mct_ldflags passed to MCT's configure.
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. add defaults for low res SE grid: ne5np4
models/atm/cam/bld/namelist_files/namelist_definition.xml
models/atm/cam/src/physics/cam/micro_mg1_0.F90
models/atm/cam/src/physics/cam/micro_mg1_5.F90
models/atm/cam/src/physics/cam/micro_mg_cam.F90
models/atm/cam/src/physics/cam/microp_aero.F90
models/atm/cam/src/physics/cam/nucleate_ice_cam.F90
models/atm/cam/src/physics/cam/phys_control.F90
. hoose_hetfrz -> hetfrz_classnuc
models/atm/cam/src/physics/cam/nucleate_ice.F90
. hoose_hetfrz -> hetfrz_classnuc
. improve comments
. Put change to conditional statement inside CPP USE_XLIU_MOD. This mod
needs a climate evaluation before being added to cam5 physics.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS
goldbach/nag: All PASS
goldbach/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: none.
===============================================================
===============================================================
Tag name: cam5_3_49
Originator(s): eaton
Date: Mon Sep 15 10:46:16 MDT 2014
One-line Summary: add pre-existing ice and Hoose heterogeneous freezing options
Purpose of changes:
. Add pre-existing ice option to nucleate_ice code.
. Add option for Hoose heterogeneous freezing parameterization.
. Improve the microp_aero driver by removing code that belonged in a CAM
specific interface for the nucleate_ice parameterization and adding the
missing CAM interface layer.
. Add two new functions to the rad_constituents interfaces to make it
easier to access the mode and specie indices for specific modes and
specie types.
Bugs fixed (include bugzilla ID):
bugzilla 2036: memory leak in cam_pio_utils
Describe any changes made to build system: none
Describe any changes made to the namelist:
Add new variables:
use_preexisting_ice - turn on the pre-existing ice option
hist_preexisting_ice - add diagnostic output for the pre-existing ice
option to the history file
use_hoose_hetfrz - turn on the Hoose hetergeneous freezing option
hist_hoose_hetfrz - add diagnostic output for Hoose heterogeneous freezing
to the history file
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: CAM SEs on Wed Sep 10.
List all subroutines eliminated:
List all subroutines added and what they do:
models/atm/cam/src/physics/cam/hoose_hetfrz.F90
. module for the Hoose heterogeneous freezing parameterization
models/atm/cam/src/physics/cam/hoose_hetfrz_cam.F90
. CAM interface layer for the hoose_hetfrz module
models/atm/cam/src/physics/cam/nucleate_ice_cam.F90
. CAM interface layer for the nucleate_ice module
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/Makefile.in
. add options to gfortran debug flags to trap FPEs
models/atm/cam/bld/namelist_files/namelist_definition.xml
. add logical variable use_hoose_hetfrz to enable Hoose heterogeneous
freezing (default: false)
. add logical variable hist_hoose_hetfrz to add default history output for
Hoose heterogeneous freezing (default: false)
. add use_preexisting_ice to enable treatment of pre-existing ice in the
ice_nucleation module (default: false)
. add hist_preexisting_ice to add default history output for the treatment
of pre-existing ice in the ice_nucleation module (default: false)
models/atm/cam/src/physics/cam/micro_mg1_0.F90
models/atm/cam/src/physics/cam/micro_mg1_5.F90
models/atm/cam/src/physics/cam/micro_mg_cam.F90
. add code to optionally make use of three new inputs provided by the Hoose
heterogeneous freezing code
models/atm/cam/src/physics/cam/microp_aero.F90
. move code belonging to the CAM interface for nucleate_ice into the new
module, nucleate_ice_cam, created for that purpose.
. add method calls for the new Hoose heterogeneous freezing module.
models/atm/cam/src/physics/cam/ndrop.F90
. add new output for activation fraction for aerosol number
models/atm/cam/src/physics/cam/nucleate_ice.F90
. add option to treat pre-existing ice
models/atm/cam/src/physics/cam/phys_control.F90
. add use_hoose_hetfrz option
models/atm/cam/src/physics/cam/rad_constituents.F90
. add function rad_cnst_get_mode_idx to find the mode index for a specific
mode type.
. add function rad_cnst_get_spec_idx to find the specie index for a specific
specie type.
models/atm/cam/src/utils/cam_pio_utils.F90
. function get_phys_ldof
- change local variable localhmask from pointer to allocatable, and add
explicit deallocate at end of function body.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS
goldbach/nag: All PASS
goldbach/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: bit-for-bit. There are currently no
tests that turn on the new options. May be good to add this with the MG2
code option.
===============================================================
===============================================================
Tag name: cam5_3_48
Originator(s): cacraig
Date: 09/11/2014
One-line Summary: Update CAM regression tests, default compilation is now caldera, update homme and minor bug fix to gw_drag output
Purpose of changes:
- Update CAM regression tests based on review by CAM SE group
- Since the gpgpu queue (exclusive node) can sometimes hold in the queue a long time, the default
is now caldera. A test of compilation times showed a difference of a few seconds in a series of
compilations which took 1.5 hours to compile. Introduced CALDERA_BATCHQ environment variable if
user wants to override.
- homme updated with bugs fixed to allow to work in CAM
- minor bug fix to gw_drag output
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
D models/atm/cam/test/system/tests_pretag_bluefire
D models/atm/cam/test/system/tests_posttag_bluefire
D models/atm/cam/test/system/tests_pretag_jaguarpf_pgi_cb
D models/atm/cam/test/system/tests_pretag_jaguarpf_pgi
D models/atm/cam/test/system/tests_posttag_jaguar
D models/atm/cam/test/system/tests_posttag_jaguar_cb
D models/atm/cam/test/system/posttag_cron_bangkok.sh
D models/atm/cam/test/system/posttag_cron_bluevista.sh
- deleted tests for old machines
D models/atm/cam/test/system/config_files/hn16c5aqdm
- renamed test to h16c5naqdm
D models/atm/cam/test/system/tests_pretag_titan_pgi
D models/atm/cam/test/system/tests_posttag_titan_cb
D models/atm/cam/test/system/tests_posttag_titan
D models/atm/cam/test/system/tests_pretag_hopper_pgi
D models/atm/cam/test/system/tests_pretag_edison_intel
- removed pretag and posttag from names for remote machine tests
List all subroutines added and what they do:
A + models/atm/cam/test/system/config_files/h16c5naqdm
- renamed test from hn16c5aqdm
A models/atm/cam/test/system/config_files/h16adh
A models/atm/cam/test/system/config_files/f10idm
- configurations for new tests
A models/atm/cam/test/system/gen-test-coverage
A models/atm/cam/test/system/gen-test-table
- perl scripts to create web page CAM test information
A models/atm/cam/test/system/nl_files/outfrq3s_lb0
A models/atm/cam/test/system/nl_files/outfrq3s_lb2
A models/atm/cam/test/system/nl_files/outfrq9s
- namelists for new tests
A + models/atm/cam/test/system/tests_titan_pgi
A + models/atm/cam/test/system/tests_hopper_pgi
A + models/atm/cam/test/system/tests_edison_intel
A + models/atm/cam/test/system/tests_titan
A + models/atm/cam/test/system/tests_titan_cb
- removed pretag and posttag from names for remote machine tests
List all existing files that have been modified, and describe the changes:
M models/atm/cam
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
- updated to new homme trunk tag
M models/atm/cam/test/system/test_driver.sh
- introduced CALDERA_BATCHQ and changed default compilation queue for yellowstone jobs to caldera from gpgpu
M models/atm/cam/test/system/tests_pretag_goldbach_nag
M models/atm/cam/test/system/tests_pretag_yellowstone
- introduced new tests and removed old ones
M models/atm/cam/test/system/TEQ_ccsm.sh
- now compares cam.i files also
M models/atm/cam/test/system/TSM_ccsm.sh
- selects outfrq file based on run length and output frequency type
M models/atm/cam/test/system/nl_files/fcase
- changed to 9 time steps and request initial files
M models/atm/cam/test/system/input_tests_master
- introduced new tests
M models/atm/cam/bld/cam.buildnml.csh
- Introduced a check for NTHRDS_ATM > 1 and sets smp
M models/atm/cam/bld/configure
- add CPP def -DHORIZ_OPENMP if se and smp is on
M models/atm/cam/src/physics/cam/gw_drag.F90
- changed format statement from i2 to i0 as integer could be > 99 and intel exits if not large enough
M models/atm/cam/src/dynamics/se/dyn_comp.F90
- ifdef was changed from ELEMENT_OPENMP to COLUMN_OPENMP or HORIZ_OPENMP
M models/atm/cam/src/dynamics/se/interp_mod.F90
M models/atm/cam/src/dynamics/se/inidat.F90
M models/atm/cam/src/dynamics/se/gravity_waves_sources.F90
M models/atm/cam/src/dynamics/se/nctopo_util_mod.F90
M models/atm/cam/src/dynamics/se/stepon.F90
- interface change for initEdgeBuffer
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
009 bl203 TBL.sh fsd1.9c4wtsmlth outfrq3s_sd+sd_1975-2010_ccmi_refc1_waccm_tsmlt 9s .....FAIL! rc= 7 at Tue Sep 9 19:09:31 MDT 2014
012 bl322 TBL.sh f1.9c5nonedh outfrq3s_lb0 9s ...........................................FAIL! rc= 7 at Tue Sep 9 19:21:05 MDT 2014
045 bl390 TBL.sh f1.9c4wmxh outfrq3s+waccmx_2000_cam4 9s ................................FAIL! rc= 7 at Tue Sep 9 20:49:20 MDT 2014
- new tests
063 bl379 TBL.sh f1.9c4wmh outfrq24h+waccm_2005-2100_cam4_rcp45 2d ......................FAIL! rc= 7 at Tue Sep 9 21:52:24 MDT 2014
- Changed use case
074 bl740 TBL.sh h30c5h outfrq3s_lb0 9s .................................................FAIL! rc= 7 at Tue Sep 9 22:09:51 MDT 2014
- Changed the loadbalance and caused automatic cprnc to fail. Hand compared select files and they are BFB
goldbach/nag:
003 bl335 TBL.sh f10idm idphys 9s .......................................................FAIL! rc= 7 at Thu Sep 11 10:33:45 MDT 2014
- New test
046 bl734 TBL.sh h16c5naqdm outfrq3s_bam+aquaplanet_cam5 9s .............................FAIL! rc= 7 at Wed Sep 10 19:28:51 MDT 2014
- Renamed test caused automatic cprnc to fail. Hand compared select files and they are BFB
goldbach/pgi or jaguar/pgi:
062 bl734 TBL.sh h16c5naqdm outfrq3s_bam+aquaplanet_cam5 9s .............................FAIL! rc= 7 at Tue Sep 9 17:42:42 MDT 2014
- Renamed test caused automatic cprnc to fail. Hand compared select files and they are BFB
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_47
Originator(s): santos
Date: 2014/09/05
One-line Summary: Minor bug fix for MG microphysics.
Purpose of changes:
- Fix a change in cam5_3_26 that inadvertently made ice limiters for
outputs to the radiation inconsistent with the value of dcs internally
used by MG1. ("dcs" is the ice->snow autoconversion threshold.)
- Tweak test_driver.sh so that it does not load two incompatible netcdf
modules on yellowstone.
Bugs fixed (include bugzilla ID):
- Bug 2038: MG uses wrong limiter on cloud ice particle size
Describe any changes made to build system: None
Describe any changes made to the namelist: None
List any changes to the defaults for the boundary datasets: None
Describe any substantial timing or memory changes: None
Code reviewed by: None (straightforward bug fix)
List all subroutines eliminated: None
List all subroutines added and what they do: None
List all existing files that have been modified, and describe the changes:
M models/atm/cam/src/physics/cam/micro_mg_cam.F90
M models/atm/cam/src/physics/cam/micro_mg_utils.F90
- Set cloud ice particle properties using version-specific value of
dcs.
M models/atm/cam/test/system/test_driver.sh
- Remove load of the netcdf module, so that netcdf-mpi is correctly
loaded for yellowstone. (The "error" from the second load is not
fatal, and the netCDF version does not affect answers, so this
fix is only for I/O performance, and to match the environment
used for running with the CESM scripts.)
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
011 bl137 TBL.sh e48c5h outfrq24h 2d ..........................................................FAIL! rc= 7 at Thu Sep 4 23:09:53 MDT 2014
014 bl322 TBL.sh f1.9c5nonedh outfrq3s 9s .....................................................FAIL! rc= 7 at Thu Sep 4 23:16:15 MDT 2014
019 bl330 TBL.sh f4c5paqdh aqpgro+aquaplanet_cam5 3s ..........................................FAIL! rc= 7 at Thu Sep 4 23:27:54 MDT 2014
027 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...................................................FAIL! rc= 7 at Thu Sep 4 23:37:37 MDT 2014
042 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ..........................................................FAIL! rc= 7 at Thu Sep 4 23:56:27 MDT 2014
046 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Fri Sep 5 00:09:03 MDT 2014
049 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .........................................................FAIL! rc= 7 at Fri Sep 5 00:17:25 MDT 2014
056 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...............................................FAIL! rc= 7 at Fri Sep 5 00:40:57 MDT 2014
065 bl421 TBL.sh f1.9c5mam4dh outfrq3s 9s .....................................................FAIL! rc= 7 at Fri Sep 5 01:09:52 MDT 2014
067 bl430 TBL.sh f1.9c5mzmam3dh outfrq3s_NEUwetdep 9s .........................................FAIL! rc= 7 at Fri Sep 5 01:18:46 MDT 2014
086 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ........................................FAIL! rc= 7 at Fri Sep 5 02:16:46 MDT 2014
088 bl740 TBL.sh h30c5h outfrq3s 9s ...........................................................FAIL! rc= 7 at Fri Sep 5 02:19:29 MDT 2014
goldbach/nag:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s ...........................................................FAIL! rc= 7 at Thu Sep 4 14:27:28 MDT 2014
010 bl112 TBL.sh e8c5dm ghgrmp 9s .............................................................FAIL! rc= 7 at Thu Sep 4 14:37:22 MDT 2014
014 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Thu Sep 4 14:46:33 MDT 2014
018 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Thu Sep 4 15:01:37 MDT 2014
026 bl312 TBL.sh f2.5c5dm ghgrmp 9s ...........................................................FAIL! rc= 7 at Thu Sep 4 15:33:04 MDT 2014
033 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .........................................FAIL! rc= 7 at Thu Sep 4 15:47:34 MDT 2014
042 bl325 TBL.sh f10c5t5mdmsc outfrq3s_subcol 9s ..............................................FAIL! rc= 7 at Thu Sep 4 16:21:51 MDT 2014
046 bl734 TBL.sh hn16c5aqdm outfrq3s_bam+aquaplanet_cam5 9s ...................................FAIL! rc= 7 at Thu Sep 4 16:46:22 MDT 2014
049 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Thu Sep 4 16:59:57 MDT 2014
052 bl805 TBL.sh f4c5carmtt1dm outfrq3s_carma 9s ..............................................FAIL! rc= 7 at Thu Sep 4 17:27:17 MDT 2014
goldbach/pgi or jaguar/pgi:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s ...........................................................FAIL! rc= 7 at Thu Sep 4 20:17:10 MDT 2014
010 bl112 TBL.sh e8c5dm ghgrmp 9s .............................................................FAIL! rc= 7 at Thu Sep 4 20:17:12 MDT 2014
012 bl113 TBL.sh e8c5paqdm aqpgro+aquaplanet_cam5 3s ..........................................FAIL! rc= 7 at Thu Sep 4 20:17:12 MDT 2014
016 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Thu Sep 4 20:17:15 MDT 2014
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Thu Sep 4 20:17:16 MDT 2014
031 bl312 TBL.sh f2.5c5dm ghgrmp 9s ...........................................................FAIL! rc= 7 at Thu Sep 4 20:17:26 MDT 2014
041 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .........................................FAIL! rc= 7 at Thu Sep 4 20:17:27 MDT 2014
044 bl317 TBL.sh f10c5cdm outfrq3s 9s .........................................................FAIL! rc= 7 at Thu Sep 4 20:17:29 MDT 2014
047 bl319 TBL.sh f10c5nonedm outfrq3s_bam 9s ..................................................FAIL! rc= 7 at Thu Sep 4 20:17:30 MDT 2014
050 bl320 TBL.sh f10c5nonedm rad_diag 9s ......................................................FAIL! rc= 7 at Thu Sep 4 20:17:31 MDT 2014
053 bl321 TBL.sh f10c5cdm atrain 9s ...........................................................FAIL! rc= 7 at Thu Sep 4 20:17:32 MDT 2014
056 bl420 TBL.sh f4c5mam7dm outfrq3s 9s .......................................................FAIL! rc= 7 at Thu Sep 4 20:17:37 MDT 2014
059 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Thu Sep 4 20:17:41 MDT 2014
062 bl734 TBL.sh hn16c5aqdm outfrq3s_bam+aquaplanet_cam5 9s ...................................FAIL! rc= 7 at Thu Sep 4 20:44:35 MDT 2014
- All failures are CAM5 baseline changes due to MG changing answers.
Summarize any changes to answers, i.e.,
- what code configurations: All with MG microphysics (CAM5)
- what platforms/compilers: All
- nature of change (roundoff; larger than roundoff but same climate; new
climate): Larger than roundoff, same climate.
===============================================================
===============================================================
Tag name: cam5_3_46
Originator(s): fvitt, Steve Ghan
Date: 28 Aug 2014
One-line Summary: Include dust in the Aitken aerosol mode of MAM4
Purpose of changes:
- add dust to the Aitken aerosol mode in MAM4
- corrections to build-namelist use cases
- change the PE threading in the regression tests for yellowstone to
improve performance
- misc bug fixes
Bugs fixed (include bugzilla ID):
- fixed bug #2020 - online interpolation of vector fields from SE dynamics
grid broken (Jim Edwards)
- fixed bug #2003 - microp_uniform needs to not be hardwired to false (Cheryl Craig)
- a bug fix for cases where short lived (not transported) chemistry species
are not found in the IC file
- corrections in build namelist use cases:
. 1850-PD_cam5 (used in compset F1850PDC5)
. 1950-2100_ccmi_refc2_rcp85_waccm_ma
. 1950-2100_ccmi_refc2_waccm_tsmlt
. 1950-2100_ccmi_refc2_trop_strat_soa
. 2000_cam4_trop_strat_soa (used in compset FSSOA)
. 2000_cam5_trop_strat_mam3 (used in compsets BSTRATMAM3, FSTRATMAM3)
. 2000_cam5_trop_strat_mam7 (used in compsets BSTRATMAM7, FSTRATMAM7)
. 2000_cam4_trop_chem (used in compsets BMOZ, FMOZ)
. waccmx_2000_cam4 (used in compset FWX)
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton, santos, cacraig
List all subroutines eliminated:
D models/atm/cam/test/system/nl_files/user_nl_cam
- now use outfrq24h for CESM runs with run lengths in days
List all subroutines added and what they do:
A models/atm/cam/test/system/nl_files/outfrq1s
- namelist settings for high-frequency history output for 1-step tests
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/TCB_ccsm.sh
- move setting the run-time options in user_nl_cam to TSM_ccsm.sh
M models/atm/cam/test/system/test_driver.sh
- set CAM_THREADS=2 and CAM_RESTART_THREADS=1 and min_cpus_per_task=1 (not over subscribing the yellowstone nodes)
-- leaving CAM_THREADS=32 CAM_RESTART_THREADS=64 avoids performance issues that
arise when OMP_NUM_THREADS=1 or in pure MPI runs
M models/atm/cam/test/system/TSM_ccsm.sh
- set user_nl_cam differently for run times lengths in steps -- high-frequency history output
. nl_files/outfrq24h is copied to user_nl_cam for run times lengths in days
. nl_files/outfrq1s is copied to user_nl_cam for run times lengths in steps
- remove all the cam.h* history files from the run directory (not just the *cam.h0* files)
- copy cam.i.* and cam.r.* files (in addition to cam.h* files) into the test dir
M models/atm/cam/test/system/input_tests_master
- adjustments to use case files used in cam-chem regression tests
- run length of CESM chemistry preprocessor test changed from 2 days to 1 step
M models/atm/cam/test/system/TNE_ccsm.sh
- compare not only cam.h0 files but also all cam.h*, cam.i.* and cam.r.* files
M models/atm/cam/bld/configure
- trop_mam4 number of advect species increased from 23 to 24
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- update to default mam4_mode2_file
M models/atm/cam/bld/namelist_files/master_aer_drydep_list.xml
- include new dst_a2 (Aitken mode of dust) tracer in dry dep list
M models/atm/cam/bld/namelist_files/master_aer_wetdep_list.xml
- include new dst_a2 (Aitken mode of dust) tracer in wet dep list
M models/atm/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_rcp85_waccm_ma.xml
- replaced 'SRFRAD' with 'FLDS'
- removed the 'D' from the OddOx* rxtn groups specifications
M models/atm/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_waccm_tsmlt.xml
M models/atm/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_trop_strat_soa.xml
- corrections in OC and BC emissions
M models/atm/cam/bld/namelist_files/use_cases/1850-PD_cam5.xml
- fixed dates in 1850-2100 DMS emis file
M models/atm/cam/bld/namelist_files/use_cases/2000_cam4_trop_strat_soa.xml
M models/atm/cam/bld/namelist_files/use_cases/2000_cam5_trop_strat_mam3.xml
M models/atm/cam/bld/namelist_files/use_cases/2000_cam4_trop_chem.xml
- correction in specifying prescribed stratospheric aerosols
M models/atm/cam/bld/namelist_files/use_cases/2000_cam5_trop_strat_mam7.xml
- correction in specifying prescribed stratospheric aerosols
- correction on pom_a3 vertical emissions file name
M models/atm/cam/bld/namelist_files/use_cases/waccmx_2000_cam4.xml
- correction in 1.9x2.5 IC file
M models/atm/cam/bld/build-namelist
- include Aitken mode of dust in MAM4 mode specifier
M models/atm/cam/src/chemistry/pp_trop_mam4/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_trop_mam4/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_mam4/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_mam4/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_trop_mam4/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mam4/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mam4/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_trop_mam4/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_mam4/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_trop_mam4/m_spc_id.F90
M models/atm/cam/src/chemistry/pp_trop_mam4/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_mam4/chem_mech.in
- include Aitken mode of dust in MAM4 chemistry mechanism
M models/atm/cam/src/chemistry/modal_aero/modal_aero_data.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_initialize_data.F90
M models/atm/cam/src/chemistry/modal_aero/dust_model.F90
- include Aitken mode of dust in MAM4 aerosol model
M models/atm/cam/src/chemistry/mozart/short_lived_species.F90
- a bug fix for cases where short lived (not transported) chemistry species
are not found in the IC file
M models/atm/cam/src/dynamics/se/interp_mod.F90
- fixed bug #2020 - Jim Edwards
M models/atm/cam/src/physics/cam/micro_mg_cam.F90
- fixed bug #2003 - Cheryl Craig
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
065 bl421 TBL.sh f1.9c5mam4dh outfrq3s 9s .....................................................FAIL! rc= 7 at Mon Aug 25 20:51:24 MDT 2014
goldbach/nag:
059 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Wed Aug 27 13:26:39 MDT 2014
goldbach/pgi:
049 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 7 at Wed Aug 27 13:19:14 MDT 2014
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_45
Originator(s): cacraig
Date: 07/21/14
One-line Summary: autogeneration of .F90 files from .F90.in, goldbach now uses modules, misc cleanup
Purpose of changes:
- Files with .in.F90 extension are now automatically generated into .F90 by the CAM makefile. This mimics the
functionality which is done with CESM for these files.
- goldbach compilations in test_driver.sh are now using modules
- wei96 had a deprecated PAUSE command which has been commented out
- two non-existent directories were being referenced in configure. These have been removed
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
- In the past, .F90.in files needed to be run by hand through genf90 to create the .F90 files. This has now been
incoporated in the CAM Makefile.in/mkSrcFiles/mkDepends in much that same way that CESM has incorporated it.
- test_driver.sh now uses modules on goldbach to select the appropriate libraries
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
D models/atm/cam/src/physics/cam/subcol_utils.F90
D models/atm/cam/src/physics/cam/physics_buffer.F90
D models/atm/cam/src/utils/buffer.F90
- These three modules are now autogenerated from the .F90.in versions during the make process
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/test_driver.sh
- goldbach now uses modules to identify library versions
M models/atm/cam/bld/configure
- removed two obsolete directories which no longer exist
M models/atm/cam/bld/Makefile.in
M models/atm/cam/bld/mkSrcfiles
M models/atm/cam/bld/mkDepends
- generation of .F90 from .F90.in files now occurs automatically during the compilation process
M models/atm/cam/src/chemistry/mozart/wei96.F90
- commented out error check which had obsolete PAUSE statement
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all PASS
goldbach/nag: all PASS
goldbach/pgi or jaguar/pgi: all PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_44
Originator(s): cacraig
Date: 07/16/14
One-line Summary: Update externals to cesm1_3_beta11
Purpose of changes:
- Update externals to cesm1_3_beta11
- Conflicting namespace for abortutils in CAM and CLM which both use endrun (with different implementations)
Renamed CAM's abortutils module to cam_abortutils
- Updated to new chem_proc external which includes changes to cam_abortutils
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
- coupler namelist variable "ocean_tight_coupling" replaced with "cpl_seq_option"
'CESM1_MOD_TIGHT' is the setting which mimics turning on the old ocean_tight_coupling
- CLM namelist variable "fpftdyn" renamed to flanduse_timeseries - no other mods
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
D models/atm/cam/src/utils/abortutils.F90
- Removed due to namespace collision with CLM version
List all subroutines added and what they do:
A models/atm/cam/src/utils/cam_abortutils.F90
- old abortutils.F90 module is now named cam_abortutils.F90 - no other change
List all existing files that have been modified, and describe the changes:
M SVN_EXTERNAL_DIRECTORIES
- updated externals to match cesm1_3_beta11
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
- updated to CAM's externals which had changes to point to cam_abortutils
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
M models/atm/cam/bld/namelist_files/namelist_definition.xml
M models/atm/cam/bld/build-namelist
- Contain changes for coupler and CLM nameslist changes
M models/atm/cam/doc/ReleaseNotes
- Contains information for cam5_3_43 release
M models/atm/cam/tools/interpic_new/interpolate_data.F90
M models/atm/cam/tools/interpaerosols/Makefile
M models/atm/cam/bld/Makefile.in
M models/atm/cam/src/control/runtime_opts.F90
M models/atm/cam/src/control/ncdio_atm.F90
M models/atm/cam/src/control/bnddyi.F90
M models/atm/cam/src/control/sat_hist.F90
M models/atm/cam/src/control/interpolate_data.F90
M models/atm/cam/src/control/wrap_nf.F90
M models/atm/cam/src/control/filenames.F90
M models/atm/cam/src/control/wrap_mpi.F90
M models/atm/cam/src/control/error_messages.F90
M models/atm/cam/src/control/readinitial.F90
M models/atm/cam/src/control/physconst.F90
M models/atm/cam/src/control/cam_history_buffers.F90
M models/atm/cam/src/control/intp_util.F90
M models/atm/cam/src/control/camsrfexch.F90
M models/atm/cam/src/control/units.F90
M models/atm/cam/src/control/gauaw_mod.F90
M models/atm/cam/src/control/cam_restart.F90
M models/atm/cam/src/control/ioFileMod.F90
M models/atm/cam/src/control/vrtmap.F90
M models/atm/cam/src/control/cam_history_support.F90
M models/atm/cam/src/control/cam_comp.F90
M models/atm/cam/src/control/cam_history.F90
M models/atm/cam/src/physics/cam/subcol.F90
M models/atm/cam/src/physics/cam/carma_flags_mod.F90
M models/atm/cam/src/physics/cam/microp_aero.F90
M models/atm/cam/src/physics/cam/tropopause.F90
M models/atm/cam/src/physics/cam/ndrop_bam.F90
M models/atm/cam/src/physics/cam/micro_mg_cam.F90
M models/atm/cam/src/physics/cam/boundarydata.F90
M models/atm/cam/src/physics/cam/aer_rad_props.F90
M models/atm/cam/src/physics/cam/radsw.F90
M models/atm/cam/src/physics/cam/phys_prop.F90
M models/atm/cam/src/physics/cam/check_energy.F90
M models/atm/cam/src/physics/cam/restart_physics.F90
M models/atm/cam/src/physics/cam/chem_surfvals.F90
M models/atm/cam/src/physics/cam/constituents.F90
M models/atm/cam/src/physics/cam/radconstants.F90
M models/atm/cam/src/physics/cam/subcol_utils.F90.in
M models/atm/cam/src/physics/cam/ndrop.F90
M models/atm/cam/src/physics/cam/clubb_intr.F90
M models/atm/cam/src/physics/cam/rad_constituents.F90
M models/atm/cam/src/physics/cam/convect_shallow.F90
M models/atm/cam/src/physics/cam/cloud_fraction.F90
M models/atm/cam/src/physics/cam/pkg_cld_sediment.F90
M models/atm/cam/src/physics/cam/dadadj.F90
M models/atm/cam/src/physics/cam/gw_drag.F90
M models/atm/cam/src/physics/cam/phys_grid.F90
M models/atm/cam/src/physics/cam/zm_conv.F90
M models/atm/cam/src/physics/cam/unicon_cam.F90
M models/atm/cam/src/physics/cam/eddy_diff.F90
M models/atm/cam/src/physics/cam/majorsp_diffusion.F90
M models/atm/cam/src/physics/cam/cam3_aero_data.F90
M models/atm/cam/src/physics/cam/comsrf.F90
M models/atm/cam/src/physics/cam/tphysidl.F90
M models/atm/cam/src/physics/cam/radiation.F90
M models/atm/cam/src/physics/cam/tracers_suite.F90
M models/atm/cam/src/physics/cam/carma_model_flags_mod.F90
M models/atm/cam/src/physics/cam/phys_control.F90
M models/atm/cam/src/physics/cam/hk_conv.F90
M models/atm/cam/src/physics/cam/cam_diagnostics.F90
M models/atm/cam/src/physics/cam/stratiform.F90
M models/atm/cam/src/physics/cam/ghg_data.F90
M models/atm/cam/src/physics/cam/radiation_data.F90
M models/atm/cam/src/physics/cam/modal_aer_opt.F90
M models/atm/cam/src/physics/cam/physics_types.F90
M models/atm/cam/src/physics/cam/subcol_utils.F90
M models/atm/cam/src/physics/cam/radlw.F90
M models/atm/cam/src/physics/cam/co2_data_flux.F90
M models/atm/cam/src/physics/cam/radae.F90
M models/atm/cam/src/physics/cam/physpkg.F90
M models/atm/cam/src/physics/cam/phys_debug_util.F90
M models/atm/cam/src/physics/cam/cam3_ozone_data.F90
M models/atm/cam/src/physics/cam/macrop_driver.F90
M models/atm/cam/src/physics/cam/cldwat2m_macro.F90
M models/atm/cam/src/physics/cam/convect_deep.F90
M models/atm/cam/src/physics/cam/cospsimulator_intr.F90
M models/atm/cam/src/physics/cam/co2_cycle.F90
M models/atm/cam/src/physics/cam/microp_driver.F90
M models/atm/cam/src/physics/cam/wv_saturation.F90
M models/atm/cam/src/physics/cam/physics_buffer.F90
M models/atm/cam/src/physics/cam/rad_solar_var.F90
M models/atm/cam/src/physics/cam/physics_buffer.F90.in
M models/atm/cam/src/physics/cam/aoa_tracers.F90
M models/atm/cam/src/physics/cam/subcol_tstcp.F90
M models/atm/cam/src/physics/cam/ref_pres.F90
M models/atm/cam/src/physics/cam/conv_water.F90
M models/atm/cam/src/physics/cam/ionosphere.F90
M models/atm/cam/src/physics/cam/cldwat.F90
M models/atm/cam/src/physics/cam/uwshcu.F90
M models/atm/cam/src/physics/cam/vertical_diffusion.F90
M models/atm/cam/src/physics/waccmx/ionosphere.F90
M models/atm/cam/src/physics/carma/cam/carma_getH2SO4.F90
M models/atm/cam/src/physics/carma/cam/carma_getT.F90
M models/atm/cam/src/physics/carma/cam/carma_getH2O.F90
M models/atm/cam/src/physics/carma/cam/carma_intr.F90
M models/atm/cam/src/physics/carma/models/test_tracers2/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/test_tracers2/carma_model_flags_mod.F90
M models/atm/cam/src/physics/carma/models/test_detrain/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/tholin/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/tholin/carma_model_flags_mod.F90
M models/atm/cam/src/physics/carma/models/bc_strat/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/sulfate/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/cirrus_dust/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/cirrus_dust/carma_model_flags_mod.F90
M models/atm/cam/src/physics/carma/models/cirrus_dust/carma_cloudfraction.F90
M models/atm/cam/src/physics/carma/models/cirrus/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/cirrus/carma_model_flags_mod.F90
M models/atm/cam/src/physics/carma/models/cirrus/carma_cloudfraction.F90
M models/atm/cam/src/physics/carma/models/test_radiative/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/test_growth/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/test_passive/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/pmc/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/pmc/carma_model_flags_mod.F90
M models/atm/cam/src/physics/carma/models/dust/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/dust/carma_model_flags_mod.F90
M models/atm/cam/src/physics/carma/models/test_swelling/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/meteor_impact/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/meteor_impact/carma_model_flags_mod.F90
M models/atm/cam/src/physics/carma/models/mixed_sulfate/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/mixed_sulfate/carma_model_flags_mod.F90
M models/atm/cam/src/physics/carma/models/meteor_smoke/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/meteor_smoke/carma_model_flags_mod.F90
M models/atm/cam/src/physics/carma/models/sea_salt/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/sea_salt/carma_model_flags_mod.F90
M models/atm/cam/src/physics/carma/models/pmc_sulfate/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/pmc_sulfate/carma_model_flags_mod.F90
M models/atm/cam/src/physics/carma/models/test_tracers/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/test_tracers/carma_model_flags_mod.F90
M models/atm/cam/src/physics/waccm/nlte_fomichev.F90
M models/atm/cam/src/physics/waccm/iondrag.F90
M models/atm/cam/src/physics/waccm/nlte_lw.F90
M models/atm/cam/src/physics/waccm/radheat.F90
M models/atm/cam/src/physics/waccm/waccm_forcing.F90
M models/atm/cam/src/physics/waccm/qbo.F90
M models/atm/cam/src/physics/rrtmg/ebert_curry.F90
M models/atm/cam/src/physics/rrtmg/slingo.F90
M models/atm/cam/src/physics/rrtmg/radconstants.F90
M models/atm/cam/src/physics/rrtmg/oldcloud.F90
M models/atm/cam/src/physics/rrtmg/radlw.F90
M models/atm/cam/src/physics/rrtmg/radiation.F90
M models/atm/cam/src/physics/rrtmg/rad_solar_var.F90
M models/atm/cam/src/physics/rrtmg/cloud_rad_props.F90
M models/atm/cam/src/physics/rrtmg/radsw.F90
M models/atm/cam/src/utils/time_manager.F90
M models/atm/cam/src/utils/sgexx.F90
M models/atm/cam/src/utils/buffer.F90
M models/atm/cam/src/utils/buffer.F90.in
M models/atm/cam/src/utils/spmd_utils.F90
M models/atm/cam/src/utils/cam_pio_utils.F90
M models/atm/cam/src/utils/cam_dom/sst_data.F90
M models/atm/cam/src/utils/cam_dom/ocn_time_manager.F90
M models/atm/cam/src/utils/cam_dom/ocn_comp.F90
M models/atm/cam/src/utils/cam_aqua/ocn_comp.F90
M models/atm/cam/src/utils/hycoef.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_sulfur/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_sulfur/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_sim_dat.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_rename.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_data.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_coag.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_gasaerexch.F90
M models/atm/cam/src/chemistry/modal_aero/dust_model.F90
M models/atm/cam/src/chemistry/modal_aero/sox_cldaero_mod.F90
M models/atm/cam/src/chemistry/modal_aero/aero_model.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_initialize_data.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_newnuc.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_sim_dat.F90
M models/atm/cam/src/chemistry/bulk_aero/aero_model.F90
M models/atm/cam/src/chemistry/bulk_aero/mo_setsoa.F90
M models/atm/cam/src/chemistry/bulk_aero/dust_model.F90
M models/atm/cam/src/chemistry/bulk_aero/sox_cldaero_mod.F90
M models/atm/cam/src/chemistry/aerosol/dust_common.F90
M models/atm/cam/src/chemistry/aerosol/dust_sediment_mod.F90
M models/atm/cam/src/chemistry/aerosol/soil_erod_mod.F90
M models/atm/cam/src/chemistry/pp_trop_ghg/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_ghg/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_waccm_ghg/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_waccm_ghg/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_none/mo_imp_sol.F90
M models/atm/cam/src/chemistry/utils/aerodep_flx.F90
M models/atm/cam/src/chemistry/utils/prescribed_ozone.F90
M models/atm/cam/src/chemistry/utils/mo_msis_ubc.F90
M models/atm/cam/src/chemistry/utils/modal_aero_deposition.F90
M models/atm/cam/src/chemistry/utils/prescribed_ghg.F90
M models/atm/cam/src/chemistry/utils/tracer_data.F90
M models/atm/cam/src/chemistry/utils/prescribed_volcaero.F90
M models/atm/cam/src/chemistry/utils/modal_aero_calcsize.F90
M models/atm/cam/src/chemistry/utils/prescribed_aero.F90
M models/atm/cam/src/chemistry/utils/horizontal_interpolate.F90
M models/atm/cam/src/chemistry/utils/mo_flbc.F90
M models/atm/cam/src/chemistry/utils/prescribed_strataero.F90
M models/atm/cam/src/chemistry/utils/aircraft_emit.F90
M models/atm/cam/src/chemistry/utils/solar_data.F90
M models/atm/cam/src/chemistry/utils/mo_solar_parms.F90
M models/atm/cam/src/chemistry/utils/time_utils.F90
M models/atm/cam/src/chemistry/utils/modal_aero_wateruptake.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_mam4/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_mam4/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_bam/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_bam/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_sim_dat.F90
M models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
M models/atm/cam/src/chemistry/mozart/mo_tgcm_ubc.F90
M models/atm/cam/src/chemistry/mozart/mo_lightning.F90
M models/atm/cam/src/chemistry/mozart/mo_seto2.F90
M models/atm/cam/src/chemistry/mozart/lin_strat_chem.F90
M models/atm/cam/src/chemistry/mozart/wei96.F90
M models/atm/cam/src/chemistry/mozart/mo_jshort.F90
M models/atm/cam/src/chemistry/mozart/mo_cph.F90
M models/atm/cam/src/chemistry/mozart/mo_chemini.F90
M models/atm/cam/src/chemistry/mozart/tracer_srcs.F90
M models/atm/cam/src/chemistry/mozart/mo_airplane.F90
M models/atm/cam/src/chemistry/mozart/gas_wetdep_opts.F90
M models/atm/cam/src/chemistry/mozart/mo_heatnirco2.F90
M models/atm/cam/src/chemistry/mozart/chemistry.F90
M models/atm/cam/src/chemistry/mozart/linoz_data.F90
M models/atm/cam/src/chemistry/mozart/mo_sethet.F90
M models/atm/cam/src/chemistry/mozart/apex_subs.F90
M models/atm/cam/src/chemistry/mozart/mo_usrrxt.F90
M models/atm/cam/src/chemistry/mozart/mo_jeuv.F90
M models/atm/cam/src/chemistry/mozart/spehox.F90
M models/atm/cam/src/chemistry/mozart/mo_synoz.F90
M models/atm/cam/src/chemistry/mozart/chlorine_loading_data.F90
M models/atm/cam/src/chemistry/mozart/mo_ghg_chem.F90
M models/atm/cam/src/chemistry/mozart/mo_mean_mass.F90
M models/atm/cam/src/chemistry/mozart/mo_airglow.F90
M models/atm/cam/src/chemistry/mozart/mo_jlong.F90
M models/atm/cam/src/chemistry/mozart/mo_photoin.F90
M models/atm/cam/src/chemistry/mozart/spedata.F90
M models/atm/cam/src/chemistry/mozart/noy_ubc.F90
M models/atm/cam/src/chemistry/mozart/upper_bc.F90
M models/atm/cam/src/chemistry/mozart/rate_diags.F90
M models/atm/cam/src/chemistry/mozart/mo_sulf.F90
M models/atm/cam/src/chemistry/mozart/tracer_cnst.F90
M models/atm/cam/src/chemistry/mozart/mo_setz.F90
M models/atm/cam/src/chemistry/mozart/mo_extfrc.F90
M models/atm/cam/src/chemistry/mozart/mo_apex.F90
M models/atm/cam/src/chemistry/mozart/mo_fstrat.F90
M models/atm/cam/src/chemistry/mozart/mo_snoe.F90
M models/atm/cam/src/chemistry/mozart/mo_sad.F90
M models/atm/cam/src/chemistry/mozart/euvac.F90
M models/atm/cam/src/chemistry/mozart/mo_drydep.F90
M models/atm/cam/src/chemistry/mozart/mo_photo.F90
M models/atm/cam/src/chemistry/mozart/cfc11star.F90
M models/atm/cam/src/chemistry/mozart/efield.F90
M models/atm/cam/src/chemistry/mozart/mo_neu_wetdep.F90
M models/atm/cam/src/chemistry/mozart/mo_tuv_inti.F90
M models/atm/cam/src/chemistry/mozart/mo_xsections.F90
M models/atm/cam/src/chemistry/mozart/mo_inter.F90
M models/atm/cam/src/chemistry/mozart/gcr_ionization.F90
M models/atm/cam/src/chemistry/mozart/mo_srf_emissions.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_waccm_tsmlt/mo_sim_dat.F90
M models/atm/cam/src/dynamics/sld/spmd_dyn.F90
M models/atm/cam/src/dynamics/sld/sld_control_mod.F90
M models/atm/cam/src/dynamics/sld/trunc.F90
M models/atm/cam/src/dynamics/sld/restart_dynamics.F90
M models/atm/cam/src/dynamics/sld/sltint.F90
M models/atm/cam/src/dynamics/sld/dyn_grid.F90
M models/atm/cam/src/dynamics/sld/interp_mod.F90
M models/atm/cam/src/dynamics/sld/initcom.F90
M models/atm/cam/src/dynamics/sld/dp_coupling.F90
M models/atm/cam/src/dynamics/sld/scanslt.F90
M models/atm/cam/src/dynamics/sld/inidat.F90
M models/atm/cam/src/dynamics/sld/nmmatrix.F90
M models/atm/cam/src/dynamics/eul/spmd_dyn.F90
M models/atm/cam/src/dynamics/eul/iop.F90
M models/atm/cam/src/dynamics/eul/cubydr.F90
M models/atm/cam/src/dynamics/eul/trunc.F90
M models/atm/cam/src/dynamics/eul/herxin.F90
M models/atm/cam/src/dynamics/eul/restart_dynamics.F90
M models/atm/cam/src/dynamics/eul/sphdep.F90
M models/atm/cam/src/dynamics/eul/lagyin.F90
M models/atm/cam/src/dynamics/eul/eul_control_mod.F90
M models/atm/cam/src/dynamics/eul/dyn_grid.F90
M models/atm/cam/src/dynamics/eul/interp_mod.F90
M models/atm/cam/src/dynamics/eul/initcom.F90
M models/atm/cam/src/dynamics/eul/dp_coupling.F90
M models/atm/cam/src/dynamics/eul/settau.F90
M models/atm/cam/src/dynamics/eul/scanslt.F90
M models/atm/cam/src/dynamics/eul/inidat.F90
M models/atm/cam/src/dynamics/eul/getinterpnetcdfdata.F90
M models/atm/cam/src/dynamics/eul/limdx.F90
M models/atm/cam/src/dynamics/se/dyn_grid.F90
M models/atm/cam/src/dynamics/se/interp_mod.F90
M models/atm/cam/src/dynamics/se/native_mapping.F90
M models/atm/cam/src/dynamics/se/dp_coupling.F90
M models/atm/cam/src/dynamics/se/inidat.F90
M models/atm/cam/src/dynamics/se/spmd_dyn.F90
M models/atm/cam/src/dynamics/se/nctopo_util_mod.F90
M models/atm/cam/src/dynamics/se/stepon.F90
M models/atm/cam/src/dynamics/se/restart_dynamics.F90
M models/atm/cam/src/dynamics/fv/dyn_comp.F90
M models/atm/cam/src/dynamics/fv/mapz_module.F90
M models/atm/cam/src/dynamics/fv/spmd_dyn.F90
M models/atm/cam/src/dynamics/fv/ctem.F90
M models/atm/cam/src/dynamics/fv/te_map.F90
M models/atm/cam/src/dynamics/fv/fv_control_mod.F90
M models/atm/cam/src/dynamics/fv/restart_dynamics.F90
M models/atm/cam/src/dynamics/fv/metdata.F90
M models/atm/cam/src/dynamics/fv/pfixer.F90
M models/atm/cam/src/dynamics/fv/dyn_grid.F90
M models/atm/cam/src/dynamics/fv/interp_mod.F90
M models/atm/cam/src/dynamics/fv/initcom.F90
M models/atm/cam/src/dynamics/fv/dp_coupling.F90
M models/atm/cam/src/dynamics/fv/inidat.F90
M models/atm/cam/src/dynamics/fv/cd_core.F90
M models/atm/cam/src/dynamics/fv/stepon.F90
M models/atm/cam/src/advection/slt/qmassa.F90
M models/atm/cam/src/unit_drivers/drv_input_data.F90
M models/atm/cam/src/unit_drivers/rad/unit_driver.F90
M models/atm/cam/src/unit_drivers/offline_driver.F90
M models/atm/cam/src/cpl/atm_comp_mct.F90
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all PASS
goldbach/nag: all PASS
goldbach/pgi or jaguar/pgi: all PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_43
Originator(s): fvitt
Date: 11 July 2014
One-line Summary: Include MAM4 aerosol model
Purpose of changes:
Place 4-mode version of the modal aerosol model (MAM4) on the development trunk.
Bugs fixed (include bugzilla ID): 1994
- ocphiwet deposition missing contributions from soa_a2 and soa_c2
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton, santos
List all subroutines eliminated:
List all subroutines added and what they do:
A models/atm/cam/src/chemistry/pp_trop_mam4
A models/atm/cam/src/chemistry/pp_trop_mam4/mo_lu_solve.F90
A models/atm/cam/src/chemistry/pp_trop_mam4/mo_setrxt.F90
A models/atm/cam/src/chemistry/pp_trop_mam4/mo_adjrxt.F90
A models/atm/cam/src/chemistry/pp_trop_mam4/mo_rxt_rates_conv.F90
A models/atm/cam/src/chemistry/pp_trop_mam4/mo_imp_sol.F90
A models/atm/cam/src/chemistry/pp_trop_mam4/chem_mods.F90
A models/atm/cam/src/chemistry/pp_trop_mam4/mo_prod_loss.F90
A models/atm/cam/src/chemistry/pp_trop_mam4/mo_lin_matrix.F90
A models/atm/cam/src/chemistry/pp_trop_mam4/m_rxt_id.F90
A models/atm/cam/src/chemistry/pp_trop_mam4/mo_phtadj.F90
A models/atm/cam/src/chemistry/pp_trop_mam4/mo_nln_matrix.F90
A models/atm/cam/src/chemistry/pp_trop_mam4/mo_lu_factor.F90
A models/atm/cam/src/chemistry/pp_trop_mam4/mo_sim_dat.F90
A models/atm/cam/src/chemistry/pp_trop_mam4/mo_indprd.F90
A models/atm/cam/src/chemistry/pp_trop_mam4/m_spc_id.F90
A models/atm/cam/src/chemistry/pp_trop_mam4/chem_mech.doc
A models/atm/cam/src/chemistry/pp_trop_mam4/chem_mech.in
- chemistry preprocessor generated source code for MAM4 chemistry package
A models/atm/cam/test/system/config_files/f4mode1.9c5dh
A models/atm/cam/test/system/config_files/f4mode4c5dm
- regressions tests added for MAM4
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/namelist_files/use_cases/sd_cam4_trop_strat_soa.xml
- fixed issues for the FSDSSOA compset (problems introduced in tag cam5_3_41)
M models/atm/cam/src/chemistry/utils/modal_aero_deposition.F90
- included contributions from MAM4 aerosols and added soa_a2 to the ocphiwet fluxes
M models/atm/cam/bld/configure
M models/atm/cam/bld/config_files/definition.xml
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
M models/atm/cam/bld/namelist_files/master_aer_drydep_list.xml
M models/atm/cam/bld/namelist_files/master_aer_wetdep_list.xml
M models/atm/cam/bld/build-namelist
M models/atm/cam/src/physics/cam/phys_control.F90
M models/atm/cam/src/physics/cam/modal_aer_opt.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_coag.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_data.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_gasaerexch.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_initialize_data.F90
M models/atm/cam/src/chemistry/modal_aero/seasalt_model.F90
M models/atm/cam/src/chemistry/modal_aero/dust_model.F90
M models/atm/cam/src/chemistry/modal_aero/sox_cldaero_mod.F90
- mods for MAM4
M models/atm/cam/test/system/tests_pretag_yellowstone
M models/atm/cam/test/system/tests_pretag_goldbach_nag
M models/atm/cam/test/system/tests_pretag_goldbach_pgi
M models/atm/cam/test/system/input_tests_master
- regressions tests added for MAM4
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
065 bl421 TBL.sh f1.9c5mam4dh outfrq3s 9s .....................................................FAIL! rc= 5 at Thu Jul 10 13:25:06 MDT 2014
goldbach/nag:
049 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 5 at Thu Jul 10 12:17:47 MDT 2014
goldbach/pgi:
056 bl420 TBL.sh f4c5mam7dm outfrq3s 9s .......................................................FAIL! rc= 7 at Thu Jul 10 11:42:00 MDT 2014
059 bl422 TBL.sh f4c5mam4dm outfrq3s 9s .......................................................FAIL! rc= 5 at Thu Jul 10 11:55:20 MDT 2014
** Answer changes for MAM7 configurations are due to change in n_so4_monolayers_pcage parameter in modal_aero_gasaerech
and change in default primary organic mater aerosol optics input file (ocpho_rrtmg)
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_42
Originator(s): jedwards, cacraig
Date: 07/01/2014
One-line Summary: Fix homme compilation issues and add homme NAG test
Purpose of changes:
- Fix homme compilation issues
- add homme NAG test
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/tests_pretag_goldbach_nag
- add homme NAG regression test
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
- Use homme branch that Jim provided which fixes compilation issues
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: Just ran homme tests, 731, 735, 740 and 751 all PASS and are BFB
goldbach/nag: Ran newly introduced 734 test and it PASSes. Could not run baseline due to compilation bug in cam5_3_41 homme.
goldbach/pgi or jaguar/pgi: Only ran 734 tests and all PASS and BFB.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_41
Originator(s): fvitt, dkin, tilmes, lamar
Date: 25 June 2014
One-line Summary: CCMI chemistry and bug fixes
Purpose of changes:
Bring in changes used in the Chemistry-Climate Model Initiative (CCMI)
simulations. These changes are summarized as follows:
- new chemistry package:
waccm_tsmlt
- CCMI diagnostic tracers added to:
trop_mozart_soa
trop_strat_soa
- chemistry mechanism updates to:
waccm_mozart
- updates to mo_sad (from Doug Kinnison)
- rate updates in usrrxt
- enthalpies for exothermic reaction chemicals are now specified
in mechanism file (for WACCM)
- new specified stratospheric aerosols (volcanic)
- give user ability to specify a different rhmin poleward of 60 deg
and above a specified pressure level (300 mbar default)
- CAM-Chem now has ability to include NOx and HOx sources from solar protons
- ability to include NOx and HOx sources from galactic cosmic ray ionization
- background ionization from star light added to WACCM
- corrections to age-of-air tracers (from Valery)
- specified dynamics namelist option met_max_rlx replaced by met_rlx_time (hours)
- history changes:
. increased number of fields in fincls from 750 to 1000
. can have upto 10 simultaneous history files (or streams)
- corrections to aerosol surface area used in usrrxt
- neu_wetdep changes
. set TICE to 263
. disable wet deposition poleward of 60 degrees and pressures < 200 mbar
- chemistry preprocessor updates:
-- enthalpies for chemical potential heating now specified in mechanism files
-- added ability to put comments at the end of reactions in mechanism file following '#' or '!'
-- bug fixes for species names longer than 8 characters (up to 16 characters)
-- mo_imp_sol changes
. compute o3 loss rates for middle atm chem
. changes to appease NAG compiler (lines cannot be longer than 128 chars)
. added 'H_PEROX_CHMP' diagnostic
. changes needed for changes in reaction tag names convention
. added o3s_loss optional argument -- o3s loss below tropopause is implimented in chemdr
Fix bugs:
- remove duplicate line "g1=g1sqrt*g1" in ndrop
- correction to Henry's Law parameters used for SO2 deposition
Regression testing capability added for CGD leehill machine (cacraig)
Regression tests configuration files renamed to follow naming convention:
(dycore)(sd?)(res)(phys)(chem)(debug?)(mpi/openmp/hybrid)
Bugs fixed (include bugzilla ID):
Bug 1983:
- remove duplicate line "g1=g1sqrt*g1" in maxsat subroutine
models/atm/cam/src/physics/cam/ndrop.F90
Bug 2000:
- correction to Henry's Law parameters used for SO2 deposition
models/drv/shr/seq_drydep_mod.F90
models/atm/cam/src/chemistry/mozart/mo_neu_wetdep.F90
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton, santos
List all subroutines eliminated:
D models/atm/cam/src/chemistry/mozart/mo_strato_sad.F90
- removed -- prescribed_strataero provides SADs
D models/atm/cam/test/system/config_files/fst7mode1.9c5h
D models/atm/cam/test/system/config_files/fsoa1.9c4dh
D models/atm/cam/test/system/config_files/wx1.9c4dh
D models/atm/cam/test/system/config_files/fmz4c5dm
D models/atm/cam/test/system/config_files/fm1.9c5dh
D models/atm/cam/test/system/config_files/fs1.9c4dh
D models/atm/cam/test/system/config_files/fsf1.9c5dh
D models/atm/cam/test/system/config_files/wm1.9c4h
D models/atm/cam/test/system/config_files/wg1.9c4dm
D models/atm/cam/test/system/config_files/wm1.9c5dh
D models/atm/cam/test/system/config_files/wm1.9c4m
D models/atm/cam/test/system/config_files/fst1.9c4dh
D models/atm/cam/test/system/config_files/fm1.9c4m
D models/atm/cam/test/system/config_files/fma1.9c4m
D models/atm/cam/test/system/config_files/fprgspc4c4dm
D models/atm/cam/test/system/config_files/wm4c4h
D models/atm/cam/test/system/config_files/wg4c4dm
D models/atm/cam/test/system/config_files/fsf4c4dm
D models/atm/cam/test/system/config_files/fmo1.9c4dh
D models/atm/cam/test/system/config_files/f7mode1.9c5dh
D models/atm/cam/test/system/config_files/wg10c4dm
D models/atm/cam/test/system/config_files/fst4c5dm
D models/atm/cam/test/system/config_files/fm1.9c4dh
D models/atm/cam/test/system/config_files/fsf1.9c4dh
D models/atm/cam/test/system/config_files/fn10c5dm
D models/atm/cam/test/system/config_files/f7mode4c5dm
D models/atm/cam/test/system/config_files/fsm4c5dm
D models/atm/cam/test/system/config_files/fsoa4c4dm
D models/atm/cam/test/system/config_files/fst7mode4c5dm
D models/atm/cam/test/system/config_files/wx1.9c4h
D models/atm/cam/test/system/config_files/fst1.9c5dh
D models/atm/cam/test/system/config_files/fmgpa1.9c4dm
D models/atm/cam/test/system/config_files/fn1.9c5dh
- renamed following convention :
(dycore)(sd?)(res)(phys)(chem)(debug?)(mpi/openmp/hybrid)
new file names are below
List all subroutines added and what they do:
A models/atm/cam/src/chemistry/mozart/gcr_ionization.F90
- new GCR sources of NOx and HOx
A models/atm/cam/src/chemistry/mozart/noy_ubc.F90
- new module for CAM-Chem NOy upper boundary intended for NOy from solar proton
ionization sources above model top
A models/atm/cam/src/chemistry/mozart/photo_bkgrnd.F90
- new module for WACCM to prameterize background ionization by star light
A models/atm/cam/src/chemistry/utils/prescribed_strataero.F90
- new module for prescribed stratospheric aerosols (volcanic)
A models/atm/cam/src/chemistry/pp_waccm_tsmlt/mo_lu_solve.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt/mo_setrxt.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt/mo_adjrxt.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt/mo_rxt_rates_conv.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt/mo_imp_sol.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt/chem_mods.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt/mo_prod_loss.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt/mo_lin_matrix.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt/m_rxt_id.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt/mo_phtadj.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt/mo_nln_matrix.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt/mo_lu_factor.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt/mo_sim_dat.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt/mo_indprd.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt/m_spc_id.F90
A models/atm/cam/src/chemistry/pp_waccm_tsmlt/chem_mech.doc
A models/atm/cam/src/chemistry/pp_waccm_tsmlt/chem_mech.in
A models/atm/cam/src/chemistry/pp_waccm_tsmlt
- new chem pkg
A models/atm/cam/bld/namelist_files/use_cases/2000_cam4_trop_moz_soa.xml
A models/atm/cam/bld/namelist_files/use_cases/2004-2100_ccmi_refc2_rcp45_trop_strat_soa.xml
A models/atm/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_waccm_tsmlt.xml
A models/atm/cam/bld/namelist_files/use_cases/sd_1975-2010_ccmi_refc1_trop_strat_soa.xml
A models/atm/cam/bld/namelist_files/use_cases/1950-2010_ccmi_refc1_waccm_ma.xml
A models/atm/cam/bld/namelist_files/use_cases/sd_1975-2010_ccmi_refc1_waccm_tsmlt.xml
A models/atm/cam/bld/namelist_files/use_cases/sd_2007-2011_htap_trop_strat_soa.xml
A models/atm/cam/bld/namelist_files/use_cases/2000_ccmi_refc1_waccm_tsmlt.xml
A models/atm/cam/bld/namelist_files/use_cases/1979-2010_sd_trop_strat_soa.xml
A models/atm/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_rcp85_waccm_ma.xml
A models/atm/cam/bld/namelist_files/use_cases/sd_waccm_tsmlt_geos5.xml
A models/atm/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_waccm_ma.xml
A models/atm/cam/bld/namelist_files/use_cases/1950-2010_ccmi_refc1_waccm_tsmlt.xml
A models/atm/cam/bld/namelist_files/use_cases/2007-2011_htap_trop_strat_soa.xml
A models/atm/cam/bld/namelist_files/use_cases/2005-2100_cam4_strataero_rcp45.xml
A models/atm/cam/bld/namelist_files/use_cases/1950-2010_ccmi_refc1_trop_strat_soa.xml
A models/atm/cam/bld/namelist_files/use_cases/sd_1975-2010_ccmi_refc1_waccm_ma.xml
A models/atm/cam/bld/namelist_files/use_cases/1950-2100_ccmi_refc2_trop_strat_soa.xml
- new use cases for CCMI simulations
A models/atm/cam/test/system/tests_pretag_leehill
- added for CGD leehill machine (cacraig)
A models/atm/cam/test/system/nl_files/outfrq3s_neu
A models/atm/cam/test/system/nl_files/outfrq3s_sd
- regression tests namelist files added
A models/atm/cam/test/system/tests_ccmi_hybrid
A models/atm/cam/test/system/tests_ccmi_mpi
- new tests for the CCMI configurations
A models/atm/cam/test/system/config_files/f10c4wtsmltdm
A models/atm/cam/test/system/config_files/f1.9c4wtsmltdh
A models/atm/cam/test/system/config_files/fsd1.9c4wtsmltdh
A models/atm/cam/test/system/config_files/f1.9c4wtsmlth
A models/atm/cam/test/system/config_files/f4c4wtsmltdh
A models/atm/cam/test/system/config_files/fsd1.9c4wtsmlth
A models/atm/cam/test/system/config_files/f1.9c4wtssoadh
A models/atm/cam/test/system/config_files/fsd1.9c4wtssoadh
A models/atm/cam/test/system/config_files/f4c4wtssoadh
A models/atm/cam/test/system/config_files/f4c4wtssoadm
A models/atm/cam/test/system/config_files/fsd1.9c4wtssoah
A models/atm/cam/test/system/config_files/f1.9c4tssoadh
A models/atm/cam/test/system/config_files/f4c4wmdh
A models/atm/cam/test/system/config_files/fsd1.9c4wmdh
A models/atm/cam/test/system/config_files/fsd1.9c4wmh
- new
A models/atm/cam/test/system/config_files/f1.9c4wghgdm
A models/atm/cam/test/system/config_files/f1.9c4mozdh
A models/atm/cam/test/system/config_files/f10c5nonedm
A models/atm/cam/test/system/config_files/f4c4wghgdm
A models/atm/cam/test/system/config_files/f1.9c5mam7dh
A models/atm/cam/test/system/config_files/f10c4wghgdm
A models/atm/cam/test/system/config_files/f1.9c4wmh
A models/atm/cam/test/system/config_files/f1.9c4wmdh
A models/atm/cam/test/system/config_files/f1.9c4wmm
A models/atm/cam/test/system/config_files/f4c5mam7dm
A models/atm/cam/test/system/config_files/f1.9c5tsmam3dh
A models/atm/cam/test/system/config_files/f1.9c5fstmam3dh
A models/atm/cam/test/system/config_files/f4c4soadm
A models/atm/cam/test/system/config_files/f1.9c4bamm
A models/atm/cam/test/system/config_files/f1.9c5mzmam3dh
A models/atm/cam/test/system/config_files/f4c5tsmam3dm
A models/atm/cam/test/system/config_files/f4c5tsmam7dm
A models/atm/cam/test/system/config_files/fsd1.9c4mozdh
A models/atm/cam/test/system/config_files/f1.9c5tsmam7h
A models/atm/cam/test/system/config_files/f4c5mzmam3dm
A models/atm/cam/test/system/config_files/f1.9c4ghgdm
A models/atm/cam/test/system/config_files/f1.9c5wmam3dh
A models/atm/cam/test/system/config_files/f1.9c4soadh
A models/atm/cam/test/system/config_files/f1.9c4fastdh
A models/atm/cam/test/system/config_files/f4c4prgspcdm
A models/atm/cam/test/system/config_files/f1.9c4mozm
A models/atm/cam/test/system/config_files/f1.9c4wmxh
A models/atm/cam/test/system/config_files/f1.9c4wmxdh
A models/atm/cam/test/system/config_files/f4c4wmh
A models/atm/cam/test/system/config_files/f4c4fastdm
A models/atm/cam/test/system/config_files/f1.9c5nonedh
A models/atm/cam/test/system/config_files/f4c5fstmam3dm
- renamed
List all existing files that have been modified, and describe the changes:
M SVN_EXTERNAL_DIRECTORIES
- update models/drv external to drvseq5_0_13 -- has correction to Henry's Law parameters for SO2
in shr/seq_drydep_mod.F90
M models/atm/cam/bld/cam.buildnml.csh
- added warning messages (cacraig)
M models/atm/cam/bld/configure
- added waccm_tsmlt
- changes in chem_nadv for
trop_mozart_soa
trop_strat_soa
waccm_mozart
- remove useless DUST and PROGSSLT cpp defs
M models/atm/cam/bld/config_files/definition.xml
- added waccm_tsmlt
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- simplified/reduced many defaults
- updates in defaults for photolysis, stratospheric aerosols
- defaults for new waccm_tsmlt chem
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- met_max_rlx --> met_rlx_time (hours)
Relaxation time (hours) applied to specified meteorology.
- positive values less then time step size gives 100% nudging
- negative values gives 0.0% nudging (infinite relaxation time)
- expanded history streams from 6 to 10
- increased number of fields in fincls from 750 to 1000
- added namelist options:
rxt_rate_sums -- to specify rate families (or groupings) diagnostics based
on reaction tag names
cldfrc_rhminp -- min rh for high stable clouds poleward of 60 degrees
cldfrc_rhminp_botmb -- max pressure (mbar) where cldfrc_rhminp is applied
noy_ubc_* options for NOy upper boundary conditions for CAM-Chem to include SPE
NOx source obove the model top
gcr_ionization_* options for GCR ionizations for NOx and HOx producation
sulf_* options (serial, cyclic, or fixed) for tropospheric sulfate aerosols
prescribed_strataero_* options for prescribed stratospheric aerosols
- removed
sad_* options
M models/atm/cam/bld/build-namelist
- die if cldfrc_rhminp is set and microphys is not 'rk'
- waccm_tsmlt changes
- set default wetdep method
- rad_climate includes CFC11STAR for trop_strat waccm_mozart waccm_tsmlt
- if strat_aero_feedback use VOLC_MMR rather than H2SO4M in rad_climate
- "sad_file" replaced by "prescribed_strataero_file"
- set cam_chempkg to "waccm_mozart" for waccm_tsmlt as well as waccm_mozart_sulfur
M models/atm/cam/src/chemistry/bulk_aero/aero_model.F90
- surface area corrections
M models/atm/cam/src/chemistry/bulk_aero/mo_setsoa.F90
- rate tags changes
- corrections in soa_yield()
M models/atm/cam/src/chemistry/modal_aero/aero_model.F90
- removed reference to ref_pres module -- not used
M models/atm/cam/src/chemistry/mozart/cfc11star.F90
- expanded for the additional waccm_tsmlt CFC species
M models/atm/cam/src/chemistry/mozart/chemistry.F90
- include GRCs
- no longer fix CO and CO2 at upper boundary (now have zero flux)
- invoke register_cfc11star for CAM-Chem as well as WACCM
- include SPEs for CAM-Chem (and NOy upper boundary)
- sad namelist vars removed (mo_sad now reads its own namelist)
- change find tropopuase method for non-super_fast mechanisms
M models/atm/cam/src/chemistry/mozart/mo_apex.F90
- added module-level initialized flag
M models/atm/cam/src/chemistry/mozart/mo_chemini.F90
- cam-chem can init spe module, which initializes mo_apex if needed
- strato_sad_inti removed (now prescribed_strataero reads SADs)
M models/atm/cam/src/chemistry/mozart/mo_chm_diags.F90
- diagnostics added for waccm_tsmlt
- added DOCHM_LMS and DOCHM_TRP for CCMI simulations
M models/atm/cam/src/chemistry/mozart/mo_cph.F90
- refactored to be more flexable -- enthalpy data are now specified in
chemistry preprocessor input files (cph=nnnn)
M models/atm/cam/src/chemistry/mozart/mo_drydep.F90
- some code cleanup
- added O3S
- unstructured grid fix
M models/atm/cam/src/chemistry/mozart/mo_extfrc.F90
- alt_data columns are now reversed in tracer_data
M models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
- added code for CCMI diagnostic tracers
- NOy uppper boundary
- SAD from phys buffer
M models/atm/cam/src/chemistry/mozart/mo_ghg_chem.F90
- correction in diagnostic GHG_CH4_R
M models/atm/cam/src/chemistry/mozart/mo_neu_wetdep.F90
- set TICE to 263.
- set dtwr to zero poleward of 60 degrees and press < 200 mbar
- added SO2t and NH_50W
- added HF COF2 and COFCL
- removed gamma function -- use csm_share version
- correction to Henry's Law parameters used for SO2 deposition (in seq_drydep_mod)
M models/atm/cam/src/chemistry/mozart/mo_photo.F90
- added background ionization for waccm
M models/atm/cam/src/chemistry/mozart/mo_sad.F90
- science changes from Doug Kinnison
M models/atm/cam/src/chemistry/mozart/mo_setext.F90
- added GCR sources of NOx and HOx
- make SPE sources available to CAM-Chem as well as WACCM
- AOA_NH forcings for CCMI simulations
M models/atm/cam/src/chemistry/mozart/mo_solarproton.F90
- added a check for presence of SPEs in simulation
M models/atm/cam/src/chemistry/mozart/mo_sulf.F90
- options added for cyclical and serial inputs
M models/atm/cam/src/chemistry/mozart/mo_tgcm_ubc.F90
- now only specify H2 at upper boundary (removed CO and CO2)
M models/atm/cam/src/chemistry/mozart/mo_usrrxt.F90
- rate updates for
cl2o2 + m --> 2*clo + m
ho2 + ho2 --> h2o2
- added aerosol losses above tropopause
M models/atm/cam/src/chemistry/mozart/rate_diags.F90
- added functionality to provide rate groupings (sumations)
M models/atm/cam/src/chemistry/mozart/spedata.F90
- make spe_run public/protected
- initialize apex module if needed
M models/atm/cam/src/chemistry/mozart/upper_bc.F90
- added F and HF fixed upper boundaries
M models/atm/cam/src/chemistry/utils/tracer_data.F90
- added geop_alt option
- reverse vertical coordinate for alt_data in interpolated
- pgi 13.9 bug workaround (santos)
M models/atm/cam/src/control/cam_history_support.F90
- ptapes changed from 8 to 12 (to have 10 history streams)
- pflds increased from 750 to 1000 (fincls can have 1000 fields)
M models/atm/cam/src/control/runtime_opts.F90
- invoke _readnl for modules:
prescribed_strataero
rate_diags
- expanded history namelist arrays from 6 to 10
- default for collect_column_output changed from .false. to .true. to avoid memory leak
M models/atm/cam/src/dynamics/fv/metdata.F90
- met_max_rlx replaced by met_rlx_time
M models/atm/cam/src/dynamics/se/interp_mod.F90
- mods to deal with the length of interpolate_analysis is NOT equal to mtapes
(which is now increased to 12)
M models/atm/cam/src/physics/cam/aoa_tracers.F90
- corrections to age-of-air tracers (Valery Yudin)
M models/atm/cam/src/physics/cam/cloud_fraction.F90
- give user ability to specify a different rhmin poleward of 60 deg
and above a specified pressure level (300 mbar default)
M models/atm/cam/src/physics/cam/micro_mg_cam.F90
- use molecular weight of H2O rather than dry air for the water constituents
M models/atm/cam/src/physics/cam/stratiform.F90
- use molecular weight of H2O rather than dry air for the water constituents
M models/atm/cam/src/physics/cam/ndrop.F90
- remove duplicate line "g1=g1sqrt*g1" in maxsat subroutine ( Bug 1983 )
M models/atm/cam/src/physics/cam/physpkg.F90
- invoke prescribed_strataero methods
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
- chem preprocessor updates
-- enthalpies for chemical potential heating now specified in mechaism files
-- added ability to put comments at the end of reactions in mechanism file following '#' or '!'
-- bug fixes for species names longer than 8 characters (up to 16 characters)
-- mo_imp_sol changes
. compute o3 loss rates for middle atm chem
. changes to appease NAG compiler (lines cannot be longer than 128 chars)
. added 'H_PEROX_CHMP' diagnostic
. changes needed for changes in reaction tag names convention
. added o3s_loss optional argument -- o3s loss below tropopause is implimented in chemdr
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_setrxt.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_rxt_rates_conv.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/m_spc_id.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/chem_mech.in
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_sulfur/mo_setrxt.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_sulfur/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_sulfur/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_sulfur/chem_mods.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_sulfur/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_sulfur/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_sulfur/chem_mech.doc
M models/atm/cam/src/chemistry/pp_waccm_mozart_sulfur/chem_mech.in
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_setrxt.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_rxt_rates_conv.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/chem_mods.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_phtadj.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/m_spc_id.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/chem_mech.doc
M models/atm/cam/src/chemistry/pp_waccm_mozart/chem_mech.in
M models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/chem_mods.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_phtadj.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/m_spc_id.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_none/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_none/mo_phtadj.F90
M models/atm/cam/src/chemistry/pp_none/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_none/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_none/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_none/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_none/mo_rxt_rates_conv.F90
M models/atm/cam/src/chemistry/pp_none/m_spc_id.F90
M models/atm/cam/src/chemistry/pp_none/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_none/chem_mods.F90
M models/atm/cam/src/chemistry/pp_none/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_none/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_phtadj.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/m_spc_id.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_setrxt.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/chem_mods.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_phtadj.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/m_spc_id.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/chem_mech.doc
M models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/chem_mech.in
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_setrxt.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_rxt_rates_conv.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/m_spc_id.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_strat_soa/chem_mech.in
M models/atm/cam/src/chemistry/pp_trop_bam/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_trop_bam/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_bam/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_bam/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/mo_phtadj.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/m_spc_id.F90
M models/atm/cam/src/chemistry/pp_trop_ghg/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_trop_ghg/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_trop_ghg/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_ghg/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_ghg/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_trop_ghg/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_ghg/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_trop_ghg/mo_phtadj.F90
M models/atm/cam/src/chemistry/pp_trop_ghg/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_ghg/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_trop_ghg/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_ghg/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_trop_ghg/m_spc_id.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/chem_mods.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_waccm_ghg/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_waccm_ghg/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_waccm_ghg/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_waccm_ghg/chem_mods.F90
M models/atm/cam/src/chemistry/pp_waccm_ghg/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_waccm_ghg/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_waccm_ghg/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_waccm_ghg/mo_phtadj.F90
M models/atm/cam/src/chemistry/pp_waccm_ghg/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_waccm_ghg/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_waccm_ghg/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_waccm_ghg/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_waccm_ghg/m_spc_id.F90
- mechanism updates
M models/atm/cam/bld/namelist_files/use_cases/sd_waccm_geos5.xml
M models/atm/cam/bld/namelist_files/use_cases/sd_waccm_tslt_geos5.xml
M models/atm/cam/bld/namelist_files/use_cases/2000_cam4_trop_strat_soa.xml
M models/atm/cam/bld/namelist_files/use_cases/2000_cam5_trop_strat_mam3.xml
M models/atm/cam/bld/namelist_files/use_cases/2000_cam4_trop_chem.xml
M models/atm/cam/bld/namelist_files/use_cases/2000_cam5_trop_strat_mam7.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp26.xml
M models/atm/cam/bld/namelist_files/use_cases/2000_cam4_super_fast_llnl.xml
M models/atm/cam/bld/namelist_files/use_cases/cam4_trop_strat_chem.xml
M models/atm/cam/bld/namelist_files/use_cases/cam5_trop_strat_chem.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1955-2005_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1974-2005_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1850_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1850_cam5.xml
M models/atm/cam/bld/namelist_files/use_cases/waccmx_1996_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/sd_waccm_sulfur.xml
M models/atm/cam/bld/namelist_files/use_cases/waccmx_2000_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_refb1.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1850-2005_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2000_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2000_cam5.xml
M models/atm/cam/bld/namelist_files/use_cases/cam4_chem_radpsv_geos5.xml
M models/atm/cam/bld/namelist_files/use_cases/1850-2005_cam4_super_fast_llnl.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp45.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp85.xml
M models/atm/cam/bld/namelist_files/use_cases/1850_cam4_super_fast_llnl.xml
M models/atm/cam/bld/namelist_files/use_cases/1850_cam5_super_fast_llnl.xml
- use case changes for SAD inputs and misc cleanup
M models/atm/cam/test/system/CAM_runcmnd.sh
- added CGD leehill machine (cacraig)
M models/atm/cam/test/system/test_driver.sh
- increased status standard output line to 95 characters
- added CGD leehill machine (cacraig)
M models/atm/cam/test/system/input_tests_master
- added new regression tests
- many existing regression test config files renamed
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
011 bl137 TBL.sh e48c5h outfrq24h 2d ..........................................................FAIL! rc= 7 at Tue Jun 24 16:59:13 MDT 2014
- expected failure -- trop_mam3 chem changes, ndrop bug fix
014 bl322 TBL.sh f1.9c5nonedh outfrq3s 9s .....................................................FAIL! rc= 7 at Tue Jun 24 17:05:22 MDT 2014
- no baseline to compare against -- regression test name changed
017 bl326 TBL.sh f1.9c4wghgdm outfrq3s_2005+waccm_sc_2005-2100_cam4_rcp45 9s ..................FAIL! rc= 7 at Tue Jun 24 17:15:52 MDT 2014
- no baseline to compare against -- regression test name changed
019 bl330 TBL.sh f4c5paqdh aqpgro+aquaplanet_cam5 3s ..........................................FAIL! rc= 7 at Tue Jun 24 17:17:23 MDT 2014
- expected failure -- trop_mam3 chem changes, ndrop bug fix
027 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...................................................FAIL! rc= 7 at Tue Jun 24 17:28:44 MDT 2014
- expected failure -- trop_mam3 chem changes, ndrop bug fix
039 bl336 TBL.sh f4c4wmdh outfrq3s+waccm_1850_cam4 9s .........................................FAIL! rc= 7 at Tue Jun 24 17:44:16 MDT 2014
- no baseline to compare against -- regression test name changed
042 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ..........................................................FAIL! rc= 7 at Tue Jun 24 17:51:54 MDT 2014
- expected failure -- trop_mam3 chem changes, ndrop bug fix
046 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ................................................FAIL! rc= 7 at Tue Jun 24 18:06:50 MDT 2014
- expected failure -- trop_mam3 chem changes, ndrop bug fix
049 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .........................................................FAIL! rc= 7 at Tue Jun 24 18:15:14 MDT 2014
- expected failure -- trop_mam3 chem changes, ndrop bug fix
054 bl373 TBL.sh f1.9c4mozdh outfrq3s 9s ......................................................FAIL! rc= 7 at Tue Jun 24 18:35:31 MDT 2014
- no baseline to compare against -- regression test name changed
056 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...............................................FAIL! rc= 7 at Tue Jun 24 18:38:10 MDT 2014
- expected failure -- trop_mam3 chem changes, ndrop bug fix
058 bl376 TBL.sh f1.9c4mozdh+testmech outfrq3s 9s .............................................FAIL! rc= 7 at Tue Jun 24 18:44:19 MDT 2014
- no baseline to compare against -- regression test name changed
061 bl374 TBL.sh f1.9c4fastdh outfrq3s+1850-2005_cam4_super_fast_llnl 9s ......................FAIL! rc= 7 at Tue Jun 24 18:52:00 MDT 2014
- no baseline to compare against -- regression test name changed
063 bl430 TBL.sh f1.9c5mzmam3dh outfrq3s_NEUwetdep 9s .........................................FAIL! rc= 7 at Tue Jun 24 19:00:57 MDT 2014
- no baseline to compare against -- regression test name changed
066 bl440 TBL.sh f1.9c4soadh outfrq3s+soa_chem_megan_emis 9s ..................................FAIL! rc= 7 at Tue Jun 24 19:17:54 MDT 2014
- no baseline to compare against -- regression test name changed
076 bl379 TBL.sh f1.9c4wmh outfrq24h+waccm_1850-2005_cam4 2d ..................................FAIL! rc= 7 at Tue Jun 24 19:43:52 MDT 2014
- no baseline to compare against -- regression test name changed
082 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ........................................FAIL! rc= 7 at Tue Jun 24 19:58:22 MDT 2014
- expected failure -- trop_mam3 chem changes, ndrop bug fix
084 bl740 TBL.sh h30c5h outfrq3s 9s ...........................................................FAIL! rc= 7 at Tue Jun 24 20:01:07 MDT 2014
- expected failure -- trop_mam3 chem changes, ndrop bug fix
094 bl010 TBL.sh f4c5portdh outfrq24h_port 2d .................................................FAIL! rc= 7 at Tue Jun 24 22:15:49 MDT 2014
- change cfc11 to cfc11star in rad_climate specification (trop_start_mam3 chem configuration)
goldbach/nag:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s ...........................................................FAIL! rc= 7 at Tue Jun 24 15:05:51 MDT 2014
- expected failure -- trop_mam3 chem changes, ndrop bug fix
010 bl112 TBL.sh e8c5dm ghgrmp 9s .............................................................FAIL! rc= 7 at Tue Jun 24 15:15:26 MDT 2014
- expected failure -- trop_mam3 chem changes, ndrop bug fix
014 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Tue Jun 24 15:25:00 MDT 2014
- expected failure -- trop_mam3 chem changes, ndrop bug fix
018 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Tue Jun 24 15:38:11 MDT 2014
- expected failure -- trop_mam3 chem changes, ndrop bug fix
026 bl312 TBL.sh f2.5c5dm ghgrmp 9s ...........................................................FAIL! rc= 7 at Tue Jun 24 16:09:16 MDT 2014
- expected failure -- trop_mam3 chem changes, ndrop bug fix
031 bl314 TBL.sh f10c4wghgdm outfrq3s 9s ......................................................FAIL! rc= 7 at Tue Jun 24 16:21:53 MDT 2014
- no baseline to compare against -- regression test name changed
033 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .........................................FAIL! rc= 7 at Tue Jun 24 16:23:58 MDT 2014
- expected failure -- trop_mam3 chem changes, ndrop bug fix
042 bl325 TBL.sh f10c5t5mdmsc outfrq3s_subcol 9s ..............................................FAIL! rc= 7 at Tue Jun 24 17:00:45 MDT 2014
- expected failure -- trop_mam3 chem changes, ndrop bug fix
046 bl805 TBL.sh f4c5carmtt1dm outfrq3s_carma 9s ..............................................FAIL! rc= 7 at Tue Jun 24 17:32:53 MDT 2014
- expected failure -- trop_mam3 chem changes, ndrop bug fix
goldbach/pgi:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s ...........................................................FAIL! rc= 7 at Tue Jun 24 15:05:54 MDT 2014
- expected failure -- trop_mam3 chem changes, ndrop bug fix
010 bl112 TBL.sh e8c5dm ghgrmp 9s .............................................................FAIL! rc= 7 at Tue Jun 24 15:14:33 MDT 2014
- expected failure -- trop_mam3 chem changes, ndrop bug fix
012 bl113 TBL.sh e8c5paqdm aqpgro+aquaplanet_cam5 3s ..........................................FAIL! rc= 7 at Tue Jun 24 15:20:49 MDT 2014
- expected failure -- trop_mam3 chem changes, ndrop bug fix
016 bl114 TBL.sh e8c5dm co2rmp 9s .............................................................FAIL! rc= 7 at Tue Jun 24 15:23:45 MDT 2014
- expected failure -- trop_mam3 chem changes, ndrop bug fix
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ..........................................................FAIL! rc= 7 at Tue Jun 24 15:39:07 MDT 2014
- expected failure -- trop_mam3 chem changes, ndrop bug fix
031 bl312 TBL.sh f2.5c5dm ghgrmp 9s ...........................................................FAIL! rc= 7 at Tue Jun 24 16:07:09 MDT 2014
- expected failure -- trop_mam3 chem changes, ndrop bug fix
036 bl314 TBL.sh f10c4wghgdm outfrq3s 9s ......................................................FAIL! rc= 7 at Tue Jun 24 16:22:18 MDT 2014
- no baseline to compare against -- regression test name changed
041 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .........................................FAIL! rc= 7 at Tue Jun 24 16:39:10 MDT 2014
- expected failure -- trop_mam3 chem changes, ndrop bug fix
044 bl317 TBL.sh f10c5cdm outfrq3s 9s .........................................................FAIL! rc= 7 at Tue Jun 24 16:50:17 MDT 2014
- expected failure -- trop_mam3 chem changes, ndrop bug fix
047 bl319 TBL.sh f10c5nonedm outfrq3s_bam 9s ..................................................FAIL! rc= 7 at Tue Jun 24 16:59:35 MDT 2014
- no baseline to compare against -- regression test name changed
050 bl320 TBL.sh f10c5nonedm rad_diag 9s ......................................................FAIL! rc= 7 at Tue Jun 24 17:02:00 MDT 2014
- no baseline to compare against -- regression test name changed
053 bl321 TBL.sh f10c5cdm atrain 9s ...........................................................FAIL! rc= 7 at Tue Jun 24 17:05:20 MDT 2014
- expected failure -- trop_mam3 chem changes, ndrop bug fix
056 bl420 TBL.sh f4c5mam7dm outfrq3s 9s .......................................................FAIL! rc= 7 at Tue Jun 24 17:21:54 MDT 2014
- no baseline to compare against -- regression test name changed
061 bl011 TBL.sh f4c5portdm outfrq24h_port 2d .................................................FAIL! rc= 7 at Tue Jun 24 17:55:39 MDT 2014
- change cfc11 to cfc11star in rad_climate specification (trop_start_mam3 chem configuration)
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: cam5, waccm, cam-chem
- what platforms/compilers: All
- nature of change: larger than roundoff but same climate
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
https://svn-ccsm-models.cgd.ucar.edu/cam1/branch_tags/ccmi_cam5_3_40_tags/ccmi07_cam5_3_40
- platform/compilers:
yellowstone / intel
- configure commandline:
create_newcase -mach yellowstone -compset FC5 -res f19_f19 -case /glade/p/acd/fvitt/cesm/cases/f.e13.FC5.f19_f19.test.011
- build-namelist command (or complete namelist):
defaults for the FC5 f19_f19 configuration
- MSS location of output:
/home/fvitt/csm/f.e13.FC5.f19_f19.test.011
MSS location of control simulations used to validate new climate:
/home/fvitt/csm/f.e13.FC5.f19_f19.cntl.001
URL for AMWG diagnostics output used to validate new climate:
http://webext.cgd.ucar.edu/FCLIMO/f.e13.FC5.f19_f19.test.011/atm/f.e13.FC5.f19_f19.test.011-f.e13.FC5.f19_f19.cntl.001/
===============================================================
===============================================================
Tag name: cam5_3_40
Originator(s): eaton
Date: Thu May 22 11:26:49 MDT 2014
One-line Summary: Add hooks for UNICON; add ne120 aquaplanet IC files.
Purpose of changes:
. Add hooks for running the UNICON code along with some associated clean
up. The UNICON parameterization is not yet on the trunk, just the CAM
interface code.
. Add IC files for ne120 aquaplanet runs (one for cam4, one for cam5).
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system:
. extensions to turn unicon on/off
Describe any changes made to the namelist:
. Add some variables that are used as tuning knobs by UNICON to the namelist:
- add sol_factb_interstitial and sol_factic_interstitial for tuning wet
scavening of modal aerosol
- add conv_water_frac_limit
- add eddy_moist_entrain_a2l
List any changes to the defaults for the boundary datasets:
. add IC files:
atm/cam/inic/homme/cami_0003-01-01_ne120np4_L26_ape_c140519.nc
atm/cam/inic/homme/cami_0003-01-01_ne120np4_L30_ape_c140519.nc
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do:
models/atm/cam/src/physics/cam/unicon_cam.F90
. CAM interface code for UNICON
. use USE_UNICON macro to create stub interfaces for when unicon not enabled
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/build-namelist
. add sol_factb_interstitial and sol_factic_interstitial for tuning wet
scavening of modal aerosol
. add conv_water_frac_limit
. add eddy_moist_entrain_a2l
models/atm/cam/bld/config_files/definition.xml
. add parameter for unicon on/off switch
models/atm/cam/bld/configure
. add unicon on/off switch.
- When unicon enabled check that Park macrophysics, UW PBL, and MG
microphysics are on.
- Add 5 advected tracers for unicon.
. add USE_UNICON to cpp macros when unicon is enabled
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. add UNICON specific settings for
- conv_water_frac_limit
- eddy_moist_entrain_a2l
- sol_factb_interstitial and sol_factic_interstitial
- deep_scheme and shallow_scheme
- cldfrc_rhminl
- cldwat_icritc
models/atm/cam/bld/namelist_files/namelist_definition.xml
. add UNICON as valid value for deep_scheme, shallow_scheme
. add definition for
- eddy_moist_entrain_a2l
- conv_water_frac_limit
- sol_factb_interstitial and sol_factic_interstitial
. conv_water_in_rad moved from phys_ctl_nl to conv_water_nl
models/atm/cam/src/chemistry/modal_aero/aero_model.F90
. sol_factb_interstitial and sol_factic_interstitial added to namelist
models/atm/cam/src/control/cam_comp.F90
models/atm/cam/src/control/cam_restart.F90
models/atm/cam/src/cpl/atm_import_export.F90
models/atm/cam/src/physics/cam/restart_physics.F90
. fix to restart
- add cam_in cflx values to restart
- don't zero out cam_in%cflx before adding values from the coupler
models/atm/cam/src/control/runtime_opts.F90
. add call to conv_water_readnl
models/atm/cam/src/physics/cam/cloud_diagnostics.F90
. remove conv_water_in_rad from conv_water_4rad actual args
models/atm/cam/src/physics/cam/cloud_fraction.F90
. change do loop limit from pver-1 to pver (not answer changing)
models/atm/cam/src/physics/cam/conv_water.F90
. add namelist
. conv_water_4rad
- remove conv_water_mode dummy arg and get info from module data
- add some pointers into the state obj to cleanup code
models/atm/cam/src/physics/cam/convect_deep.F90
. add treatment of deep_scheme='UNICON'. This acts the same as
deep_scheme='off' since the actual call to the unicon code is made from
the convect_shallow driver.
models/atm/cam/src/physics/cam/convect_shallow.F90
. add treatment of shallow_scheme='UNICON'
. convect_shallow_register
- add call to unicon_cam_register
. convect_shallow_init
- add call to unicon_cam_init
. add convect_shallow_implements_cnst/convect_shallow_init_cnst methods
. convect_shallow_tend
- add sgh30, cam_in dummy args
- add call to unicon_cam_tend
models/atm/cam/src/physics/cam/eddy_diff.F90
. set a2l from init_eddy_diff
. add UW_wsed to master field list
. compute_eddy_diff
- add intent(out) args tkes, went
- change ipbl, kpblh to integer types since the pbuf supports it
. caleddy
- add intent(out) args went, wsed_CL
models/atm/cam/src/physics/cam/macrop_driver.F90
. macrop_driver_init
- add dummy arg pbuf2d and add initializers for pbuf fields declared in
this module.
. macrop_driver_tend
- remove unused conv_water_in_rad
models/atm/cam/src/physics/cam/micro_mg1_0.F90
. add stratiform evap area diagnostic output
models/atm/cam/src/physics/cam/micro_mg_cam.F90
. mods to add stratiform evap diagnostics to pbuf
models/atm/cam/src/physics/cam/phys_control.F90
. remove conv_water_in_rad
. remove checks for valid values that are done by build-namelist
models/atm/cam/src/physics/cam/phys_debug.F90
. add phys_debug_state_init and phys_debug_state_out methods to help with
output of state variables from various points in the code.
models/atm/cam/src/physics/cam/physpkg.F90
. some cleanup of how phys_getopts is called. Mainly collected the duplicated calls
and just do it once from the registration routine.
. phys_init
- add pbuf2d arg to call convect_shallow_init
- add pbuf2d arg to call macrop_driver_init
. tphysac
- add call to unicon_cam_org_diags
. tphysbc
- add sgh30, cam_in args to convect_shallow_tend
models/atm/cam/src/physics/cam/vertical_diffusion.F90
. add eddy_moist_entrain_a2l to namelist
. add fields to pbuf for use by UNICON
models/atm/cam/src/physics/cam/wv_saturation.F90
. add findsp to public methods
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS
goldbach/nag: All PASS
goldbach/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_3_39
Originator(s): santos
Date: 2014/05/19
One-line Summary: SC-WACCM development, reduce bare add_default calls.
Purpose of changes:
- Merge changes from the CESM 1.2.2 release, mostly SC-WACCM changes and
WACCM bug fixes.
- Control all add_default calls with namelist flags.
- Remove last traces of CSIM, and the COUP_DOM CPP macro.
Bugs fixed (include bugzilla ID):
- #1837: WACCM: QBO forcing gives wrong output with do_circulation_diags = .false.
- #1841: WACCM5: zisocl failures for some compsets
Describe any changes made to build system:
Describe any changes made to the namelist:
- solar_htng_spctrl_scl is now properly set for SC-WACCM.
- do_circulation_diags no longer defaults to .true. for SC-WACCM or
WACCM-SE.
- Add new xs_short_file and xs_long_file (includes Lyman-alpha and
quantum yields for jh2so4, h/t Chuck Bardeen).
- Add new history flags:
- history_waccmx
- history_chemistry
- history_carma
- history_clubb
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
A models/atm/cam/bld/namelist_files/use_cases/waccm_sc_1850_cam4.xml
A models/atm/cam/bld/namelist_files/use_cases/waccm_sc_1955-2005_cam4.xml
A models/atm/cam/bld/namelist_files/use_cases/waccm_sc_2005-2100_cam4_rcp26.xml
A models/atm/cam/bld/namelist_files/use_cases/waccm_sc_2005-2100_cam4_rcp45.xml
A models/atm/cam/bld/namelist_files/use_cases/waccm_sc_2005-2100_cam4_rcp85.xml
- New SC-WACCM use_cases for 1850, 1955-2005, and RCP 2.6, 4.5,
and 8.5.
A models/atm/cam/src/dynamics/fv/zonal_mean.F90
- Short utility to do zonal averages, extracted from the uzm
calculation in ctem.F90.
A models/atm/cam/test/system/config_files/wg1.9c4dm
A models/atm/cam/test/system/nl_files/outfrq3s_2005
- Files used for new SC-WACCM RCP 4.5 test.
List all existing files that have been modified, and describe the changes:
M SVN_EXTERNAL_DIRECTORIES
- Update scripts and Machines externals to get new compsets.
M models/atm/cam/bld/build-namelist
- Fix SC-WACCM and WACCM-SE defaults.
- Add new history flags.
- Remove mention of CSIM.
M models/atm/cam/bld/configure
- Age of air tracers are no longer on by default for SC-WACCM.
- Remove COUP_DOM macro and CSIM paths.
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- Add solar_data_file for waccm_ghg.
- xs_* file changes for H2SO4.
- TMS settings now trigger based on waccm_phys, not specific
chemistry packages.
- Defaults for new history outputs (currently all .true., to match
previous behavior when the corresponding configure options were
set to .true.).
- Remove CSIM variables.
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- Add new history variables.
- Correct descriptions to match new/current behavior.
- Remove mentions of CSIM.
D models/atm/cam/bld/namelist_files/use_cases/waccm_cam4_nuclear_winter.xml
A models/atm/cam/bld/namelist_files/use_cases/waccm_carma_bc_2013_cam4.xml
M models/atm/cam/test/system/input_tests_master
- Rename use_case, which is now a generic black carbon use_case
instead of a nuclear winter use_case.
- New test for SC-WACCM RCP 4.5 case.
M models/atm/cam/bld/namelist_files/use_cases/waccm_sc_2000_cam4.xml
- Remove variables produced by modules that are no longer enabled
by default for SC-WACCM.
M models/atm/cam/src/chemistry/bulk_aero/seasalt_model.F90
M models/atm/cam/src/chemistry/mozart/chemistry.F90
M models/atm/cam/src/chemistry/mozart/mo_exp_sol.F90
M models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
M models/atm/cam/src/chemistry/mozart/mo_setinv.F90
M models/atm/cam/src/chemistry/pp_none/chemistry.F90
M models/atm/cam/src/control/cam_comp.F90
M models/atm/cam/src/physics/cam/cldwat2m_macro.F90
M models/atm/cam/src/physics/cam/convect_shallow.F90
M models/atm/cam/src/physics/cam/rad_constituents.F90
M models/atm/cam/src/physics/cam/tropopause.F90
M models/atm/cam/src/physics/cam/zm_conv_intr.F90
M models/atm/cam/src/physics/carma/models/cirrus_dust/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/dust/carma_model_mod.F90
- Remove unnecessary uses of add_default.
- Remove commented out add_default calls.
M models/atm/cam/src/chemistry/mozart/exbdrift.F90
M models/atm/cam/src/physics/cam/vertical_diffusion.F90
M models/atm/cam/src/physics/waccm/iondrag.F90
M models/atm/cam/src/physics/waccm/nlte_lw.F90
M models/atm/cam/src/physics/waccm/qbo.F90
M models/atm/cam/src/physics/waccm/radheat.F90
- Call add_default for WACCM variables only if history_waccm is
set.
M models/atm/cam/src/chemistry/mozart/lin_strat_chem.F90
M models/atm/cam/src/chemistry/mozart/mo_neu_wetdep.F90
M models/atm/cam/src/chemistry/utils/aircraft_emit.F90
- Call add_default for chemistry variables only if
history_chemistry is set.
M models/atm/cam/src/chemistry/mozart/mo_apex.F90
M models/atm/cam/src/chemistry/mozart/mo_aurora.F90
M models/atm/cam/src/chemistry/mozart/mo_photo.F90
- Fix erroneous usage of pbuf_set_field that affected WACCM-X.
- History fields that were added in mo_apex, but set in mo_aurora,
are now controlled exclusively by mo_aurora.
M models/atm/cam/src/chemistry/utils/tracer_data.F90
- Add ability to specify PS in zonally averaged files.
M models/atm/cam/src/dynamics/eul/dyn_comp.F90
- Only add DTH to output if history_amwg is set.
M models/atm/cam/src/dynamics/fv/ctem.F90
M models/atm/cam/src/dynamics/fv/dp_coupling.F90
- Calculation of zonal mean zonal wind for QBO nudging is done in
zonal_mean, not in ctem, so that the QBO no longer breaks down
when circulation diagnostics are turned off.
M models/atm/cam/src/physics/cam/cam_diagnostics.F90
- Move add_default for ideal_phys inside history_amwg conditional.
- Use history_waccm, not chem_is, to test for whether WACCM
add_default calls should be done.
- Remove CSIM-specific outputs.
M models/atm/cam/src/physics/cam/clubb_intr.F90
- Call add_default only if history_clubb is set.
- PBLH is a necessary part of the AMWG set, so it is added by
history_amwg instead of history_clubb.
M models/atm/cam/src/physics/cam/eddy_diff.F90
- Turn off an error condition for high-top runs, in order to
prevent WACCM5 from needlessly aborting the run.
M models/atm/cam/src/physics/cam/modal_aer_opt.F90
- Add defaults for diagnostic calls only if history_aero_optics is
set.
M models/atm/cam/src/physics/cam/ndrop_bam.F90
- Call add_default for CCN3 only if history_amwg is set, since it
is one of the AMWG diagnostic variables.
M models/atm/cam/src/physics/cam/phys_control.F90
- Add history fields to phys_control.
M models/atm/cam/src/physics/cam/physics_types.F90
- Add missing deallocation of state%cid.
M models/atm/cam/src/physics/carma/cam/carma_intr.F90
- Call add_default for CARMA variables only if history_carma is
set.
M models/atm/cam/src/physics/waccmx/ionosphere.F90
M models/atm/cam/src/physics/waccmx/majorsp_diffusion.F90
- Call add_default for WACCM-X variables only if history_waccmx is
set.
M src/utils/cam_dom/ocn_comp.F90
- Remove TSOCN output, as it is inconveniently located and seems to
have been only for CSIM.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
017 bl326 TBL.sh wg1.9c4dm outfrq3s_2005+waccm_sc_2005-2100_cam4_rcp45 9s .FAIL! rc= 7 at Mon May 19 14:31:15 MDT 2014
goldbach/nag:
031 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7 at Fri May 16 15:33:17 MDT 2014
goldbach/pgi or jaguar/pgi:
036 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7 at Fri May 16 15:59:48 MDT 2014
The goldbach SC-WACCM baselines failed because of a bug fix in solar_data
settings, and due to changes in the output. bl326 is a new test, so it has
no baseline to compare to.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: SC-WACCM with default namelist.
- what platforms/compilers: All.
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
Larger than roundoff; fixed a bug in SC-WACCM solar_data_file handling.
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_38
Originator(s): fvitt, aconley
Date: 16 May 2014
One-line Summary: Include PORT and fix use case bugs
Purpose of changes:
Place the Parallel Offline Radiation Tool (PORT) on the development trunk.
PORT is used to compute both radiative forcing and instantaneous radiative
forcing. PORT computes radiative heating rates and fluxes on model-generated
datasets (as well as user-modified datasets) without any responses or feedback
of non-radiative processes.
If the rad_data_fdh namelist variable is set to .true., radiative fluxes
and heating rates are computed under the assumption of fixed dynamical
heating (fdh). Under fdh, changes in temperature tendencies (due to
changes in composition or optical properties) above the tropopause
are applied. Only temperatures are allowed to adjust.
Fix bugs in build-namelist use cases:
. 2000_cam4_trop_strat_soa
. sd_cam4_trop_strat_soa
. cam4_bam_radpsv_geos5
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Added the following CAM configure option to activate unit driver
-offline_drv
where is:
stub -- does nothing, used for testing
rad -- invokes only radiation_tend subroutine (PORT)
Describe any changes made to the namelist:
Added namelist options:
. offline_driver_infile -- filepath of a single input file for offline driver (or PORT)
. offline_driver_fileslist -- filepath of an ascii file containing a list of input files for PORT
. rad_data_fdh -- switch to turn on "fixed dynamical heating" in PORT
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
A models/atm/cam/src/unit_drivers/drv_input_data.F90
A models/atm/cam/src/unit_drivers/rad/unit_driver.F90
A models/atm/cam/src/unit_drivers/rad/README
A models/atm/cam/src/unit_drivers/rad
A models/atm/cam/src/unit_drivers/offline_driver.F90
A models/atm/cam/src/unit_drivers/stub/unit_driver.F90
A models/atm/cam/src/unit_drivers/stub
A models/atm/cam/src/unit_drivers
- unit driver code
A models/atm/cam/test/system/nl_files/outfrq24h_port
A models/atm/cam/test/system/config_files/f4c5portdh
A models/atm/cam/test/system/config_files/f4c5portdm
- regression tests added for PORT
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/configure
- offline_drv configure option added
- set surface models to stub versions for PORT configurations
M models/atm/cam/bld/config_files/definition.xml
- offline_drv configure option added
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- added namelist options
. rad_data_fdh -- switch to turn on "Fixed Dynamical Heating" in PORT configuration
. offline_driver_infile -- filepath of a single input file for offline driver (or PORT)
. offline_driver_fileslist -- filepath of an ascii file containing a list of input files for PORT
M models/atm/cam/bld/namelist_files/use_cases/2000_cam4_trop_strat_soa.xml
M models/atm/cam/bld/namelist_files/use_cases/sd_cam4_trop_strat_soa.xml
- replaced 'SSLT*PP' fields in fincl1 list with 'SSLT*WET'
M models/atm/cam/bld/namelist_files/use_cases/cam4_bam_radpsv_geos5.xml
- flbc_list needed
M models/atm/cam/bld/build-namelist
- set defaults of iradsw and iradlw to 1 for PORT configurations
M models/atm/cam/src/control/runtime_opts.F90
- invoke offline_driver_readnl routine
M models/atm/cam/src/control/ncdio_atm.F90
- add missing timelevel optional argument to call to in_real_3dncolphys routine for unstructed grids
M models/atm/cam/src/control/cam_comp.F90
- some code cleanup
- skip dynamics and "before coupler" part of physics (tphysbc) if PORT is active
- invoke offline_driver_run instead of "after coupler" part of physics (tphysac) if PORT is active
M models/atm/cam/src/physics/rrtmg/radiation.F90
M models/atm/cam/src/physics/cam/radiation.F90
- removed unused reference to concld
- added call to rad_data_register()
- added 'FSNR' and 'FLNR' diagnostics
- removed unused landm argument in radiation_tend subroutine
M models/atm/cam/src/physics/cam/radiation_data.F90
- combined with PORT's rad_data_input module
- add pbuf fields needed for the "fixed dynamical heating" functionality of PORT
- changes for cam5 physics (modal aerosols)
- added "fixed dynamical heating" functionality
M models/atm/cam/src/physics/cam/modal_aer_opt.F90
- add missing errcode argument to pbuf_get_index call
M models/atm/cam/src/physics/cam/physpkg.F90
- remove unused "landm" from argument list to radiation_tend
M models/atm/cam/src/physics/cam/radheat.F90
M models/atm/cam/src/physics/waccm/radheat.F90
- added method to turn off chemical heating in the upper atmosphere so the PORT can be used
in a waccm configuration
M models/atm/cam/src/chemistry/utils/tracer_data.F90
- added abort option to incr_filename function. If abort is .false. then return
'NOT_FOUND' if not able to increment the filename rather than calling endrun
M models/atm/cam/test/system/test_driver.sh
- added ability to use gpgpu queue on yellowstone
M models/atm/cam/test/system/tests_pretag_yellowstone
M models/atm/cam/test/system/tests_pretag_goldbach_pgi
M models/atm/cam/test/system/input_tests_master
- regression tests added for PORT
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
091 bl010 TBL.sh f4c5portdh outfrq24h_port 2d .............................FAIL! rc= 7 at Fri May 16 03:52:04 MDT 2014
-no baseline to compare against
goldbach/nag: All pass
goldbach/pgi:
061 bl011 TBL.sh f4c5portdm outfrq24h_port 2d .............................FAIL! rc= 7 at Thu May 15 19:10:14 MDT 2014
-no baseline to compare against
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_37
Originator(s): dennis, srinathv, eaton
Date: Wed May 7 07:57:45 MDT 2014
One-line Summary: cleanup and vectorize wetdepa_v2; BGQ mods for standalone build
Purpose of changes:
. Mods to vectorize the wetdepa_v2 subroutine used by the modal aerosol
code were contributed by John Dennis and Srinath Vadlamani. Eaton
subsequently cleaned the unused code out of the subroutine. The answers
are different at the roundoff level.
. Mods for the BGQ platform were added to configure and Makefile.in for the
CAM standalone build.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system:
. Add BGQ option to CAM standalone build
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes:
. The wetdepa_v2 code is 70% faster on yellowstone. The overall speedup of
CAM is about 1.5%
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/Makefile.in
. add -DHAVE_MPI when building with MPI -- used by timing lib code
. add section for BGQ
models/atm/cam/bld/config_files/sys_defaults.xml
. add entry for bgq to turn on both mpi and threading by default
models/atm/cam/bld/configure
. add bgq as possible value for target_os
. add default compilers for bgq
models/atm/cam/src/chemistry/aerosol/wetdep.F90
. wetdepa_v2 - vectorization mods
- combine the four occurances of the 3-way conditional into a single
occurance.
- replace scalar temporaries by arrays as necessary for vectorization
. wetdepa_v2 - cleanup mods
- remove unused weight factor (hardcoded to 0.0)
- remove unused dummy args: t
- remove set but unused vars: adjfac
- remove unused dummy arg rate1ord_cw2pr_st
- remove unused local vars: aqfrac, cwatc, cwats, cwatp, gafrac, hconst,
mpla, mplb, part, patm, precbl, precmin
- remove set but not used vars: tracab, sol_factii, sol_factiic, sol_factbi
- remove unused dummy arg sol_factbi_in, sol_factii_in, sol_factiic_in
- add local vars to remove some duplicated calculations:
clds for stratiform cloud fraction
trac_qqcw - common to tracer_incu and tracer_mean
- remove cldmabc - not needed since cldvcu always supplies its values
- remove cldmabs - not needed since cldvst always supplies its values
- remove dummy arg cldv - not used
- replace overloaded vars scrs1, scrs2 with variables specific to
convective and stratiform contributions to incloud and below cloud
scavenging... conv_scav_ic, conv_scav_bc, st_scav_ic, st_scav_bc
models/atm/cam/src/chemistry/modal_aero/aero_model.F90
. aero_model_wetdep
- remove actual args for t from calls to wetdepa_v2
- remove wetdepa_v2 actual args: rate1ord_cw2pr_st
- remove optional args sol_factbi, sol_factii, sol_factiic
- remove optional args qqcw and f_act_conv in the call for the cloud
borne aerosols.
- remove actual arg dep_inputs%cldv in wetdepa_v2 calls
models/atm/cam/src/physics/carma/cam/carma_intr.F90
. carma_wetdep_tend
- remove actual args for t from call to wetdepa_v2
- remove actual arg cldv in wetdepa_v2 call
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS except:
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Tue May 6 04:09:39 MDT 2014
016 bl330 TBL.sh f4c5paqdh aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Tue May 6 04:17:24 MDT 2014
024 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...............................FAIL! rc= 7 at Tue May 6 04:28:38 MDT 2014
039 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ......................................FAIL! rc= 7 at Tue May 6 04:42:04 MDT 2014
043 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ............................FAIL! rc= 7 at Tue May 6 04:56:50 MDT 2014
046 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .....................................FAIL! rc= 7 at Tue May 6 05:05:09 MDT 2014
053 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...........................FAIL! rc= 7 at Tue May 6 05:27:41 MDT 2014
060 bl430 TBL.sh fm1.9c5dh outfrq3s_NEUwetdep 9s ..........................FAIL! rc= 7 at Tue May 6 05:50:29 MDT 2014
079 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ....................FAIL! rc= 7 at Tue May 6 06:43:50 MDT 2014
081 bl740 TBL.sh h30c5h outfrq3s 9s .......................................FAIL! rc= 7 at Tue May 6 06:46:33 MDT 2014
goldbach/nag: All PASS except:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Mon May 5 15:15:15 MDT 2014
010 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Mon May 5 15:21:40 MDT 2014
014 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Mon May 5 15:28:13 MDT 2014
018 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Mon May 5 15:37:57 MDT 2014
026 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Mon May 5 16:04:29 MDT 2014
033 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 7 at Mon May 5 16:17:46 MDT 2014
042 bl325 TBL.sh f10c5t5mdmsc outfrq3s_subcol 9s ..........................FAIL! rc= 7 at Mon May 5 16:45:46 MDT 2014
046 bl805 TBL.sh f4c5carmtt1dm outfrq3s_carma 9s ..........................FAIL! rc= 7 at Mon May 5 17:13:20 MDT 2014
goldbach/pgi: All PASS except:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Mon May 5 15:23:50 MDT 2014
010 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Mon May 5 15:31:42 MDT 2014
012 bl113 TBL.sh e8c5paqdm aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Mon May 5 15:37:29 MDT 2014
016 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Mon May 5 15:41:09 MDT 2014
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Mon May 5 15:55:31 MDT 2014
031 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Mon May 5 16:22:58 MDT 2014
041 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 7 at Mon May 5 16:50:08 MDT 2014
044 bl317 TBL.sh f10c5cdm outfrq3s 9s .....................................FAIL! rc= 7 at Mon May 5 16:59:10 MDT 2014
053 bl321 TBL.sh f10c5cdm atrain 9s .......................................FAIL! rc= 7 at Mon May 5 17:14:27 MDT 2014
056 bl420 TBL.sh f7mode4c5dm outfrq3s 9s ..................................FAIL! rc= 7 at Mon May 5 17:28:41 MDT 2014
The baseline failures are all due to roundoff level differences introduced
into subroutine wetdepa_v2 which is used by the modal aerosol code.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: Roundoff differences in configurations
that use the prognostic modal aerosol code.
===============================================================
===============================================================
Tag name: cam5_3_36
Originator(s): cacraig, goldy
Date: May 5, 2014
One-line Summary: Bug fixes to subcolumns
Purpose of changes: Bug fixes and cleanup for subcolumns
Bugs fixed (include bugzilla ID):1952 (incorrect allocation of state_sc)
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/src/physics/cam/subcol_utils.F90
M models/atm/cam/src/physics/cam/subcol_utils.F90.in
- Changed all state intents to in from inout
- Allocation of state_sc was incorrect and decided it should be an endrun instead
M models/atm/cam/src/physics/cam/physics_buffer.F90
M models/atm/cam/src/physics/cam/physics_buffer.F90.in
- Bug in setting of dimsizes(:) and then overwriting dimsizes(1) when threading
in pbuf1d_field_by_index. Corrected to set only dimsizes(2:).
- Cleanup of logic when using optional start/count in get_pbuf1d_field_by_index
- Corrected spelling of persistence
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All BFB
goldbach/nag: All BFB
goldbach/pgi or jaguar/pgi: All BFB
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_35
Originator(s): fvitt, tilmes
Date: 02 May 2014
One-line Summary: Improve performance of satellite history output
and bug fix in modal aerosol surface area routine
Purpose of changes:
- It has been observed that CAM's run-time performance is degraded
when satellite history output is turned on, especially when running
on higher resolution grids. The sat_hist_write method was spending
the bulk of its time doing a "find nearest neighbor" operation to
determine the nearest column(s) to output. Each PE was searching the
entire global grid (both dynamics grid and physics grid). Modifications
here parallelizes the find nearest neighbor operation and performance
is greatly improved.
- Fix a bug in modal aerosol surface area routine used for heterogeneous
chemistry reactions.
- Build-namelist use case changes to be consistent with 1.2.2 release.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/namelist_files/use_cases/2000_cam4_trop_strat_soa.xml
M models/atm/cam/bld/namelist_files/use_cases/2000_cam5_trop_strat_mam3.xml
M models/atm/cam/bld/namelist_files/use_cases/sd_cam4_trop_strat_soa.xml
M models/atm/cam/bld/namelist_files/use_cases/sd_cam5_trop_strat_mam3.xml
Changes to be consistent with 1.2.2 release
M models/atm/cam/src/chemistry/mozart/mo_usrrxt.F90
The use of dgnumwet in MODAL version of surfarea needs to be converted from "m" to "cm"
M models/atm/cam/src/control/sat_hist.F90
M models/atm/cam/src/dynamics/sld/dyn_grid.F90
M models/atm/cam/src/dynamics/eul/dyn_grid.F90
M models/atm/cam/src/dynamics/se/dyn_grid.F90
M models/atm/cam/src/dynamics/fv/dyn_grid.F90
Improve performance of sat_hist
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
060 bl430 TBL.sh fm1.9c5dh outfrq3s_NEUwetdep 9s ..........................FAIL! rc= 7 at Thu May 1 19:07:34 MDT 2014
- expected failure due to the bug fix in surfarea
goldbach/nag: all pass
goldbach/pgi: all pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_34
Originator(s): cacraig, eaton, gettelman
Date: April 25, 2014
One-line Summary: Fix broken COSP, namelist defaults, ideal physics, PGI on hopper, micro_mg_cam
Purpose of changes:
- COSP had include files with .f instead of .inc. Use the new COSP external to allow CESM to compile it
- CAM standalone whn running SE ne16np4 was broken - updated namelist defaults to fix problem
- ideal physics with threading was broken. Eliminated ptend as a thread private variable and made
it an array with a chunk index
- Fixed PGI compilation on hopper
- Andrew provided a bug fix for micro_mg_cam for vprao, vprco, and racau
Bugs fixed (include bugzilla ID): 1894
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/configure
- fix PGI compilation on hopper
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- fix namelist defaults for CAM SE ne16np4 in CAM standalone
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
- use updated COSP build which allows CESM to compile it directly with their new mods
M models/atm/cam/src/physics/cam/micro_mg_cam.F90
- Changes from Andrew for vprco, vprao and racau
M models/atm/cam/src/physics/cam/physpkg.F90
- Fix for running ideal physics using threading
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB
goldbach/nag: all BFB
goldbach/pgi or jaguar/pgi: all BFB
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_33
Originator(s): fvitt
Date: 22 Apr 2014
One-line Summary: Chemistry bug fixes and use case changes
Purpose of changes:
Correction to aerosol surface area used in heterogeneous chemical
reactions, O1D+O2 reaction rate update, and updates to build-namelist
use cases
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
A models/atm/cam/bld/namelist_files/use_cases/2005-2100_cam4_rcp45_bgc.xml
A models/atm/cam/bld/namelist_files/use_cases/2005-2100_cam4_rcp85_bgc.xml
A models/atm/cam/bld/namelist_files/use_cases/sd_cam4_trop_strat_soa.xml
A models/atm/cam/bld/namelist_files/use_cases/sd_cam5_trop_strat_mam3.xml
- new build-namelist use cases
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/namelist_files/use_cases/2000_cam4_trop_strat_soa.xml
M models/atm/cam/bld/namelist_files/use_cases/2000_cam5_trop_strat_mam3.xml
- updates to build-namelist use cases
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_setrxt.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_mozart/chem_mech.in
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_setrxt.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/chem_mech.in
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_setrxt.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/chem_mech.in
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_setrxt.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/chem_mech.in
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_setrxt.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_trop_strat_soa/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_strat_soa/chem_mech.in
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_setrxt.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/chem_mech.in
- corrected rate constants for O1D + O2 reaction
- added SO4 to external forcing list
M models/atm/cam/src/chemistry/modal_aero/aero_model.F90
M models/atm/cam/src/chemistry/bulk_aero/aero_model.F90
M models/atm/cam/src/chemistry/mozart/mo_usrrxt.F90
- correction to aerosol surface area used in heterogeneous chemical
reactions when using prognostic modal aerosols
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
051 bl373 TBL.sh fm1.9c4dh outfrq3s 9s ....................................FAIL! rc= 7 at Tue Apr 22 10:44:29 MDT 2014
055 bl376 TBL.sh fm1.9c4dh+testmech outfrq3s 9s ...........................FAIL! rc= 7 at Tue Apr 22 10:53:13 MDT 2014
060 bl430 TBL.sh fm1.9c5dh outfrq3s_NEUwetdep 9s ..........................FAIL! rc= 7 at Tue Apr 22 11:10:20 MDT 2014
063 bl440 TBL.sh fsoa1.9c4dh outfrq3s+soa_chem_megan_emis 9s ..............FAIL! rc= 7 at Tue Apr 22 11:27:14 MDT 2014
- these failures are expected due to changes in O1D+O2 reaction rate and
changes to aerosol surface area used in heterogeneous chemical reactions
goldbach/nag: All Pass
goldbach/pgi: All Pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_32
Originator(s): olson, cacraig
Date: 4/18/14
One-line Summary: Bug fixes for bugzilla 1958
Purpose of changes: Fix computation of the diagnostic term "DTCORE" - total dynamics T-tendency
Bugs fixed (include bugzilla ID): 1958
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton, olson
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/src/physics/cam/check_energy.F90
M models/atm/cam/src/physics/cam/cam_diagnostics.F90
M models/atm/cam/src/physics/cam/physpkg.F90
- bug fix for DTCORE
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
036 bl336 TBL.sh wm4c4h outfrq3s+waccm_1850_cam4 9s .......................FAIL! rc= 7 at Thu Apr 17 20:45:09 MDT 2014
073 bl379 TBL.sh wm1.9c4h outfrq24h+waccm_1850_cam4 2d ....................FAIL! rc= 7 at Thu Apr 17 22:40:06 MDT 2014
Failed due to differences with DTCORE. Jerry Olson approved the new numbers
goldbach/nag: all BFB
goldbach/pgi or jaguar/pgi: all BFB
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_31
Originator(s): santos, Marcus Wagner, cacraig
Date: 04/14/2014
One-line Summary: Bug fixes for bugzilla 1954 and 1919
Purpose of changes:
- Fix write outside array bounds in pilgrim code (1919)
- Fix lon-lat history file from multi-instance CAM-SE (1954)
Bugs fixed (include bugzilla ID): 1919 and 1954
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by:
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
M models/atm/cam/src/utils/pilgrim/parutilitiesmodule.F90
- bug fix from Marcus Wagner for bugzilla 1919 (write outside array bounds in pilgrim code)
M models/atm/cam/src/dynamics/se/interp_mod.F90
- bug fix from Sean Santos for bugzilla 1954 (lon-lat history file from multi-instance CAM-SE: bug and fix)
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS
goldbach/nag: All PASS
goldbach/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_3_30
Originator(s): cacraig, eaton
Date: 04/04/2014
One-line Summary: Changed compiler settings to match CESM to fix failing TEQ test
Purpose of changes:
- Fix failing TEQ test by changing compiler settings
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/test_driver.sh
- Updated modules for yellowstone to match CESM modules
M models/atm/cam/bld/Makefile.in
- Updated compiler flags for yellowstone to match CESM modules
- handful of routines have -O3 optimization
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS except:
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Thu Apr 3 19:59:46 MDT 2014
053 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...........................FAIL! rc= 7 at Thu Apr 3 21:19:13 MDT 2014
081 bl740 TBL.sh h30c5h outfrq3s 9s .......................................FAIL! rc= 7 at Thu Apr 3 22:39:16 MDT 2014
- Answers changed due to modified compiler settings
goldbach/nag: All PASS
goldbach/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_3_29
Originator(s): cacraig
Date: 03/31/2014
One-line Summary: Update to cesm1_3beta08 externals
Purpose of changes:
- Updated to cesm1_3beta08 externals
- Needed to update to CLM branch to eliminate some introduced bugs
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by:
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/test_driver.sh
- need to "module load perlmods" for yellowstone
M SVN_EXTERNAL_DIRECTORIES
- updated to beta08 externals
- Needed to upgrade to CLM branch tag to correct several bugs discovered during CAM testing
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS except:
036 bl336 TBL.sh wm4c4h outfrq3s+waccm_1850_cam4 9s .......................FAIL! rc= 7 at Fri Mar 28 14:04:13 MDT 2014
051 bl373 TBL.sh fm1.9c4dh outfrq3s 9s ....................................FAIL! rc= 7 at Fri Mar 28 14:56:53 MDT 2014
055 bl376 TBL.sh fm1.9c4dh+testmech outfrq3s 9s ...........................FAIL! rc= 7 at Fri Mar 28 15:06:36 MDT 2014
058 bl374 TBL.sh fs1.9c4dh outfrq3s+1850-2005_cam4_super_fast_llnl 9s .....FAIL! rc= 7 at Fri Mar 28 15:14:36 MDT 2014
060 bl430 TBL.sh fm1.9c5dh outfrq3s_NEUwetdep 9s ..........................FAIL! rc= 7 at Fri Mar 28 15:24:34 MDT 2014
063 bl440 TBL.sh fsoa1.9c4dh outfrq3s+soa_chem_megan_emis 9s ..............FAIL! rc= 7 at Fri Mar 28 15:42:57 MDT 2014
073 bl379 TBL.sh wm1.9c4h outfrq24h+waccm_1850_cam4 2d ....................FAIL! rc= 7 at Fri Mar 28 16:06:52 MDT 2014
- Jobs with cam_chempkg='trop_mozart', 'waccm_mozart', 'super_fast_llnl', 'trop_mozart_mam3', 'trop_mozart_soa' all
have answer changes due to CLM mods
088 eq993 TEQ_ccsm.sh ne30_ne30 FC5 h30c5h fcase 2d .......................FAIL! rc= 7 at Fri Mar 28 18:11:41 MDT 2014
- Machines compiler updates on yellowstone cause answer changes. Will change in next tag as may be answer changing to
other tests as well.
goldbach/nag: All PASS except
001 fm001 TFM.sh ..........................................................FAIL! rc= 2 at Thu Mar 27 13:56:02 MDT 2014
Failed because checked out on yellowstone which is using an older svn repository. Will check out next version on
goldbach for this test.
goldbach/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_3_28
Originator(s): cacraig, goldy (with input from many others)
Date: 03/06/2014
One-line Summary: Routines to support sub-columns introduced and sub-columninzation of microphysics
Purpose of changes:
- Last checkin of initial subcolumn development work
- Introduces the subcolumn interface and utility routines
- microphysics is now subcolumnized
- when subcolumns are not specified, no changes
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton, plus numerous group reviews
List all subroutines eliminated: none
D models/atm/cam/src/utils/grid_flag_utils.F90
- Got rid of confusing bit_field_kind and its supporting routines as no longer needed
- Moved col_type parameters to physics_buffer.F90
- Removed obsolete module
List all subroutines added and what they do: none
A models/atm/cam/test/system/config_files/f10c5t5mdmsc
A models/atm/cam/test/system/nl_files/outfrq3s_subcol
- Added regression testing for subcolumns
A + models/atm/cam/src/physics/cam/subcol.F90
- interface layer for subcolumns
- Has the ability to redirect subcolumn-scheme specific routines
- public routines:
subcol_register ! Read scheme from namelist and initialize any scheme-global variables
subcol_init ! Initialize any variables or fields specific to the active scheme
subcol_gen ! Generate subcol fields from GBA fields
subcol_field_avg ! Average subcol fields back into GBA fields
subcol_ptend_avg ! Average sub-column ptend to grid ptend
subcol_readnl ! Namelist reader for subcolumns
subcol_init_restart ! Initialize restart with subcolumn specific fields
subcol_read_restart ! Read subcolumn specific fields from restart
subcol_write_restart ! Write subcolumn specific fields for restart
A + models/atm/cam/src/physics/cam/subcol_utils.F90.in
A + models/atm/cam/src/physics/cam/subcol_utils.F90
- Utilities layer for subcolumns
- public routines:
subcol_utils_init ! Initialize module data (e.g., nsubcol2d)
subcol_get_nsubcol ! Copy chunk from nsubcol2d
subcol_set_nsubcol ! Copy chunk to nsubcol2d
subcol_get_indcol ! Copy chunk from indcol2d
subcol_get_filter ! return the filter values
subcol_set_filter ! set the filter values
subcol_get_weight ! return the weight values
subcol_set_weight ! set the weight values
subcol_field_copy ! copy a physics buffer field into one with subcolumn dimensions
subcol_ptend_copy ! copy a physics_ptend object into one with subcolumn dimensions
subcol_set_subcols ! set nsubcols and copy state & tend objects into one with subcolumn dimensions
subcol_field_avg_shr ! Average subcol fields back into GBA fields
subcol_ptend_avg_shr ! average subcolumn ptend to grid ptend
subcol_field_get_firstsubcol ! Retrieve the first subcolumn and assign to grid
subcol_ptend_get_firstsubcol ! retrieve the first subcolumn from the ptend fields and assign to grid ptend
subcol_unpack ! Unpack a subcolumn field
subcol_pack ! Pack a subcolumn field
subcol_utils_init_restart ! Initialize restart with subcolumn specific fields
subcol_utils_read_restart ! Read subcolumn specific fields from restart
subcol_utils_write_restart ! Write subcolumn specific fields for restart
is_filter_set ! True if filters for averaging have been set
is_weight_set ! True if weights for averaging have been set
is_subcol_on ! true is any subcol_scheme other than "off" is set
subcol_get_scheme ! Return the active subcolumn scheme name
subcol_utils_readnl ! Set the active scheme based on namelist
A + models/atm/cam/src/physics/cam/subcol_tstcp.F90
- Testing module for subcolumns
- Has more robust setting of random indices
- Introduced on fly test of subcolumn indices
List all existing files that have been modified, and describe the changes:
M models/atm/cam
M models/atm/cam/test/system/tests_pretag_goldbach_nag
M models/atm/cam/test/system/input_tests_master
- Added regression test for subcolumns
M models/atm/cam/bld/configure
- Added SILHS to configure for SILHS developer's ease
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- Moved microp_uniform initialized to false
- Introduced a number of subcolumn namelist variables
- Initialized subcol_scheme to "off"
M models/atm/cam/bld/build-namelist
- Introduced subcolumn namelist variables
M models/atm/cam/src/control/runtime_opts.F90
- Added subcol_readnl
M models/atm/cam/src/control/cam_history_support.F90
- Added is_subcol field to field_info strtucture
- Only defines the pio variables if they are associated
M models/atm/cam/src/control/cam_history.F90
- Utilizes the new is_subcol field and handles subcolumns properly
- Introduced an interface for subcol_field_avg_handler to eliminate a circular dependency
M models/atm/cam/src/physics/cam/micro_mg_cam.F90
- Allow microphysics to have grid or subcolumns
- moved microp_uniform the this namelist
- Lower portion of micro_mg_cam reverts back to the grid
M models/atm/cam/src/physics/cam/check_energy.F90
- registers that TEOUT and DTCORE need to have subcolumn fields as well if subcolumns turned on
M models/atm/cam/src/physics/cam/restart_physics.F90
- if subcolumns are on, then call appropriate routines
M models/atm/cam/src/physics/cam/physics_types.F90
- Removed fields from physics_state: nsubcol and indcol. This information is now stored in subcol_utils module
- cpariv and rairv now have local copies of cpair and rair.
If subcolumns, then allocates to subcolumn size and assigns the scalar value
If grid columns, then allocates to pcols and checks that all of array version matches scalar value
If grid columns, and cpairv and/or rairv is not all equal to scalar value, exit with an error
M models/atm/cam/src/physics/cam/physpkg.F90
- uses the new subcolumn routines for init and register
- tphysbc now has logic to handle optional subcolumns for microp_driver
M models/atm/cam/src/physics/cam/microp_driver.F90
- minor cleanup
M models/atm/cam/src/physics/cam/physics_buffer.F90
M models/atm/cam/src/physics/cam/physics_buffer.F90.in
- Now contains col_type parameters
- buffer_hdr has additional fields:
is_copy(ngrid_types): contains info on whether subcolumn field is a copy or not
added: used to flag if pbuf_add_field was called on this field
dimsizes and vardesc have dimension for ngrid_types
- bfg_sc is gone and bfg now dimensioned over ngrid_types
- number of possible dimensions one less due to subcolumns needing a dimension
- new routines which are public:
pbuf_col_type_index: retrieves col_type index based on whether subcolumns turned on or not
pbuf_register_subcol: used to make sure a subcolumn field is created - does not care if it
already was created or not.
- new internal routines:
pbuf_field_has_gridcols: returns true only if it is an allocated grid field
pbuf_field_has_subcols: returns true only if it is an allocated subcol field (and not a copy)
pbuf_field_has_col_type: returns true only if col_type field is allocated
find_pbuf_header: finds a pbuf header pointer based on name and tacks on extra header if name not found
pbuf_register_field_int: register a field in the pbuf (used by pbuf_add_field and was
guts of pbuf_add_field in previous versions). Has additional functionality
to register both grid and subcolumn fields.
get_pbuf2d_field_restart: used by write_restart_field. If subcolumn data, unpacks it before returning field.
Grid data is just returned
write_restart_field: replacement of pbuf_write_field with support for subcolumns
read_restart_field: replacement for pio_read_restart_field with support for subcolumns
- removed obsolete routines: pbuf_set_test_field and pbuf_check_test_field
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS
goldbach/nag: All Pass except newly introduced BL325
goldbach/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_3_27
Originator(s): cchen, hannay, eaton
Date: Tue Mar 4 10:52:02 MST 2014
One-line Summary: Namelist control for AMWG variability
diagnostics; remove rad_diag_1 from CAM5/RCP use cases; add new
history_amwg defaults.
Purpose of changes:
. Add a namelist variable, history_vdiag, which adds the specific fields at
daily, 6-hourly, and 3-hourly output frequency required by the AMWG
variability diagnostics. Setting history_vdiag=.true. in the namelist is
the only action required on the user's part, unless there is a conflict
with the setting of nhtfrq. The history_vdiag option writes fields to
auxilliary history files h1, h2, and h3, and it assumes that the output
frequency for these files is set to daily, 6-hourly, and 3-hourly
respectively. If the value of nhtfrq is set by the user, the first 4
elements must be 0,-24,-6,-3. Otherwise build-namelist will flag an
error. In the case of an error be sure to check whether the use case is
the source of the inconsistent values of nhtfrq.
. All CAM5/RCP use case files have been modified to remove the diagnostic
radiation calculation. This is an expensive diagnostic and it was
decided that it should not be enabled by default.
. Add variables TAUGWX, TAUGWY, TAUTMSX, TAUTMSY to the history_amwg field
list.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist:
. add history_vdiag
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/build-namelist
. add default for history_vdiag
. check that nhtfrq hasn't been set inconsistently, and set it's default if
not set yet.
models/atm/cam/bld/namelist_files/namelist_definition.xml
. add history_vdiag to phys_ctl_nl group
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. add default for history_vdiag (.false.)
models/atm/cam/src/physics/cam/cam_diagnostics.F90
. add module variable history_vdiag
. diag_init
- init history_vdiag via phys_getopts call
- add add_default calls for U200, V200, U850, OMEGA500, PRECT inside of
history_vdiag conditionals. The addfld calls for all these variables
sets them as time averaged. The add_default calls are hardcoded with
the specific auxillary hist file numbers.
models/atm/cam/src/physics/cam/gw_drag.F90
models/atm/cam/src/physics/cam/vertical_diffusion.F90
. add variables TAUGWX, TAUGWY, TAUTMSX, TAUTMSY to the history_amwg field list.
models/atm/cam/src/physics/cam/phys_control.F90
. add module variable history_vdiag
. add history_vdiag to namelist group phys_ctl_nl
. add history_vdiag to phys_getopts accessor method
models/atm/cam/src/physics/cam/radiation.F90
models/atm/cam/src/physics/rrtmg/radiation.F90
. radiation_init
- add add_default calls for FLUT inside history_vdiag conditional.
addfld call adds FLUT as time average field.
models/atm/cam/bld/namelist_files/use_cases/2006-2100_cam5_rcp26.xml
models/atm/cam/bld/namelist_files/use_cases/2006-2100_cam5_rcp60.xml
models/atm/cam/bld/namelist_files/use_cases/2006-2100_cam5_rcp45.xml
models/atm/cam/bld/namelist_files/use_cases/2006-2100_cam5_rcp85.xml
. remove the following setting for rad_diag_1:
'A:Q:H2O','N:O2:O2','N:CO2:CO2','N:ozone:O3','N:N2O:N2O','N:CH4:CH4','N:CFC11:CFC11','N:CFC12:CFC12'
. remove the fincl1 fields that are produced by rad_diag_1:
'FSNT_d1','FSNTC_d1','FSNS_d1','FSNSC_d1'
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS
goldbach/nag: All PASS
goldbach/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_3_26
Originator(s): santos cacraig eaton
Date: 2014/02/21
One-line Summary: Subcolumn prep for MG, update frankfurt to goldbach
Purpose of changes:
- Size distribution calculations for the radiation and diagnostic output
are now done in micro_mg_cam, instead of directly using the outputs from
micro_mg_tend. This will allow them to be done after all subcolumn and
substepping averages, which presents a more consistent state for RRTMG.
- Add new test to check that svn:mergeinfo exists in the root directory of
CAM, and nowhere else.
- Transition from frankfurt to goldbach as a test system.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
A models/atm/cam/test/system/TFM.sh
A models/atm/cam/test/system/find_mergeinfo.sh
- Add find_mergeinfo.sh utility for detecting mergeinfo.
- Add test to ensure that the svn:mergeinfo property only exists on
certain directories.
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/Makefile.in
M models/atm/cam/bld/configure
- Work around issue with PGI and the MPI compiler wrapper on
goldbach.
- When running MCT's configure script, pass along the value of CC
and user-specified LDFLAGS as well as FC.
M models/atm/cam/doc/ChangeLog_template
- Change frankfurt to goldbach.
M models/atm/cam/src/physics/cam/micro_mg_cam.F90
M models/atm/cam/src/physics/cam/micro_mg_utils.F90
- Have micro_mg_cam initialize micro_mg_utils for MG 1 (in future
versions, the portable layer of MG will do this).
- Define mincld in micro_mg_utils. In future versions, this will
allow micro_mg_cam and the portable layer to share this value.
For now, only micro_mg_cam uses it.
- Do size distribution parameter calculations for the radiation
in micro_mg_cam, rather than using the MG outputs. Reorder some
output statements for consistency with this.
M models/atm/cam/test/system/CAM_decomp.sh
M models/atm/cam/test/system/CAM_runcmnd.sh
M models/atm/cam/test/system/TCB.sh
M models/atm/cam/test/system/TCB_ccsm.sh
M models/atm/cam/test/system/archive_baseline.sh
M models/atm/cam/test/system/input_tests_master
M models/atm/cam/test/system/test_driver.sh
D models/atm/cam/test/system/tests_posttag_frankfurt
A models/atm/cam/test/system/tests_posttag_goldbach
D models/atm/cam/test/system/tests_pretag_frankfurt_nag
A models/atm/cam/test/system/tests_pretag_goldbach_nag
D models/atm/cam/test/system/tests_pretag_frankfurt_pgi
A models/atm/cam/test/system/tests_pretag_goldbach_pgi
- Fix issue where ptile=32 does not work consistently on
yellowstone, due to a recent LSF upgrade.
- Update from frankfurt to goldbach.
- Add TFM (mergeinfo test) to NAG pretag tests on goldbach.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Wed Feb 19 20:09:42 MST 2014
014 bl322 TBL.sh fn1.9c5dh outfrq3s 9s ....................................FAIL! rc= 7 at Wed Feb 19 20:15:58 MST 2014
016 bl330 TBL.sh f4c5paqdh aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Wed Feb 19 20:17:22 MST 2014
024 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...............................FAIL! rc= 7 at Wed Feb 19 20:28:53 MST 2014
039 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ......................................FAIL! rc= 7 at Wed Feb 19 20:42:39 MST 2014
043 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ............................FAIL! rc= 7 at Wed Feb 19 20:57:31 MST 2014
046 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .....................................FAIL! rc= 7 at Wed Feb 19 21:05:43 MST 2014
053 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...........................FAIL! rc= 7 at Wed Feb 19 21:29:26 MST 2014
060 bl430 TBL.sh fm1.9c5dh outfrq3s_NEUwetdep 9s ..........................FAIL! rc= 7 at Wed Feb 19 21:53:57 MST 2014
063 bl440 TBL.sh fsoa1.9c4dh outfrq3s+soa_chem_megan_emis 9s ..............FAIL! rc= 7 at Wed Feb 19 22:11:53 MST 2014
079 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ....................FAIL! rc= 7 at Wed Feb 19 22:48:40 MST 2014
081 bl740 TBL.sh h30c5h outfrq3s 9s .......................................FAIL! rc= 7 at Wed Feb 19 22:51:25 MST 2014
goldbach/nag:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Thu Feb 20 13:26:47 MST 2014
010 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Thu Feb 20 13:32:29 MST 2014
014 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Thu Feb 20 13:38:46 MST 2014
018 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Thu Feb 20 13:47:49 MST 2014
026 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Thu Feb 20 14:10:49 MST 2014
033 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 7 at Thu Feb 20 14:20:19 MST 2014
042 bl805 TBL.sh f4c5carmtt1dm outfrq3s_carma 9s ..........................FAIL! rc= 7 at Thu Feb 20 14:57:21 MST 2014
goldbach/pgi or jaguar/pgi:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Thu Feb 20 13:28:30 MST 2014
010 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Thu Feb 20 13:36:01 MST 2014
012 bl113 TBL.sh e8c5paqdm aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Thu Feb 20 13:41:24 MST 2014
016 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Thu Feb 20 13:44:47 MST 2014
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Thu Feb 20 13:58:49 MST 2014
031 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Thu Feb 20 14:25:48 MST 2014
041 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 7 at Thu Feb 20 14:49:53 MST 2014
044 bl317 TBL.sh f10c5cdm outfrq3s 9s .....................................FAIL! rc= 7 at Thu Feb 20 14:58:45 MST 2014
047 bl319 TBL.sh fn10c5dm outfrq3s_bam 9s .................................FAIL! rc= 7 at Thu Feb 20 15:05:23 MST 2014
050 bl320 TBL.sh fn10c5dm rad_diag 9s .....................................FAIL! rc= 7 at Thu Feb 20 15:07:44 MST 2014
053 bl321 TBL.sh f10c5cdm atrain 9s .......................................FAIL! rc= 7 at Thu Feb 20 15:11:19 MST 2014
056 bl420 TBL.sh f7mode4c5dm outfrq3s 9s ..................................FAIL! rc= 7 at Thu Feb 20 15:24:21 MST 2014
059 bl734 TBL.sh hn16c5aqdm outfrq3s_bam+aquaplanet_cam5 9s ...............FAIL! rc= 7 at Thu Feb 20 15:43:58 MST 2014
All CAM 5 baselines were expected to fail due to microphysics changes.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
All MG (therefore all CAM5).
- what platforms/compilers:
All.
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
Larger than roundoff, same climate.
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_25
Originator(s): santos
Date: 2014/01/27
One-line Summary: Add settings for new 60 layer AMIP compset.
Purpose of changes:
- Add namelist settings for new AMIP compset, with high vertical resolution
and spectral gravity waves.
- Clean up work and fix of a minor DEBUG mode bug in CARMA.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
- Add gravity wave settings for new compset, including the new tau_0_ubc
option.
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M SVN_EXTERNAL_DIRECTORIES
- Use https instead of http to get PIO external (fixes Subversion
1.7 error).
M models/atm/cam/bld/build-namelist
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
M models/atm/cam/bld/namelist_files/namelist_definition.xml
M models/atm/cam/src/physics/cam/gw_common.F90
M models/atm/cam/src/physics/cam/gw_drag.F90
- Add option to enforce an upper boundary condition of tau = 0 in
the gravity waves, and turn it on for CAM runs with spectral
gravity waves.
- Add gravity wave settings for high vertical resolution,
non-WACCM runs.
M models/atm/cam/src/physics/carma/models/bc_strat/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/cirrus/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/cirrus_dust/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/dust/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/meteor_impact/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/meteor_smoke/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/mixed_sulfate/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/pmc/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/pmc_sulfate/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/sea_salt/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/sulfate/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/test_detrain/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/test_growth/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/test_passive/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/test_radiative/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/test_swelling/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/test_tracers/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/test_tracers2/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/tholin/carma_model_mod.F90
- Remove unnecessary calls to phys_grid routines.
- Fix bug where uninitialized data was used in a multiplication,
causing a floating point exception (this bug did not affect
results).
M models/atm/cam/doc/ChangeLog
- Fixed the date on the previous commit (Happy New Year!).
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS.
frankfurt/nag: All PASS.
frankfurt/pgi or jaguar/pgi: All PASS.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_24
Originator(s): santos
Date: 2014/01/23
One-line Summary: Improve stability of the ZM scheme.
Purpose of changes:
- In the ZM deep convection scheme, use Brent's method to invert the
entropy equation, as it converges faster.
- In the ZM scheme, for the mixing ratio of water vapor at saturation, use
a ratio over total mass rather than over dry mass, to avoid numerical
overflow that can occur at high temperature.
- Limit RRTMG so that it only operates below 1 Pa, rather than using the
lid of 0.01 Pa that was in place before. (Only impacts WACCM5 and other
high-top RRTMG runs.)
- Prevent rare divide by zero error for very cold grid points in
modal_aero_wateruptake.
Bugs fixed (include bugzilla ID):
#1891 - The "top" of RRTMG is too high in high-top models
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/src/chemistry/utils/modal_aero_wateruptake.F90
- Work around divide-by-zero error by checking for water vapor
saturation pressures that are 0. (If this happens, the saturation
pressure is wrong anyway because the temperature is unrealistic,
but we give the model the chance to recover, or to throw a better
error elsewhere.)
M models/atm/cam/src/physics/cam/wv_sat_methods.F90
M models/atm/cam/src/physics/cam/wv_saturation.F90
M models/atm/cam/src/physics/cam/zm_conv.F90
- Use improved entropy inverter in zm_conv.
- Remove qmmr, and have zm_conv use qsat_water instead.
M models/atm/cam/src/physics/rrtmg/rrtmg_state.F90
- Lower the lid on RRTMG to 1 Pa.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
004 bl132 TBL.sh e48c4dh ghgrmp+1850_cam4 9s ..............................FAIL! rc= 7 at Wed Jan 22 16:30:24 MST 2014
006 bl133 TBL.sh e48c4paqdm aqpgro+aquaplanet_cam4 3s .....................FAIL! rc= 7 at Wed Jan 22 16:31:33 MST 2014
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Wed Jan 22 16:34:29 MST 2014
014 bl322 TBL.sh fn1.9c5dh outfrq3s 9s ....................................FAIL! rc= 7 at Wed Jan 22 16:46:44 MST 2014
016 bl330 TBL.sh f4c5paqdh aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Wed Jan 22 16:48:19 MST 2014
020 bl331 TBL.sh f4c4dh co2rmp+1850_cam4 9s ...............................FAIL! rc= 7 at Wed Jan 22 16:56:05 MST 2014
024 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...............................FAIL! rc= 7 at Wed Jan 22 17:10:37 MST 2014
026 bl333 TBL.sh f4c4paqdh aqpgro+aquaplanet_cam4 3s ......................FAIL! rc= 7 at Wed Jan 22 17:11:14 MST 2014
036 bl336 TBL.sh wm4c4h outfrq3s+waccm_1850_cam4 9s .......................FAIL! rc= 7 at Wed Jan 22 17:18:15 MST 2014
039 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ......................................FAIL! rc= 7 at Wed Jan 22 17:26:23 MST 2014
043 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ............................FAIL! rc= 7 at Wed Jan 22 17:45:20 MST 2014
046 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .....................................FAIL! rc= 7 at Wed Jan 22 18:02:32 MST 2014
051 bl373 TBL.sh fm1.9c4dh outfrq3s 9s ....................................FAIL! rc= 7 at Wed Jan 22 18:33:04 MST 2014
053 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...........................FAIL! rc= 7 at Wed Jan 22 18:37:35 MST 2014
055 bl376 TBL.sh fm1.9c4dh+testmech outfrq3s 9s ...........................FAIL! rc= 7 at Wed Jan 22 18:44:54 MST 2014
058 bl374 TBL.sh fs1.9c4dh outfrq3s+1850-2005_cam4_super_fast_llnl 9s .....FAIL! rc= 7 at Wed Jan 22 18:55:12 MST 2014
060 bl430 TBL.sh fm1.9c5dh outfrq3s_NEUwetdep 9s ..........................FAIL! rc= 7 at Wed Jan 22 19:05:38 MST 2014
063 bl440 TBL.sh fsoa1.9c4dh outfrq3s+soa_chem_megan_emis 9s ..............FAIL! rc= 7 at Wed Jan 22 19:26:59 MST 2014
067 bl387 TBL.sh f1.9c4dm outfrq3s+1850-2005_cam4 9s ......................FAIL! rc= 7 at Wed Jan 22 19:39:13 MST 2014
070 bl388 TBL.sh f1.9c4m outfrq24h+1850_cam4 2d ...........................FAIL! rc= 7 at Wed Jan 22 19:41:58 MST 2014
073 bl379 TBL.sh wm1.9c4h outfrq24h+waccm_1850_cam4 2d ....................FAIL! rc= 7 at Wed Jan 22 19:58:22 MST 2014
076 bl731 TBL.sh h16c3aqdh aqua+aquaplanet_cam3 9s ........................FAIL! rc= 7 at Wed Jan 22 20:02:49 MST 2014
079 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ....................FAIL! rc= 7 at Wed Jan 22 20:19:54 MST 2014
081 bl740 TBL.sh h30c5h outfrq3s 9s .......................................FAIL! rc= 7 at Wed Jan 22 20:23:06 MST 2014
084 bl751 TBL.sh h30c4aqdm outfrq3s+aquaplanet_cam4 9s ....................FAIL! rc= 7 at Wed Jan 22 20:40:08 MST 2014
frankfurt/nag:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Wed Jan 22 14:43:08 MST 2014
009 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Wed Jan 22 14:49:33 MST 2014
013 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Wed Jan 22 14:58:37 MST 2014
017 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Wed Jan 22 15:26:40 MST 2014
025 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Wed Jan 22 15:53:26 MST 2014
027 bl313 TBL.sh f10c3aqdm outfrq3s+aquaplanet_cam3 3s ....................FAIL! rc= 7 at Wed Jan 22 15:54:48 MST 2014
030 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7 at Wed Jan 22 16:07:12 MST 2014
032 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 7 at Wed Jan 22 16:09:36 MST 2014
035 bl318 TBL.sh f10c4cdm sat_hist 9s .....................................FAIL! rc= 7 at Wed Jan 22 16:26:25 MST 2014
038 bl324 TBL.sh f4c4rrtmgdm outfrq3s 9s ..................................FAIL! rc= 7 at Wed Jan 22 16:35:26 MST 2014
041 bl805 TBL.sh f4c5carmtt1dm outfrq3s_carma 9s ..........................FAIL! rc= 7 at Wed Jan 22 17:10:20 MST 2014
frankfurt/pgi or jaguar/pgi:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Wed Jan 22 14:42:53 MST 2014
010 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Wed Jan 22 14:51:44 MST 2014
012 bl113 TBL.sh e8c5paqdm aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Wed Jan 22 14:57:17 MST 2014
016 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Wed Jan 22 15:04:41 MST 2014
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Wed Jan 22 15:20:23 MST 2014
031 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Wed Jan 22 15:56:52 MST 2014
033 bl313 TBL.sh f10c3aqdm outfrq3s+aquaplanet_cam3 3s ....................FAIL! rc= 7 at Wed Jan 22 16:01:35 MST 2014
036 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7 at Wed Jan 22 16:23:41 MST 2014
039 bl315 TBL.sh f10c3dm outfrq3s+cam3 9s .................................FAIL! rc= 7 at Wed Jan 22 16:29:37 MST 2014
041 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 7 at Wed Jan 22 16:37:16 MST 2014
044 bl317 TBL.sh f10c5cdm outfrq3s 9s .....................................FAIL! rc= 7 at Wed Jan 22 16:54:05 MST 2014
047 bl319 TBL.sh fn10c5dm outfrq3s_bam 9s .................................FAIL! rc= 7 at Wed Jan 22 17:00:47 MST 2014
050 bl320 TBL.sh fn10c5dm rad_diag 9s .....................................FAIL! rc= 7 at Wed Jan 22 17:04:38 MST 2014
053 bl321 TBL.sh f10c5cdm atrain 9s .......................................FAIL! rc= 7 at Wed Jan 22 17:13:05 MST 2014
056 bl420 TBL.sh f7mode4c5dm outfrq3s 9s ..................................FAIL! rc= 7 at Wed Jan 22 17:43:31 MST 2014
059 bl734 TBL.sh hn16c5aqdm outfrq3s_bam+aquaplanet_cam5 9s ...............FAIL! rc= 7 at Wed Jan 22 18:08:46 MST 2014
ZM modifications expected to change answers (but not climate) for all
runs except adiabatic and ideal physics.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: All CAM3/4/5.
- what platforms/compilers: All.
- nature of change (roundoff; larger than roundoff but same climate; new
climate): Larger than roundoff, same climate.
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_23
Originator(s): Po-Lun Ma, eaton
Date: Thu Jan 9 13:37:51 MST 2014
One-line Summary: changes to defaults for soil_erod and dust_emis_fact
Purpose of changes:
A problem with too low dust burdens in high res (1/2 and 1/4 deg) cam5 runs
was traced to the soil_erod datasets being used at those resolutions. The
proposed solution involves using the 2 deg (1.9x2.5) dataset at all
resolutions higher than 2 deg. Since we have already tuned the 1 deg FV
runs using the 0.9x1.25 dataset, an exception will be made for that case to
maintain backwards compatibility. There are also several low resolution
specific versions of soil_erod that will be maintained for backwards
compatibility (and because there is no known problem with them).
It is unnecessary to produce specific high resolution version of soil_erod by
interpolating the 1.9x2.5 dataset since CAM is already set up to do this on
the fly internally (this was done to support the SE dycore and not require
resolution specific datasets for all the SE grids). Therefore the fix is
to simply modify the build-namelist defaults so that the 1.9x2.5 dataset is
used for all dycores at all resolutions except the 1 deg FV case, and for
the existing low res specific datasets.
There is also a change to the dust_emis_fact tuning parameter. Po-Lun Ma
has done simulations showing that the dust burdens for the 1/2 and 1/4 deg
FV case is improved by using the value 0.45 (the value from the tuning with
the 1 deg soil_erod dataset is set to 0.35). This change has been included
as well.
The SE setting for dust_emis_fact has not been changed from the value of
0.55 which was from tuning ne30np4 using the 1 deg soil_erod dataset.
Since the tuning of CAM-SE is ongoing we decided to change all the SE
resolutions to use the 1.9x2.5 soil erod dataset recognizing that
dust_emis_fact will need retuning for SE.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist:
. defaults changed for soil_erod and dust_emis_fact as described above.
List any changes to the defaults for the boundary datasets:
. defaults changed for soil_erod as described above.
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. updates to soil_erod defaults
. updates to dust_emis_fact
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all PASS except:
081 bl740 TBL.sh h30c5h outfrq3s 9s .......................................FAIL! rc= 7 at Thu Jan 9 01:31:35 MST 2014
All SE w/ prognostic aerosols runs change answers.
frankfurt/nag: all PASS except:
025 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Wed Jan 8 18:46:44 MST 2014
frankfurt/pgi: all PASS except:
031 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Wed Jan 8 19:38:10 MST 2014
bl312 fails becase the default dataset when no resolution dependent version
of soil_erod is now the 1.9x2.5 version. This resolution was previously
using 0.9x1.25.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers:
Answers will change for all SE w/ prognostic aerosol runs.
Answers will change for FV w/ prognostic aerosols for resolutions finer
than 1 deg, or for low resolutions that don't already have resolution
specific version of soil_erod.
===============================================================
===============================================================
Tag name: cam5_3_22
Originator(s): cbardeen, santos
Date: 2014/01/02
One-line Summary: New CARMA updates and compiler porting
Purpose of changes:
- Add six new CARMA models:
- cirrus_dust
- meteor_impact
- mixed_sulfate
- pmc_sulfate
- tholin
- test_tracers2
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
- New CPP macro, N_RAD_CNST, sets the maximum allowed number of
radiatively active constituents.
- Work around new NAG fpp bug by turning off macro expansion in comments.
Describe any changes made to the namelist:
- Add defaults for the new CARMA models.
- New carma_emis_* variables for the meteor impact model.
- New carma_neutral_h2so4 flag.
- The types in namelist_definitions.xml can now contain arrays with
variable dimensions.
- The variable n_rad_cnst was introduced, and controls the size of the
rad_climate and rad_diag_* arrays.
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
A models/atm/cam/src/physics/carma/cam/carma_getH2O.F90
A models/atm/cam/src/physics/carma/cam/carma_getH2SO4.F90
- Create constituent profiles from averages over columns.
A models/atm/cam/src/physics/carma/models/cirrus_dust
A models/atm/cam/src/physics/carma/models/cirrus_dust/carma_mod.F90
A models/atm/cam/src/physics/carma/models/cirrus_dust/carma_model_mod.F90
A models/atm/cam/src/physics/carma/models/cirrus_dust/hetnucl.F90
A models/atm/cam/src/physics/carma/models/cirrus_dust/growevapl.F90
A models/atm/cam/src/physics/carma/models/cirrus_dust/carma_model_flags_mod.F90
A models/atm/cam/src/physics/carma/models/cirrus_dust/carma_cloudfraction.F90
- Add CARMA cirrus_dust model.
A models/atm/cam/src/physics/carma/models/meteor_impact
A models/atm/cam/src/physics/carma/models/meteor_impact/carma_model_mod.F90
A models/atm/cam/src/physics/carma/models/meteor_impact/carma_model_flags_mod.F90
- Add CARMA meteor_impact model.
A models/atm/cam/src/physics/carma/models/mixed_sulfate
A models/atm/cam/src/physics/carma/models/mixed_sulfate/carma_model_mod.F90
A models/atm/cam/src/physics/carma/models/mixed_sulfate/carma_model_flags_mod.F90
- Add CARMA mixed_sulfate model.
A models/atm/cam/src/physics/carma/models/pmc_sulfate
A models/atm/cam/src/physics/carma/models/pmc_sulfate/carma_model_mod.F90
A models/atm/cam/src/physics/carma/models/pmc_sulfate/carma_model_flags_mod.F90
- Add CARMA pmc_sulfate model.
A models/atm/cam/src/physics/carma/models/test_tracers2
A models/atm/cam/src/physics/carma/models/test_tracers2/carma_model_mod.F90
A models/atm/cam/src/physics/carma/models/test_tracers2/carma_model_flags_mod.F90
- Add CARMA test_tracers2 model.
A models/atm/cam/src/physics/carma/models/tholin
A models/atm/cam/src/physics/carma/models/tholin/carma_model_mod.F90
A models/atm/cam/src/physics/carma/models/tholin/carma_model_flags_mod.F90
- Add CARMA tholin model.
A models/atm/cam/test/system/config_files/f1.9c5carmc_ddm
A models/atm/cam/test/system/config_files/f4c5carmm_idm
A models/atm/cam/test/system/config_files/f4c5carmm_sdm
A models/atm/cam/test/system/config_files/f4c5carmp_sdm
A models/atm/cam/test/system/config_files/f4c5carmthodm
A models/atm/cam/test/system/config_files/f4c5carmtt1dm
A models/atm/cam/test/system/config_files/f4c5carmtt2dm
- Config files for tests of new models.
A models/atm/cam/test/system/config_files/f4c4rrtmgdm
- Test case for CAM4+RRTMG.
A models/atm/cam/test/system/nl_files/outfrq3s_carma
A models/atm/cam/test/system/nl_files/outfrq3s_carma_fractal
- Add namelist for a outfrq3s test for CARMA, and CARMA with
fractal soot enabled.
List all existing files that have been modified, and describe the changes:
NOTE: There are changes for porting to the NAG compiler, as well as
cosmetic changes (such as whitespace), throughout the CARMA
modules. The most common such change is to wrap lines that would
otherwise be longer than the Fortran standard's limit (132
characters.) These changes will not be listed individually.
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
- New CARMA external.
M models/atm/cam/bld/build-namelist
M models/atm/cam/bld/config_files/definition.xml
M models/atm/cam/bld/configure
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- Add settings for new CARMA models and features.
- Allow the user to increase the maximum allowed number of
radiatively active (or diagnosed) constituents, using the new
option "-max_n_rad_cnst".
M models/atm/cam/bld/perl5lib/Build/Namelist.pm
M models/atm/cam/bld/perl5lib/Build/NamelistDefinition.pm
- Allow namelist_definitions.xml to specify a variable as an array
dimension (for n_rad_cnst).
M models/atm/cam/bld/Makefile.in
- For NAG, don't expand macros in comments (workaround for fpp
regression).
M models/atm/cam/src/physics/cam/micro_mg_cam.F90
- Don't add rel_fn to the physics buffer, since it's only used for
output from MG itself.
M models/atm/cam/src/physics/cam/phys_prop.F90
- Remove limit on maximum number of unique input files, by using
an allocatable to store the list.
M models/atm/cam/src/physics/cam/physics_types.F90
- Allow grid cells to have negative altitude. (Fixes issue with
state_debug_checks and negative topology.)
M models/atm/cam/src/physics/cam/rad_constituents.F90
- Maximum number of radiative constituents now comes from the
parameter n_rad_cnst, set from the macro N_RAD_CNST.
M models/atm/cam/src/physics/cam/radiation_data.F90
- Don't output rel_fn; it can be output by MG, and is not actually
an input the radiation uses in the first place.
M models/atm/cam/src/physics/carma/cam/carma_getT.F90
- Spelling correction.
M models/atm/cam/src/physics/carma/cam/carma_intr.F90
- Create and use new H2O/H2SO4 profiles.
- Correct indexing errors.
- Changes for new fractal code.
- Fix fairly severe error where "nf90_get_var" was being used
instead of "nf90_put_var", leading to outputs of files
containing uninitialized data.
M models/atm/cam/src/physics/carma/models/cirrus/carma_model_mod.F90
- Fix bug where state was being inappropriately modified by calls
to MG routines.
M models/atm/cam/src/physics/carma/models/sea_salt/carma_model_mod.F90
- Address case where Weibull_k is 0 differently.
- Removed prescribed Weibull coefficients entirely from this file,
since this feature has not been completed, and the incomplete
pieces broke the NAG build.
M models/atm/cam/src/physics/carma/models/bc_strat/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/dust/carma_model_flags_mod.F90
M models/atm/cam/src/physics/carma/models/dust/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/meteor_smoke/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/pmc/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/sea_salt/carma_model_flags_mod.F90
M models/atm/cam/src/physics/carma/models/sulfate/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/test_detrain/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/test_growth/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/test_passive/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/test_radiative/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/test_swelling/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/test_tracers/carma_model_flags_mod.F90
M models/atm/cam/src/physics/carma/models/test_tracers/carma_model_mod.F90
- Only NAG port and cosmetic changes.
M models/atm/cam/src/physics/rrtmg/rad_solar_var.F90
- Removed unnecessary use of ssi_ref.
M models/atm/cam/src/physics/rrtmg/radiation.F90
- Don't retrieve REL, REL_FN, and REI from pbuf, because they are
never used.
M models/atm/cam/test/system/TER.sh
- For restart tests, copy RRTMG optical property files that have
been written out by CARMA.
D models/atm/cam/test/system/config_files/fsd1.9c4wcarmsulm
A models/atm/cam/test/system/config_files/fsd1.9c4wcarmsuldm
- Change SD-WACCM CARMA sulfate test to a debug mode test.
M models/atm/cam/test/system/input_tests_master
- Add tests for CAM4+RRTMG and new CARMA models.
- Dust and SD-WACCM tests are now 9 timestep debug tests.
M models/atm/cam/test/system/nl_files/outfrq3s_carma
- Remove year specification (it messed up the SD test, which
doesn't have data for 1995).
M models/atm/cam/test/system/test_driver.sh
- Explicitly use netCDF 4.3 on Frankfurt/NAG, since the symlink
still points to a version where getting text attributes will
corrupt data on the stack.
M models/atm/cam/test/system/tests_carma
- Add new CARMA model tests to this list.
M models/atm/cam/test/system/tests_pretag_frankfurt_nag
- Add test for CAM4+RRTMG.
- Add test for CARMA test_tracers.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
All PASS.
frankfurt/nag:
038 bl324 TBL.sh f4c4rrtmgdm outfrq3s 9s ..................................FAIL! rc= 7 at Thu Jan 2 11:15:23 MST 2014
041 bl805 TBL.sh f4c5carmtt1dm outfrq3s_carma 9s ..........................FAIL! rc= 7 at Thu Jan 2 11:15:24 MST 2014
Failures for new tests where there is no baseline.
frankfurt/pgi or jaguar/pgi:
All PASS.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
All configurations with CARMA.
- what platforms/compilers:
All.
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
Mixed, several likely to have new climate. No changes to compsets that are
scientifically supported within CESM; this is only new CARMA development.
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_21
Originator(s): cacraig
Date: 12/20/13
One-line Summary: Update yellowstone intel compiler to 13.1.2 to allow implementation of eq tests for CESM
Purpose of changes:
- Have CAM standalone tests on yellowstone use Intel 13.1.2 to allow CESM eq tests to be BFB (compiler version
CESM is currently using)
- Modified CESM regression tests on yellowstone to be eq tests for CAM4 FV and CAM5 SE
- Added missing finidat file for ne30np4 case (answers were not BFB with CESM without it)
- Updated to alpha06d externals (except CLM which has a bug) to eliminate core dump in pio
- Performed mergeinfo cleanup (svn propdel svn:mergeinfo -R; svn revert .)
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/TCB_ccsm.sh
- Added echo of all CESM commands for documentation purposes
M models/atm/cam/test/system/TSM.sh
- Added echo of command sent to TCB.sh for documentation purposes
M models/atm/cam/test/system/test_driver.sh
- Force yellowstone intel compilations to be 13.1.2 to match CESM compiler and allow eq tests to be BFB
M models/atm/cam/test/system/tests_pretag_yellowstone
M models/atm/cam/test/system/input_tests_master
- Modified CESM tests on yellowstone to be eq tests for CAM4 FV and CAM5 SE
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- Added missing finidat file for ne30np4 case (answers were not BFB with CESM without it)
M models/atm/cam/src/dynamics/sld/inidat.F90
M models/atm/cam/src/dynamics/eul/inidat.F90
M models/atm/cam/src/dynamics/fv/inidat.F90
- Change trunits to character*256 to get rid of bounds error in sm374 and sm430 with Intel 13.1.2
M SVN_EXTERNAL_DIRECTORIES
- Updated to alpha06d externals to eliminate core dump in pio
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
********* SPECIAL NOTES ON TESTING **************
** Due to time constraints due to the upcoming holidays,
** yellowstone tests were run without changes to inidat.
** The change to inidat allowed sm374 and sm430 to run to
** completion and these two group of tests were rerun.
** Other yellowstone tests were not run with this inidat
** change on yellowstone.
*************************************************
yellowstone/intel: All BFB except,
036 bl336 TBL.sh wm4c4h outfrq3s+waccm_1850_cam4 9s .......................FAIL! rc= 7 at Mon Dec 16 18:23:43 MST 2013
053 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...........................FAIL! rc= 7 at Mon Dec 16 19:41:43 MST 2013
070 bl388 TBL.sh f1.9c4m outfrq24h+1850_cam4 2d ...........................FAIL! rc= 7 at Mon Dec 16 20:25:32 MST 2013
073 bl379 TBL.sh wm1.9c4h outfrq24h+waccm_1850_cam4 2d ....................FAIL! rc= 7 at Mon Dec 16 20:42:02 MST 2013
081 bl740 TBL.sh h30c5h outfrq3s 9s .......................................FAIL! rc= 7 at Mon Dec 16 21:06:26 MST 2013
- Answer changes due to change in compiler
frankfurt/nag: All BFB and run with new inidat
frankfurt/pgi or jaguar/pgi: All BFB and run with new inidat
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_20
Originator(s): hannay, pel, mataylo, eaton
Date:
One-line Summary: update SE dycore tuning parameters and mct_libdir option
Purpose of changes:
. Update SE dycore tuning parameters
- Change time stepping method to RK5 (Kinnmark & Gray Runga-Kutta 5
stage; 3rd order accurate in time)
- Set the namelists variables as recommended for RK5 in:
http://www.cgd.ucar.edu/cms/pel/software/cam-se-dt-table.pdf
- Add "tstep_type" namelist option for SE dycore
- Turn on the FV energy fixer (This will be updated later on with
Williamson an Olson enery fixer)
- Remove the variable "energy_fixer" from the cam namelist.
To update a namelist created by an earlier CAM tag for an SE run on
grid ne30np4 one should make the following changes:
- The default settings for CESM1.2.0 were:
tstep_type =1
hypervis_subcycle = 2
qsplit =4
rsplit = 5
se_ftype =1
se_nsplit = 1
- The new recommended settings are:
tstep_type =5
hypervis_subcycle = 3
qsplit = 1
rsplit = 3
se_ftype =0
se_nsplit = 2
These changes have been approved by the AMWG co-chairs:
Rich Neale: 11/11/2013
Minghua Zhang: 11/12/2013
Mark Taylor: 11/12/2013
. Allow the CAM standalone mode to build the MCT libs outside the CAM build
directory so that they can be reused for multiple CAM builds. The
location of the MCT build is specified using the -mct_libdir argument.
If the libraries are present in the location specified by -mct_libdir
then the MCT configure and build do not need to be redone.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist:
. add tstep_type, remove energy_fixer, and change many default SE tuning
parameters
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/build-namelist
. remove default for energy_fixer
. add default for tstep_type
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. update defaults for dtime, nu, nu_p, nu_div, hypervis_subcycle, qsplit,
rsplit, se_nsplit, se_ftype
. remove energy_fixer default
. add default for tstep_type
models/atm/cam/bld/namelist_files/namelist_definition.xml
. add namelist variable tstep_type for SE timestepping method
. remove entry for energy_fixer
models/atm/cam/src/physics/cam/physpkg.F90
. tphysbc and phys_run1_adiabatic_or_ideal
- Add SE to the conditional for running the check_energy_fix global
energy fixer (also used by FV).
models/atm/cam/test/system/TCB.sh
models/atm/cam/bld/configure
. Added changes to allow mct to be built once and used for all tests
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB except
076 bl731 TBL.sh h16c3aqdh aqua+aquaplanet_cam3 9s ........................FAIL! rc= 7 at Thu Dec 5 20:45:41 MST 2013
079 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ....................FAIL! rc= 7 at Thu Dec 5 21:02:21 MST 2013
081 bl740 TBL.sh h30c5h outfrq3s 9s .......................................FAIL! rc= 7 at Thu Dec 5 21:05:37 MST 2013
084 bl751 TBL.sh h30c4aqdm outfrq3s+aquaplanet_cam4 9s ....................FAIL! rc= 7 at Thu Dec 5 21:22:22 MST 2013
090 bl997 TBL_ccsm.sh ne16_g37 FC5 2d .....................................FAIL! rc= 7 at Thu Dec 5 21:40:24 MST 2013
frankfurt/nag: All BFB
frankfurt/pgi or jaguar/pgi: all BFB except
059 bl734 TBL.sh hn16c5aqdm outfrq3s_bam+aquaplanet_cam5 9s ...............FAIL! rc= 7 at Thu Dec 5 19:24:50 MST 2013
SE tests were answer changing and approved by the AMWG co-chairs
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: All SE configurations have answer
changes. The climate is similar for ne30np4.
URL for AMWG diagnostics output used to validate new climate:
> http://webext.cgd.ucar.edu/FCLIMO/f.e12.F2000C5.ne30_ne30.dyn.001/atm/f.e12.F2000C5.ne30_ne30.dyn.001-f.e12.F2000C5.ne30_ne30.cnt.001/
===============================================================
===============================================================
Tag name: cam5_3_19
Originator(s): Cheryl Craig, Jen Kay, Jay Shollenberger, Jim Edwards, Brian Eaton
Date: 11/21/13
One-line Summary: Misc updates
Purpose of changes:
- update a use case
- updated SVN_EXTERNAL_DIRECTORIES to match alpha06c (gets the fix for Machines)
Bugs fixed (include bugzilla ID):
- 1859: bug inside physics_buffer: linked list should be advanced from 2,index instead of 1,index
- 1864: bug in lidar simulator fixed
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/namelist_files/use_cases/2000_cam4_trop_strat_soa.xml
- Changed SSLT01PP, SSLT02PP, SSLT03PP and SSLT04PP to
SSLT01WET, SSLT02WET, SSLT03WET and SSLT04WET
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
- bug fix from jenkay for lidar simulator in cosp
M models/atm/cam/src/control/cam_history.F90
- bug fix from edwards to _Filltype type written to history restart file
did not match the data type when history files were set to real in netCDF6.
M models/atm/cam/src/physics/cam/physics_buffer.F90
M models/atm/cam/src/physics/cam/physics_buffer.F90.in
- bug fix from eaton. In physics_buffer, pbuf_get_field_name the loop over the
linked list was one too many. Now loops from 2,index.
M models/atm/cam/test/system/tests_pretag_frankfurt_pgi
- Removed tests sm991 and eq991 per Brian Eaton
His feeling is that CESM testing should be done by the CESM team and this test is redundant
The CESM team is seeing this error as well.
M SVN_EXTERNAL_DIRECTORIES
- Needed new machines external to fix CESM regression tests. This tag
appears to depend on other tags (CESM test would not build). Updated
all externals to ones used in alpha06c
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB
frankfurt/nag: all BFB
frankfurt/pgi or jaguar/pgi: all BFB except:
044 bl317 TBL.sh f10c5cdm outfrq3s 9s .....................................FAIL! rc= 7 at Thu Nov 21 15:46:21 MST 2013
The cosp test was expected to be different
060 sm991 TSM_ccsm.sh f19_f19 F 2d ........................................FAIL! rc= 4 at Thu Nov 21 17:00:26 MST 2013
061 eq991 TEQ_ccsm.sh f19_f19 F f1.9c4m fcase 2d ..........................FAIL! rc= 4 at Thu Nov 21 17:00:26 MST 2013
NOTE -- These tests are permanently removed per Brian Eaton
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_18
Originator(s): cacraig
Date: 11/20/13
One-line Summary: Update externals to CESM1_3_beta05
Purpose of changes: Update externals to CESM1_3_beta05
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M SVN_EXTERNAL_DIRECTORIES
- Update externals to CESM1_3_beta05 except for csm_share.
beta05 used share3_130918 for csm_share, but CAM requires
share3_131101 which made gamma elemental and is required in micro_mg_cam.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All BFB except
085 sm992 TSM_ccsm.sh f19_g16 ETEST 2d ....................................FAIL! rc= 6 at Mon Nov 18 20:43:08 MST 2013
086 er992 TER_ccsm.sh f19_g16 ETEST 1+1d ..................................FAIL! rc= 6 at Mon Nov 18 20:43:08 MST 2013
087 bl992 TBL_ccsm.sh f19_g16 ETEST 2d ....................................FAIL! rc= 4 at Mon Nov 18 20:43:08 MST 2013
088 sm997 TSM_ccsm.sh ne16_g37 FC5 2d .....................................FAIL! rc= 6 at Mon Nov 18 21:13:31 MST 2013
089 er997 TER_ccsm.sh ne16_g37 FC5 1+1d ...................................FAIL! rc= 6 at Mon Nov 18 21:13:31 MST 2013
090 bl997 TBL_ccsm.sh ne16_g37 FC5 2d .....................................FAIL! rc= 4 at Mon Nov 18 21:13:31 MST 2013
091 ne996 TNE_ccsm.sh f45_f45 FMOZ FMOZ+testmech 2d .......................FAIL!
These CESM tests failed due to a bug in scripts/ccsm_utils/Machines/mkbatch.yellowstone. Jim provided a fix which I tested
and the tests all passed and were BFB. This fix will be in his next machines tag and will be incorporated in cam5_3_19.
frankfurt/nag: All BFB
frankfurt/pgi or jaguar/pgi: All BFB
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_17
Originator(s): santos
Date: 13/11/04
One-line Summary: Compiler fixes, prep work for new MG substepping.
Purpose of changes:
- Bring in changes from the MG microphysics development branch, to make it
easier to merge the full set of changes with the new subcolumn framework
later.
This includes micro_mg_utils, which provides routines that implement
part of the core functionality of MG version 2 (coming soon).
- Fix issues identified by NAG.
- Fix issue due to missing CMELIQ output in the CLUBB interface.
- Address other bugs that show up in rare cases.
Bugs fixed (include bugzilla ID):
- #1813: CMELIQ no longer able to be output
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
A models/atm/cam/src/physics/cam/micro_mg_utils.F90
- Module providing utility code and individual process subroutines.
Part of MG 2.0 will be based on this module.
List all existing files that have been modified, and describe the changes:
M SVN_EXTERNAL_DIRECTORIES
- Update csm_share to a version where gamma is an elemental
function.
M models/atm/cam/bld/cam.buildnml.csh
- Turn on state_debug_checks if DEBUG is TRUE.
M models/atm/cam/src/chemistry/mozart/apex_subs.F90
- Send array sections to TRILIN rather than individual elements
(i.e. pointers), to satisfy NAG checks that error due to the
dummy array exceeding the bounds of the actual argument array.
M models/atm/cam/src/chemistry/mozart/spedata.F90
M models/atm/cam/src/dynamics/fv/metdata.F90
- Nullify pointers and take steps to reduce the chance of an
uninitialized pointer reference.
M models/atm/cam/src/chemistry/utils/modal_aero_wateruptake.F90
- Change modal_aero_kohler to use assumed shape arrays to avoid a
pointless copy.
M models/atm/cam/src/control/interpolate_data.F90
- Use new integer flag in lininterp_full1d instead of a literal 1.
- Remove inappropriate check that causes failure in RRTMG for
serial cases that are not SCAM runs.
M models/atm/cam/src/control/wrap_nf.F90
- Remove arbitrary addition of "1" to a dimension; this has no
clear purpose and was causing issues for CARMA with strict
compiler checks.
M models/atm/cam/src/physics/cam/clubb_intr.F90
- Add missing addfld/outfld calls for CMELIQ.
M models/atm/cam/src/physics/cam/micro_mg1_0.F90
- Remove unnecessary output "cldo".
- Rearrange statements.
M models/atm/cam/src/physics/cam/micro_mg_cam.F90
- Reduce the number of differences with respect to the MG 2.0
development branch, including statement reordering, cleanup,
whitespace, and output changes.
M models/atm/cam/src/physics/cam/tropopause.F90
- Switch to assumed-shape arrays for twmo interface, avoiding an
unnecessary copy, and also change the name "gamma" to "gam" to
avoid any name clash with the F2008 intrinsic.
M models/atm/cam/src/physics/carma/models/cirrus/carma_model_mod.F90
- Remove routines that were duplicated from an earlier version of
the MG microphysics, and instead use versions from the new
micro_mg_utils module.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS.
frankfurt/nag: All PASS.
frankfurt/pgi or jaguar/pgi: All PASS.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
Changes answers for CARMA cirrus model only.
- what platforms/compilers:
All.
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
Roundoff.
If bitwise differences were observed, how did you show they were no worse
than roundoff?
Offline testing of the size distribution routines that caused the answer
change.
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_16
Originator(s): fvitt
Date: 21 Oct 2013
One-line Summary: Refactor the chemistry-aerosol model interface and update drv tag.
Purpose of changes:
- The chemistry-aerosol model interface was refactored to provide a more
extendable framework. This will ease incorporation of other aerosol
models (e.g., a sectional aerosol model)
This refactoring of the chemistry-aerosol model interface does away with the numerous
ifdefs that have been historically used to embed aerosol models within CAM-Chem.
Each aerosol model must provide the following modules in chemistry/name_aero (where
name is a named aerosol model, e.g., modal):
aero_model.F90 ------- aerosol model specific module with methods:
. aero_model_drydep ----- aerosol dry deposition and sediment
. aero_model_wetdep ----- aerosol wet removal
. aero_model_emissions -- aerosol emissions
. aero_model_surfarea --- aerosol surface area for chemistry
. aero_model_gasaerexch - create, grow, change, and shrink aerosols
dust_model.F90 ------- dust module with methods
. dust_names - dust species names
. dust_nbin -- number of dust bins
. dust_emis -- emissions method
seasalt_model.F90 ---- sea salt module with methods
. seasalt_names - sea salt species names
. seasalt_nbin -- number of sea salt bins
. seasalt_emis -- sea salt emissions method
sox_cldaero_mod.F90 -- aerosol model specific code for cloud aerosol/SOx aqueous chemistry
- Update drv tag to drvseq5_0_03 which fixes SCAM test
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
D models/atm/cam/bld/namelist_files/master_mam_dep_list.xml
D models/atm/cam/bld/namelist_files/master_drydep_list.xml
- re-organized master deposition list files
D models/atm/cam/src/chemistry/bulk_aero/dust_intr.F90
D models/atm/cam/src/chemistry/bulk_aero/aerosol_intr.F90
D models/atm/cam/src/chemistry/bulk_aero/progseasalts_intr.F90
- removed
D models/atm/cam/src/chemistry/bulk_aero/drydep_mod.F90
D models/atm/cam/src/chemistry/bulk_aero/wetdep.F90
D models/atm/cam/src/chemistry/bulk_aero/dust_sediment_mod.F90
- moved to chemistry/aerosol directory
D models/atm/cam/src/chemistry/mozart/mo_setsox.F90
- moved to chemistry/aerosol directory
D models/atm/cam/src/chemistry/mozart/mo_aerosols.F90
D models/atm/cam/src/chemistry/mozart/mo_setsoa.F90
- moved to chemistry/bulk_aero
D models/atm/cam/src/chemistry/mozart/mz_aerosols_intr.F90
- removed
List all subroutines added and what they do:
A models/atm/cam/bld/namelist_files/master_aer_drydep_list.xml
A models/atm/cam/bld/namelist_files/master_gas_drydep_list.xml
- re-organized master deposition list files
A models/atm/cam/src/chemistry/modal_aero/aero_model.F90
A models/atm/cam/src/chemistry/modal_aero/seasalt_model.F90
A models/atm/cam/src/chemistry/modal_aero/dust_model.F90
- modal aerosol specific code
A models/atm/cam/src/chemistry/modal_aero/sox_cldaero_mod.F90
- modal aerosol specific code for aqueous chemistry
A models/atm/cam/src/chemistry/bulk_aero/mo_aerosols.F90
- moved from chemistry/mozart
A models/atm/cam/src/chemistry/bulk_aero/aerosol_depvel.F90
A models/atm/cam/src/chemistry/bulk_aero/aero_model.F90
A models/atm/cam/src/chemistry/bulk_aero/mo_setsoa.F90
A models/atm/cam/src/chemistry/bulk_aero/seasalt_model.F90
A models/atm/cam/src/chemistry/bulk_aero/dust_model.F90
A models/atm/cam/src/chemistry/bulk_aero/sox_cldaero_mod.F90
- module for the bulk aerosol model
A models/atm/cam/src/chemistry/aerosol/mo_setsox.F90
A models/atm/cam/src/chemistry/aerosol/dust_sediment_mod.F90
- moved from chemistry/mozart
A models/atm/cam/src/chemistry/aerosol/drydep_mod.F90
A models/atm/cam/src/chemistry/aerosol/dust_common.F90
A models/atm/cam/src/chemistry/aerosol/cldaero_mod.F90
A models/atm/cam/src/chemistry/aerosol/wetdep.F90
A models/atm/cam/src/chemistry/aerosol
- utility modules common to all aerosol models
A models/atm/cam/src/chemistry/aerosol/soil_erod_mod.F90
- extracted code from deprecated dust_intr.F90 module
A models/atm/cam/src/chemistry/aerosol/sslt_sections.F90
- extracted code from deprecated progseasalts module
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/TR8.sh
- check for r8 kind in source directory src/cpl rather than the multiple
src/cpl_* directories that existed prior to cam3_5_15
M models/atm/cam/bld/configure
- change in Filepath directory structure for the aerosol models
M models/atm/cam/bld/namelist_files/master_aer_wetdep_list.xml
- re-organized master deposition list files
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- sol_farcti_cloud_borne, aerosol_drydep_list, and aer_wetdep_list
are now part of the aerosol_nl namelist group
- dust_emis_fact moved to dust_nl namelist group
- "soil_erod" --> "soil_erod_file" and is included in dust_nl namelist group
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- "soil_erod" --> "soil_erod_file"
M models/atm/cam/bld/build-namelist
- refactored how deposition lists are set
- "soil_erod" --> "soil_erod_file"
- set sol_facti_cloud_borne only if chemistry has modal aerosols
M models/atm/cam/bld/perl5lib/Build/ChemNamelist.pm
- refactored how deposition lists are set
M models/atm/cam/bld/namelist_files/use_cases/cam4_chem_radpsv_geos5.xml
M models/atm/cam/bld/namelist_files/use_cases/cam4_bam_radpsv_geos5.xml
M models/atm/cam/bld/namelist_files/use_cases/2000_cam4_trop_chem.xml
- removed ORO from fincl list
- "soil_erod" --> "soil_erod_file"
M models/atm/cam/bld/namelist_files/use_cases/soa_chem_megan_emis.xml
- *PP history fields replaced by *WET fields
M models/atm/cam/src/control/camsrfexch.F90
- extend cam_in_t for dust and MEGAN fluxes from land model
M models/atm/cam/src/cpl/atm_comp_mct.F90
M models/atm/cam/src/cpl/atm_comp_esmf.F90
M models/atm/cam/src/cpl/atm_import_export.F90
- set dust and MEGAN fluxes in the cam_in data structure
M models/atm/cam/src/physics/cam/micro_mg_cam.F90
- replaced MODAL_AERO ifdef with prog_modal_aero conditional
M models/atm/cam/src/physics/cam/clubb_intr.F90
- removed MODAL_AERO ifdef that served no purpose
M models/atm/cam/src/physics/cam/physpkg.F90
- add FRACIS to phys buffer
- register and initialize calsize and wateruptake for modal aerosol model
- removed calls to aerosol_reg and aerosol_init -- these are invoked by chemistry
- removed MODAL_AERO ifdefs
- invoke calcsize and wateruptake for prescribed modal aerosols
M models/atm/cam/src/physics/cam/qneg3.F90
- removed MODAL_AERO ifdefs and qneg3_modalx1 subroutine which was
essentially a duplication of the qneg3 subroutine
M models/atm/cam/src/physics/cam/vertical_diffusion.F90
- moved output of 'ustar' and 'obklen' history fields to this module
M models/atm/cam/src/chemistry/pp_none/chemistry.F90
- chem_register invokes aero_model_register for prescribed aerosols
- chem_init invokes aero_model_init for prescribed aerosols
- added stub chem_emissions subroutine to be consistent with mozart/chemistry.F90
M models/atm/cam/src/chemistry/utils/prescribed_aero.F90
- added comment
M models/atm/cam/src/chemistry/utils/mo_constants.F90
- added densities of dust and sea salt
M models/atm/cam/src/chemistry/utils/modal_aero_wateruptake.F90
- added register method which adds DGNUMWET, WETDENS_AP, and QAERWAT fields
to physics buffer
M models/atm/cam/src/chemistry/utils/modal_aero_calcsize.F90
- added register method which adds DGNUM field to physics buffer
M models/atm/cam/src/chemistry/modal_aero/modal_aero_rename.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_coag.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_gasaerexch.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_newnuc.F90
- removed latndx and lonndx arguments
M models/atm/cam/src/chemistry/modal_aero/modal_aero_initialize_data.F90
- moved addfld calls for aqueous chemistry to sox_cldaero_mod
- moved call to modal_aero_bcscavcoef_init to modal_aero/aero_model.F90
M models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
- removed the numerous MODAL_AERO ifdefs
- the bulk of the modal aerosol formation code moved to
aero_model_gasaerexch subroutine which include aqueous chem in setsox
- invocation of set_srf_emissions moved to chem_emissions subroutine
M models/atm/cam/src/chemistry/mozart/mo_setaer.F90
- reference sslt_names
M models/atm/cam/src/chemistry/mozart/mo_chemini.F90
- removed calls to aerosols_inti, soa_inti, and sox_int subroutines
M models/atm/cam/src/chemistry/mozart/mo_chm_diags.F90
- removed MODAL_AERO ifdefs
-- moved cloud-borne wet dep addfld calls to modal_aero/aero_model.F90
M models/atm/cam/src/chemistry/mozart/chemistry.F90
- chem_reset_fluxes --> chem_emissions
- aerosol namelist vars moved to aero_model
- general cleanup
M models/atm/cam/src/chemistry/mozart/mo_photo.F90
- reference seasalt_names
M models/atm/cam/src/chemistry/mozart/mo_sethet.F90
- determine if modal aerosols in initialization
M models/atm/cam/src/chemistry/mozart/mo_waccm_hrates.F90
- min size of 1 for col_dens and col_delta arguments
M models/atm/cam/src/chemistry/mozart/mo_mass_xforms.F90
- moved qqcw2vmr and vmr2qqcw to modal_aero/aero_model.F90
M models/atm/cam/src/chemistry/mozart/mo_usrrxt.F90
- removed surarea code -- is now in aero_model.F90
- removed MODAL_AERO ifdefs
M models/atm/cam/src/dynamics/sld/inidat.F90
M models/atm/cam/src/dynamics/eul/inidat.F90
M models/atm/cam/src/dynamics/se/inidat.F90
M models/atm/cam/src/dynamics/fv/inidat.F90
- removed calls to aerosol_implements_cnst (deprecated)
M SVN_EXTERNAL_DIRECTORIES
- update drv tag to drvseq5_0_03
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All pass
frankfurt/nag: All pass
frankfurt/pgi: All pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_15
Originator(s): cacraig, mvertens
Date: 10/17/13
One-line Summary: Merge in comptype03_cam5_3_10 and update to cesm1_3_beta04 externals
Purpose of changes:
- Merge in Mariana's changes in comptype03_cam5_3_10
svn merge $CAMREPO/trunk_tags/cam5_3_10 $CAMREPO/branch_tags/comptype_tags/comptype03_cam5_3_10
(plus additional changes below)
- Additional changes required:
cp /glade/u/home/mvertens/src/cesm1_3_beta04+comptype/models/atm/cam/src/utils/cam_dom/ocn_comp_mct.F90
svn del cpl_mct, cpl_mct_ocn_comp_mct.F90, cpl_esmf
svn add ocn_comp_mct.F90
cp /glade/u/home/mvertens/src/cescm1_3_beat04_comptype/models/atm/cam/bld/configure
Needed to modify models/atm/cam/src/cpl/atm_import_export.F90 to incorporate the changes from cam5_3_14
- Using cesm1_3_beta04 externals except for share which is using share3_130918 to get single column test to pass
- Added ability to CAM regression tests to use caldera for both compilation and running if job is started while logged on caldera
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
D models/atm/cam/src/cpl_share
D models/atm/cam/src/cpl_share/cam_cpl_indices.F90
D models/atm/cam/src/utils/cam_dom/cpl_mct
D models/atm/cam/src/utils/cam_dom/cpl_mct/ocn_comp_mct.F90
D models/atm/cam/src/utils/cam_dom/cpl_esmf
D models/atm/cam/src/utils/cam_aqua/cpl_mct
D models/atm/cam/src/utils/cam_aqua/cpl_mct/ocn_comp_mct.F90
D models/atm/cam/src/utils/cam_aqua/cpl_esmf
D models/atm/cam/src/cpl_mct
D models/atm/cam/src/cpl_mct/atm_comp_mct.F90
D models/atm/cam/src/cpl_esmf
D models/atm/cam/src/cpl_esmf/atm_comp_mct.F90
D models/atm/cam/src/cpl_esmf/atm_comp_esmf.F90
- Changes from the comptype branch merge
List all subroutines added and what they do:
A models/atm/cam/src/utils/cam_dom/ocn_comp_mct.F90
A + models/atm/cam/src/utils/cam_aqua/cpl
A + models/atm/cam/src/utils/cam_aqua/cpl/ocn_comp_mct.F90
A + models/atm/cam/src/cpl
A + models/atm/cam/src/cpl/atm_comp_mct.F90
A + models/atm/cam/src/cpl/atm_comp_esmf.F90
A + models/atm/cam/src/cpl/atm_import_export.F90
A + models/atm/cam/src/cpl/cam_cpl_indices.F90
- Changes from the comptype branch merge
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/test_driver.sh
M models/atm/cam/test/system/CAM_runcmnd.sh
- Added the ability to use caldera instead of yellowstone if job initiated on caldera
M models/atm/cam/bld/configure
- Changes from the comptype branch merge
M SVN_EXTERNAL_DIRECTORIES
- Using cesm1_3_beta04 externals except for share which is using share3_130918 to get single column test to pass
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB except:
092 sc001 TSC.sh e64c5bfbiop scm_prep scm64c5bfbiop scm_b4b_o1 7s .........FAIL! rc= 6 at Tue Oct 15 00:23:51 MDT 2013
- Not all of John Truesdale's changes made it into the driver tag. Mariana will be updating and will be in future tag.
frankfurt/nag:all BFB:
frankfurt/pgi: all BFB:
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_14
Originator(s): fvitt
Date: 30 Sep 2013
One-line Summary: Minor corrections to aerosol code and bug fixes
Purpose of changes:
This branch isolates the answer changing corrections uncovered
during the refactoring of chemistry/aerosol interface.
Answer changing features:
- Corrections to dust emissions to use unmodified dust fluxes
from the coupler each time step -- in time-steps where fluxes
from surface models were not updated modification of cam_in%cflx
were not representations of dust flux from surface models but the
modification of dust fluxes from the previous time-step. For example,
when the model advances from time step 0 to time step 1 it does not
couple with surface models and thus does not restore cflx with dust
fluxes from coupler.
- Removed kludge in ChemNamelist.pm that excluded H2SO4 from the
gas_wetdep_list namelist variable
- Register number density tracers with qmin of 1e-5 -- this allows
removal of special code in uwshcu for MAM
- Scale organic aerosol emissions off line for MAM
- Replace obsolete sulfchem routine with some chemistry for trop_bam
- Let chemistry output all aerosols in MMR units (kg/kg)
and gas species in VMR units (mole/mole) regardless of
aerosol model
- Corrections in setsox routine for chemistry mechanisms with MAM3
aerosols that include NH3
- Removed kludge in the hetrxtrate routine of mo_usrrxt which was
introduced to keep the bulk aerosol model bit-for-bit
Bugs fixed (include bugzilla ID): Bug 1824
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
D models/atm/cam/src/chemistry/bulk_aero/sulchem.F90
- this code is obsolete and thus was removed
List all subroutines added and what they do:
A models/atm/cam/test/system/config_files/fmz4c5dm
A models/atm/cam/test/system/config_files/fsf1.9c5dh
A models/atm/cam/test/system/config_files/fprgspc4c4dm
A models/atm/cam/test/system/config_files/fsf4c4dm
A models/atm/cam/test/system/config_files/fst4c5dm
A models/atm/cam/test/system/config_files/fsf1.9c4dh
A models/atm/cam/test/system/config_files/f1.9c4carmsuldh
A models/atm/cam/test/system/nl_files/outfrq3s_NEUwetdep
- updates to chemistry regression tests
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/namelist_files/use_cases/1850-PD_cam5.xml
M models/atm/cam/bld/namelist_files/use_cases/1850-2005_cam5.xml
M models/atm/cam/bld/namelist_files/use_cases/2006-2100_cam5_rcp60.xml
M models/atm/cam/bld/namelist_files/use_cases/2006-2100_cam5_rcp26.xml
M models/atm/cam/bld/namelist_files/use_cases/2000_cam5_trop_strat_mam7.xml
M models/atm/cam/bld/namelist_files/use_cases/2006-2100_cam5_rcp45.xml
M models/atm/cam/bld/namelist_files/use_cases/2006-2100_cam5_rcp85.xml
M models/atm/cam/bld/namelist_files/use_cases/1850_cam5.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1850_cam5.xml
M models/atm/cam/bld/namelist_files/use_cases/1850_cam5_super_fast_llnl.xml
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- new emissions files for MAM pom and soag
-- the emissions MAM primary organic matter and SOA are now
multiplied offline by a factor of 1.4
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- removed the obsolete "use_cam_sulfchem" namelist option
M models/atm/cam/bld/namelist_files/use_cases/1850_cam4_trop_bam.xml
- trop_bam now needs spectral irradiance solar data for photolysis
M models/atm/cam/bld/build-namelist
- references to 'use_cam_sulfchem' namelist variable were moved
- bulk aerosol model need default photolysis inputs
M models/atm/cam/bld/perl5lib/Build/ChemNamelist.pm
- removed the kludge that excluded H2SO4 from the gas_wetdep_list namelist variable
M models/atm/cam/bld/perl5lib/Build/ChemPreprocess.pm
- for BAM include the chemistry for the aerosol precursors
M models/atm/cam/src/physics/cam/uwshcu.F90
- removed the special code which set trmin to 1e-5 for number density tracers
-- this is accounted for by setting qmin to 1e-5 when the tracers
are registered to the consistuents module
M models/atm/cam/src/chemistry/pp_trop_bam/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_trop_bam/mo_setrxt.F90
M models/atm/cam/src/chemistry/pp_trop_bam/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_trop_bam/mo_rxt_rates_conv.F90
M models/atm/cam/src/chemistry/pp_trop_bam/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_bam/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_trop_bam/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_bam/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_trop_bam/mo_phtadj.F90
M models/atm/cam/src/chemistry/pp_trop_bam/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_bam/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_trop_bam/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_bam/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_trop_bam/m_spc_id.F90
M models/atm/cam/src/chemistry/pp_trop_bam/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_bam/chem_mech.in
- changes to trop_bam mechanism to include chemistry for aerosol precursors
which is needed to replace the calcuations of the obsolete sulfchem routine
M models/atm/cam/src/chemistry/bulk_aero/progseasalts_intr.F90
- added wet dep diagnostics
M models/atm/cam/src/chemistry/bulk_aero/dust_sediment_mod.F90
- corrections to if block expressions
M models/atm/cam/src/chemistry/bulk_aero/dust_intr.F90
- corrections to dust emissions to use unmodified dust fluxes
from surface models each time step -- in time-steps where fluxes
from surface models were not updated modification of cam_in%cflx
were not representations of dust flux from surface models but the
modifaction of dust fluxes from the previous time-step.
M models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
- add emissions fluxes returned from set_srf_emissions to cflx
- removed references to do_cam_sulfchem
M models/atm/cam/src/chemistry/mozart/mo_chm_diags.F90
- let chemistry output all aerosols in MMR units (kg/kg)
and gas species in VMR units (mole/mole) regardless of
aerosol model
M models/atm/cam/src/chemistry/mozart/mo_srf_emissions.F90
M models/atm/cam/src/chemistry/mozart/mo_extfrc.F90
- removed special code which scaled organic aerosol emissions
M models/atm/cam/src/chemistry/mozart/chemistry.F90
- register number density tracers with qmin set to 1.e-5
- removed use_cam_sulfchem from chemistry namelist
M models/atm/cam/src/chemistry/mozart/mz_aerosols_intr.F90
- removed obsolete sulfchem option
M models/atm/cam/src/chemistry/mozart/mo_usrrxt.F90
- removed a kludge in the hetrxtrate routine which was introduced to
keep the bulk aerosol model bit-for-bit
M models/atm/cam/src/chemistry/mozart/mo_setsox.F90
- correction for chemistry mechanisms with MAM3 aerosols that inlucde NH3
M models/atm/cam/src/control/camsrfexch.F90
M models/atm/cam/src/cpl_mct/atm_comp_mct.F90
M models/atm/cam/src/cpl_esmf/atm_comp_esmf.F90
- extend cam_in_t data structure for dust fluxes from surface models
- dust fluxes from the surface models are simply copied to cam_in%dstflx
without any modifications. This prevents an error which happens
when the model goes from time step 0 to time step 1, which does not
couple with surface models and thus does not restore cflx with dust
fluxes from coupler.
M models/atm/cam/test/system/tests_chem_hybrid
M models/atm/cam/test/system/input_tests_master
M models/atm/cam/test/system/tests_chem_mpi
- updates to chemistry regression tests
M models/atm/cam/src/chemistry/utils/tracer_data.F90
-- Bugzilla – Bug 1824 -- don't use cyc_ndx_beg before it is defined
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Fri Sep 27 21:51:05 MDT 2013
016 bl330 TBL.sh f4c5paqdh aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Fri Sep 27 22:04:01 MDT 2013
024 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...............................FAIL! rc= 7 at Fri Sep 27 22:27:31 MDT 2013
039 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ......................................FAIL! rc= 7 at Fri Sep 27 22:42:35 MDT 2013
043 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ............................FAIL! rc= 7 at Fri Sep 27 23:00:23 MDT 2013
046 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .....................................FAIL! rc= 7 at Fri Sep 27 23:17:06 MDT 2013
051 bl373 TBL.sh fm1.9c4dh outfrq3s 9s ....................................FAIL! rc= 7 at Fri Sep 27 23:47:27 MDT 2013
053 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...........................FAIL! rc= 7 at Fri Sep 27 23:51:43 MDT 2013
055 bl376 TBL.sh fm1.9c4dh+testmech outfrq3s 9s ...........................FAIL! rc= 7 at Fri Sep 27 23:58:49 MDT 2013
060 bl430 TBL.sh fm1.9c5dh outfrq3s_NEUwetdep 9s ..........................FAIL! rc= 7 at Sat Sep 28 00:17:53 MDT 2013
063 bl440 TBL.sh fsoa1.9c4dh outfrq3s+soa_chem_megan_emis 9s ..............FAIL! rc= 7 at Sat Sep 28 00:38:28 MDT 2013
079 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ....................FAIL! rc= 7 at Sat Sep 28 01:37:52 MDT 2013
081 bl740 TBL.sh h30c5h outfrq3s 9s .......................................FAIL! rc= 7 at Sat Sep 28 01:40:56 MDT 2013
090 bl997 TBL_ccsm.sh ne16_g37 FC5 2d .....................................FAIL! rc= 7 at Sat Sep 28 02:22:09 MDT 2013
- expected failures
092 sc001 TSC.sh e64c5bfbiop scm_prep scm64c5bfbiop scm_b4b_o1 7s .........FAIL! rc= 6 at Sat Sep 28 02:32:04 MDT 2013
- this failed in previous tag
frankfurt/nag:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Fri Sep 27 16:46:11 MDT 2013
009 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Fri Sep 27 16:56:14 MDT 2013
013 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Fri Sep 27 17:04:30 MDT 2013
017 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Fri Sep 27 17:15:09 MDT 2013
025 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Fri Sep 27 17:41:33 MDT 2013
032 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 7 at Fri Sep 27 17:54:59 MDT 2013
- expected failures
frankfurt/pgi:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Fri Sep 27 16:46:48 MDT 2013
010 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Fri Sep 27 16:56:32 MDT 2013
012 bl113 TBL.sh e8c5paqdm aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Fri Sep 27 17:02:49 MDT 2013
016 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Fri Sep 27 17:10:19 MDT 2013
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Fri Sep 27 17:27:21 MDT 2013
031 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Fri Sep 27 18:02:22 MDT 2013
041 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 7 at Fri Sep 27 18:27:56 MDT 2013
044 bl317 TBL.sh f10c5cdm outfrq3s 9s .....................................FAIL! rc= 7 at Fri Sep 27 18:39:34 MDT 2013
053 bl321 TBL.sh f10c5cdm atrain 9s .......................................FAIL! rc= 7 at Fri Sep 27 18:55:59 MDT 2013
056 bl420 TBL.sh f7mode4c5dm outfrq3s 9s ..................................FAIL! rc= 7 at Fri Sep 27 19:15:26 MDT 2013
- expected failures
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate): larger than roundoff but same climate
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
https://svn-ccsm-models.cgd.ucar.edu/cam1/branch_tags/aerofixes_cam5_3_12_tags/aerofixes01_cam5_3_12
- platform/compilers:
yellowstone / intel
- configure commandline:
create_newcase -case /glade/p/acd/fvitt/cesm/cases/FC5.cam5312.ne30.test.001 -compset FC5 -res ne30_ne30 -mach yellowstone
- build-namelist command (or complete namelist):
defaults for the FC5 ne30_ne30 configuration
- MSS location of output:
-fvitt/csm/FC5.cam5312.ne30.test.001
MSS location of control simulations used to validate new climate:
/CCSM/csm/f.e13.FC5.ne30_ne30.test2ans.002
URL for AMWG diagnostics output used to validate new climate:
http://webext.cgd.ucar.edu/FCLIMO/FC5.cam5312.ne30.test.001/atm/FC5.cam5312.ne30.test.001-f.e13.FC5.ne30_ne30.test2ans.002/
===============================================================
===============================================================
Tag name: cam5_3_13
Originator(s): cacraig
Date: 9/27/13
One-line Summary: SVN_EXTERNAL_DIRECTORIES updated to cesm1_3_beta03
Purpose of changes:
- Updated to all cesm1_3_beta03 directories
- Updated share to share3_130918 (cesm1_3_alpha04a) to fix the frankfurt single column tests
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M SVN_EXTERNAL_DIRECTORIES
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB except:
004 bl132 TBL.sh e48c4dh ghgrmp+1850_cam4 9s ..............................FAIL! rc= 7 at Mon Sep 23 19:10:56 MDT 2013
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Mon Sep 23 19:14:57 MDT 2013
014 bl322 TBL.sh fn1.9c5dh outfrq3s 9s ....................................FAIL! rc= 7 at Mon Sep 23 19:26:09 MDT 2013
020 bl331 TBL.sh f4c4dh co2rmp+1850_cam4 9s ...............................FAIL! rc= 7 at Mon Sep 23 19:35:35 MDT 2013
024 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...............................FAIL! rc= 7 at Mon Sep 23 19:50:23 MDT 2013
036 bl336 TBL.sh wm4c4h outfrq3s+waccm_1850_cam4 9s .......................FAIL! rc= 7 at Mon Sep 23 19:57:50 MDT 2013
039 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ......................................FAIL! rc= 7 at Mon Sep 23 20:05:37 MDT 2013
043 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ............................FAIL! rc= 7 at Mon Sep 23 20:23:33 MDT 2013
046 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .....................................FAIL! rc= 7 at Mon Sep 23 20:40:25 MDT 2013
051 bl373 TBL.sh fm1.9c4dh outfrq3s 9s ....................................FAIL! rc= 7 at Mon Sep 23 21:10:38 MDT 2013
053 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...........................FAIL! rc= 7 at Mon Sep 23 21:14:51 MDT 2013
055 bl376 TBL.sh fm1.9c4dh+testmech outfrq3s 9s ...........................FAIL! rc= 7 at Mon Sep 23 21:22:02 MDT 2013
058 bl374 TBL.sh fs1.9c4dh outfrq3s+1850-2005_cam4_super_fast_llnl 9s .....FAIL! rc= 7 at Mon Sep 23 21:31:13 MDT 2013
060 bl430 TBL.sh fm1.9c5dh outfrq3s_mozEOOH 9s ............................FAIL! rc= 7 at Mon Sep 23 21:41:21 MDT 2013
063 bl440 TBL.sh fsoa1.9c4dh outfrq3s_mozEOOH+soa_chem_megan_emis 9s ......FAIL! rc= 7 at Mon Sep 23 22:02:19 MDT 2013
067 bl387 TBL.sh f1.9c4dm outfrq3s+1850-2005_cam4 9s ......................FAIL! rc= 7 at Mon Sep 23 22:13:31 MDT 2013
070 bl388 TBL.sh f1.9c4m outfrq24h+1850_cam4 2d ...........................FAIL! rc= 7 at Mon Sep 23 22:16:13 MDT 2013
073 bl379 TBL.sh wm1.9c4h outfrq24h+waccm_1850_cam4 2d ....................FAIL! rc= 7 at Mon Sep 23 22:33:18 MDT 2013
081 bl740 TBL.sh h30c5h outfrq3s 9s .......................................FAIL! rc= 7 at Mon Sep 23 23:04:53 MDT 2013
087 bl992 TBL_ccsm.sh f19_g16 ETEST 2d ....................................FAIL! rc= 7 at Mon Sep 23 23:31:36 MDT 2013
- Changes expected due to Tony's coupler changes
092 sc001 TSC.sh e64c5bfbiop scm_prep scm64c5bfbiop scm_b4b_o1 7s .........FAIL! rc= 6 at Mon Sep 23 23:54:57 MDT 2013
- John Truesdale investigated and has proposed a temporary fix which will most likely be in the next tag. This
test runs to completion but fails in the comparison test with a regular CAM run.
frankfurt/nag:all BFB except:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Mon Sep 23 17:08:08 MDT 2013
009 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Mon Sep 23 17:15:03 MDT 2013
013 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Mon Sep 23 17:21:44 MDT 2013
017 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Mon Sep 23 17:31:52 MDT 2013
025 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Mon Sep 23 17:59:27 MDT 2013
030 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7 at Mon Sep 23 18:09:20 MDT 2013
035 bl318 TBL.sh f10c4cdm sat_hist 9s .....................................FAIL! rc= 7 at Mon Sep 23 18:22:32 MDT 2013
- Changes expected due to Tony's coupler changes
frankfurt/pgi: all BFB except:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Mon Sep 23 17:12:03 MDT 2013
010 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Mon Sep 23 17:21:52 MDT 2013
016 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Mon Sep 23 17:35:36 MDT 2013
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Mon Sep 23 17:55:59 MDT 2013
031 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Mon Sep 23 18:32:05 MDT 2013
036 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7 at Mon Sep 23 18:45:53 MDT 2013
039 bl315 TBL.sh f10c3dm outfrq3s+cam3 9s .................................FAIL! rc= 7 at Mon Sep 23 18:52:03 MDT 2013
044 bl317 TBL.sh f10c5cdm outfrq3s 9s .....................................FAIL! rc= 7 at Mon Sep 23 19:09:22 MDT 2013
047 bl319 TBL.sh fn10c5dm outfrq3s_bam 9s .................................FAIL! rc= 7 at Mon Sep 23 19:16:29 MDT 2013
050 bl320 TBL.sh fn10c5dm rad_diag 9s .....................................FAIL! rc= 7 at Mon Sep 23 19:19:22 MDT 2013
053 bl321 TBL.sh f10c5cdm atrain 9s .......................................FAIL! rc= 7 at Mon Sep 23 19:26:45 MDT 2013
056 bl420 TBL.sh f7mode4c5dm outfrq3s 9s ..................................FAIL! rc= 7 at Mon Sep 23 19:46:08 MDT 2013
- Changes expected due to Tony's coupler changes
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_12
Originator(s): santos
Date: 2013/09/22
One-line Summary: Fix radiation bug, change SE dycore to run with Pa.
Purpose of changes:
- Fix a bug in cloud_rad_props, which was rounding down in a table lookup
rather than interpolating between table values, due to declaring an integer
rather than a real as the interpolation weight.
- Change the dycore to use units of Pa instead of hPa, to remove the need
for a conversion and prevent future bugs.
- Fix minor precision issues, where complex numbers were constructed in
single precision when double was intended.
- Add clarification to error message in modal_aer_opt.
Bugs fixed (include bugzilla ID):
#1794 - Interpolation in cloud_rad_props is less precise than originally
intended.
#1797 - CMPLX has wrong precision in several places
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
- Update HOMME external to use Pa instead of hPa.
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- Clarify that do_circulation_diags must be set to .true. for QBO
forcing to work in WACCM.
M models/atm/cam/src/chemistry/utils/modal_aero_wateruptake.F90
M models/atm/cam/src/physics/cam/phys_prop.F90
- Add explicit kind to calls of the intrinsic "cmplx", to ensure
that it returns a complex number with double-precision
components.
M models/atm/cam/src/control/interpolate_data.F90
- Add parameters corresponding to the extrap_method values, so
that callers can use named constants rather than bare ints.
M models/atm/cam/src/dynamics/se/dp_coupling.F90
M models/atm/cam/src/dynamics/se/dyn_comp.F90
M models/atm/cam/src/dynamics/se/gravity_waves_sources.F90
M models/atm/cam/src/dynamics/se/inidat.F90
M models/atm/cam/src/dynamics/se/stepon.F90
- Remove conversions to hPa (factors of 100._r8).
M models/atm/cam/src/dynamics/se/spmd_dyn.F90
- Remove unnecessary use statement.
M models/atm/cam/src/physics/cam/micro_mg1_0.F90
M models/atm/cam/src/physics/cam/micro_mg1_5.F90
- Very slightly reduce the max value of lambda_c so that it will
always fall within the limits of the table in RRTMG.
M models/atm/cam/src/physics/cam/modal_aer_opt.F90
- Clarify error message for unreasonable optical depth, and abort
when optical depth is negative.
- Make calls to cmplx return double precision values.
M models/atm/cam/src/physics/rrtmg/cloud_rad_props.F90
- Use lininterp to do interpolation of lookup tables, instead of
using (buggy) custom code.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Tue Sep 17 21:30:18 MDT 2013
014 bl322 TBL.sh fn1.9c5dh outfrq3s 9s ....................................FAIL! rc= 7 at Tue Sep 17 21:41:45 MDT 2013
016 bl330 TBL.sh f4c5paqdh aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Tue Sep 17 21:43:22 MDT 2013
024 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...............................FAIL! rc= 7 at Tue Sep 17 22:06:07 MDT 2013
039 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ......................................FAIL! rc= 7 at Tue Sep 17 22:20:49 MDT 2013
043 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ............................FAIL! rc= 7 at Tue Sep 17 22:38:58 MDT 2013
046 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .....................................FAIL! rc= 7 at Tue Sep 17 22:56:00 MDT 2013
053 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...........................FAIL! rc= 7 at Tue Sep 17 23:30:24 MDT 2013
060 bl430 TBL.sh fm1.9c5dh outfrq3s_mozEOOH 9s ............................FAIL! rc= 7 at Tue Sep 17 23:57:06 MDT 2013
076 bl731 TBL.sh h16c3aqdh aqua+aquaplanet_cam3 9s ........................FAIL! rc= 7 at Wed Sep 18 00:56:25 MDT 2013
079 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ....................FAIL! rc= 7 at Wed Sep 18 01:18:00 MDT 2013
081 bl740 TBL.sh h30c5h outfrq3s 9s .......................................FAIL! rc= 7 at Wed Sep 18 01:21:15 MDT 2013
084 bl751 TBL.sh h30c4aqdm outfrq3s+aquaplanet_cam4 9s ....................FAIL! rc= 7 at Wed Sep 18 01:42:23 MDT 2013
090 bl997 TBL_ccsm.sh ne16_g37 FC5 2d .....................................FAIL! rc= 7 at Wed Sep 18 02:01:53 MDT 2013
frankfurt/nag:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Tue Sep 17 15:03:31 MDT 2013
009 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Tue Sep 17 15:10:17 MDT 2013
013 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Tue Sep 17 15:17:04 MDT 2013
017 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Tue Sep 17 15:27:32 MDT 2013
025 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Tue Sep 17 15:52:56 MDT 2013
032 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 7 at Tue Sep 17 16:05:17 MDT 2013
frankfurt/pgi or jaguar/pgi:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Tue Sep 17 15:07:17 MDT 2013
010 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Tue Sep 17 15:16:57 MDT 2013
012 bl113 TBL.sh e8c5paqdm aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Tue Sep 17 15:23:06 MDT 2013
016 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Tue Sep 17 15:30:41 MDT 2013
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Tue Sep 17 15:47:50 MDT 2013
031 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Tue Sep 17 16:22:59 MDT 2013
041 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 7 at Tue Sep 17 16:48:32 MDT 2013
044 bl317 TBL.sh f10c5cdm outfrq3s 9s .....................................FAIL! rc= 7 at Tue Sep 17 16:59:13 MDT 2013
047 bl319 TBL.sh fn10c5dm outfrq3s_bam 9s .................................FAIL! rc= 7 at Tue Sep 17 17:06:27 MDT 2013
050 bl320 TBL.sh fn10c5dm rad_diag 9s .....................................FAIL! rc= 7 at Tue Sep 17 17:09:30 MDT 2013
053 bl321 TBL.sh f10c5cdm atrain 9s .......................................FAIL! rc= 7 at Tue Sep 17 17:16:29 MDT 2013
056 bl420 TBL.sh f7mode4c5dm outfrq3s 9s ..................................FAIL! rc= 7 at Tue Sep 17 17:36:04 MDT 2013
059 bl734 TBL.sh hn16c5aqdm outfrq3s_bam+aquaplanet_cam5 9s ...............FAIL! rc= 7 at Tue Sep 17 18:14:35 MDT 2013
CAM-5 changes answers due to bug fixes, CAM-SE due to units change. This
accounts for all failed baselines.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
All CAM-SE cases, and all CAM5 (specifically due to RRTMG, MG, and MAM
changes).
- what platforms/compilers:
All.
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
The pressure unit change and the cmplx changes are at or below single
precision roundoff. RRTMG is larger than roundoff and climate changing.
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
https://svn-ccsm-models.cgd.ucar.edu/cam1/branch_tags/
santos_answer_changes_tags/santos_answer_changes_n01_cam5_3_08
- platform/compilers:
yellowstone/intel
- configure commandline:
Default FC5 ne30_ne30 run.
- build-namelist command (or complete namelist):
Default FC5 ne30_ne30 run.
- MSS location of output:
/CCSM/csm/f.e13.FC5.ne30_ne30.test2ans.002
MSS location of control simulations used to validate new climate:
/CCSM/csm/f.e13.FC5.ne30_ne30.ctl_ans.001
URL for AMWG diagnostics output used to validate new climate:
Comparison of an earlier version with the RRTMG bug fix to control:
http://webext.cgd.ucar.edu/FCLIMO/f.e13.FC5.ne30_ne30.test2ans.001/atm/
f.e13.FC5.ne30_ne30.test2ans.001-f.e13.FC5.ne30_ne30.ctl_ans.001/
Comparison of this tag to the earlier version mentioned above:
http://webext.cgd.ucar.edu/FCLIMO/f.e13.FC5.ne30_ne30.test2ans.002/atm/
f.e13.FC5.ne30_ne30.test2ans.002-f.e13.FC5.ne30_ne30.test2ans.001/
Comparison of this tag to observations:
http://webext.cgd.ucar.edu/FCLIMO/f.e13.FC5.ne30_ne30.test2ans.002/atm/
f.e13.FC5.ne30_ne30.test2ans.002-obs/
===============================================================
===============================================================
Tag name: cam5_3_11
Originator(s): cacraig
Date: 9/17/13
One-line Summary: Corrected SVN_EXTERNAL_DIRECTORIES to match cam5_3_09 version
Purpose of changes:
- Corrected the SVN_EXTERNAL_DIRECTORIES file which inadvertently was checked in with an older version
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M SVN_EXTERNAL_DIRECTORIES
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB
frankfurt/nag:all BFB
frankfurt/pgi: all BFB
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_10
Originator(s): tcraig, cacraig
Date: 9/06/13
One-line Summary: Tony's coupler changes
Purpose of changes:
- CAM trunk tag for merge of https://svn-ccsm-models.cgd.ucar.edu/cam1/branch_tags/cplupa_tags/cplupa_n01_cam5_3_09
From Tony's notes:
- modify intents on init/run/finalize coupling interfaces
- add ifdef USE_ESMF_METADATA arouind esmf attribute calls
- modify mkatmsrffile.F90 to account fo changes in shr_mct_sMatPInitnc interface
- Removed all mergeinfo files except the root
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton, andrew
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/tools/mkatmsrffile/mkatmsrffile.F90
M models/atm/cam/bld/namelist_files/namelist_definition.xml
M models/atm/cam/doc/ChangeLog
M models/atm/cam/src/utils/cam_dom/cpl_mct/ocn_comp_mct.F90
M models/atm/cam/src/utils/cam_aqua/cpl_mct/ocn_comp_mct.F90
M models/atm/cam/src/cpl_mct/atm_comp_mct.F90
M models/atm/cam/src/cpl_esmf/atm_comp_esmf.F90
M SVN_EXTERNAL_DIRECTORIES
- merge to Tony's cplupa branch
property 'svn:mergeinfo' deleted from 'models/atm/cam'.
property 'svn:mergeinfo' deleted from 'models/atm/cam/test/system/tests_pretag_frankfurt_pgi'.
property 'svn:mergeinfo' deleted from 'models/atm/cam/test/system/TCB_ccsm.sh'.
property 'svn:mergeinfo' deleted from 'models/atm/cam/test/system/tests_posttag_frankfurt'.
property 'svn:mergeinfo' deleted from 'models/atm/cam/bld/configure'.
property 'svn:mergeinfo' deleted from 'models/atm/cam/bld/Makefile.in'.
property 'svn:mergeinfo' deleted from 'models/atm/cam/bld/config_files/definition.xml'.
property 'svn:mergeinfo' deleted from 'models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml'.
property 'svn:mergeinfo' deleted from 'models/atm/cam/bld/namelist_files/namelist_definition.xml'.
property 'svn:mergeinfo' deleted from 'models/atm/cam/bld/namelist_files/use_cases/2000_cam4_trop_strat_soa.xml'.
property 'svn:mergeinfo' deleted from 'models/atm/cam/bld/namelist_files/use_cases/2000_cam5_trop_strat_mam3.xml'.
property 'svn:mergeinfo' deleted from 'models/atm/cam/bld/namelist_files/use_cases/2000_cam5_trop_strat_mam7.xml'.
property 'svn:mergeinfo' deleted from 'models/atm/cam/bld/namelist_files/use_cases/soa_chem_megan_emis.xml'.
property 'svn:mergeinfo' deleted from 'models/atm/cam/bld/namelist_files/use_cases/2000_cam5_trop_moz_mam3.xml'.
property 'svn:mergeinfo' deleted from 'models/atm/cam/bld/build-namelist'.
property 'svn:mergeinfo' deleted from 'models/atm/cam/src/control/camsrfexch.F90'.
property 'svn:mergeinfo' deleted from 'models/atm/cam/src/cpl_share/cam_cpl_indices.F90'.
property 'svn:mergeinfo' deleted from 'models/atm/cam/src/cpl_mct/atm_comp_mct.F90'.
property 'svn:mergeinfo' deleted from 'models/atm/cam/src/physics/cam/comsrf.F90'.
property 'svn:mergeinfo' deleted from 'models/atm/cam/src/physics/cam/micro_mg_cam.F90'.
property 'svn:mergeinfo' deleted from 'models/atm/cam/src/physics/cam/wv_sat_methods.F90'.
property 'svn:mergeinfo' deleted from 'models/atm/cam/src/physics/cam/micro_mg1_0.F90'.
property 'svn:mergeinfo' deleted from 'models/atm/cam/src/physics/cam/microp_driver.F90'.
property 'svn:mergeinfo' deleted from 'models/atm/cam/src/physics/cam/wv_saturation.F90'.
property 'svn:mergeinfo' deleted from 'models/atm/cam/src/physics/cam/uwshcu.F90'.
property 'svn:mergeinfo' deleted from 'models/atm/cam/src/physics/rrtmg/radsw.F90'.
property 'svn:mergeinfo' deleted from 'models/atm/cam/src/chemistry/utils/modal_aero_wateruptake.F90'.
property 'svn:mergeinfo' deleted from 'models/atm/cam/src/chemistry/utils/modal_aero_calcsize.F90'.
property 'svn:mergeinfo' deleted from 'models/atm/cam/src/chemistry/mozart/mz_aerosols_intr.F90'.
property 'svn:mergeinfo' deleted from 'models/atm/cam/src/dynamics/se/dp_coupling.F90'.
property 'svn:mergeinfo' deleted from 'models/atm/cam/src/cpl_esmf/atm_comp_esmf.F90'.
- files which had mergeinfo removed
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB
frankfurt/nag:all BFB
frankfurt/pgi: all BFB except
060 sm991 TSM_ccsm.sh f19_f19 F 2d ........................................FAIL! rc= 6 at Thu Sep 5 18:11:33 MDT 2013
061 eq991 TEQ_ccsm.sh f19_f19 F f1.9c4m fcase 2d ..........................FAIL! rc= 4 at Thu Sep 5 18:11:33 MDT 2013
Jay agreed that it made sense to check in cam5_3_10 with this failure and discover the solution during CESM testing
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_09
Originator(s): cacraig
Date: 8/30/13
One-line Summary: minor microphysics rearrangement and cleanup
Purpose of changes:
- Moved calculation of in-cloud water values for radiation into cloud_diagnostics
- Moved calls for micro_mg_tend back to inline instead of being contained at end of module
- Simplified the calling list for conv_water_4rad
- Moved a handful pbuf_add_field, outfld and addfld calls to modules where data is being set
- Renamed pbuf_times to dyn_time_lvls and itim to itim_old. These values should only be
used for leap-frogging times.
- cmeliq is now a member of pbuf and not passed on calling lists
- Removed effi, effc, effc_fn, effliq, effice and effliq_fn variables as they were exact copies
of rei, rel and rel_fn and were redundant inside micro_mg_cam.F90
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton, andrew
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/src/control/cam_history.F90
- added trim to string for calendar
M models/atm/cam/src/physics/cam/cloud_diagnostics.F90
- Contains the conv_rad_4rad call and associated calculations for radiation using mg physics
M models/atm/cam/src/physics/cam/micro_mg_cam.F90
- Moved in-cloud water calculations which are specific for radiation to cloud_diagnostics
- Moved calls for micro_mg_tend back to inline instead of being contained at end of module
- Moved a number of pbuf_add_field and addfld calls to modules where data is being set
- Removed effi, effc, effc_fn, effliq, effice and effliq_fn variables as they were exact copies
of rei, rel and rel_fn and were redundant
M models/atm/cam/src/physics/cam/clubb_intr.F90
M models/atm/cam/src/physics/cam/macrop_driver.F90
M models/atm/cam/src/physics/cam/microp_driver.F90
- cmeliq is now added to pbuf
M models/atm/cam/src/physics/cam/conv_water.F90
- simplified calling list for conv_water_4rad
- addfld calls moved from micro_mg_cam
M models/atm/cam/src/physics/cam/convect_shallow.F90
M models/atm/cam/src/physics/cam/cloud_fraction.F90
M models/atm/cam/src/physics/cam/convect_deep.F90
- pbuf_add_field, addfld and outfld call(s) moved from micro_mg_cam
M models/atm/cam/src/physics/cam/microp_aero.F90
M models/atm/cam/src/physics/cam/radiation.F90
M models/atm/cam/src/physics/cam/flux_avg.F90
M models/atm/cam/src/physics/cam/check_energy.F90
M models/atm/cam/src/physics/cam/cam_diagnostics.F90
M models/atm/cam/src/physics/cam/stratiform.F90
M models/atm/cam/src/physics/cam/radiation_data.F90
M models/atm/cam/src/physics/cam/physpkg.F90
M models/atm/cam/src/physics/cam/physics_buffer.F90
M models/atm/cam/src/physics/cam/physics_buffer.F90.in
M models/atm/cam/src/physics/cam/cospsimulator_intr.F90
M models/atm/cam/src/physics/cam/zm_conv_intr.F90
M models/atm/cam/src/physics/carma/cam/carma_intr.F90
M models/atm/cam/src/physics/carma/models/cirrus/carma_model_mod.F90
M models/atm/cam/src/physics/rrtmg/ebert_curry.F90
M models/atm/cam/src/physics/rrtmg/oldcloud.F90
M models/atm/cam/src/physics/rrtmg/radiation.F90
M models/atm/cam/src/physics/rrtmg/slingo.F90
M models/atm/cam/src/chemistry/utils/modal_aero_wateruptake.F90
M models/atm/cam/src/chemistry/bulk_aero/aerosol_intr.F90
M models/atm/cam/src/chemistry/mozart/mo_neu_wetdep.F90
- Rename of pbuf_times to dyn_time_lvls and itim to itim_old
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all BFB
frankfurt/nag:all BFB (mc111 failed on first try due to mangled log file due to multiprocessing) Passed on second try
frankfurt/pgi: all BFB
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
Inside micro_mg_cam: calculation of tgliqwp now uses iclwpst instead of iclwp. This changes answers for
the diagnostic variables APRL, PE and PEFRAC. No regression tests show any changes.
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_08
Originator(s): santos
Date: 2013/08/29
One-line Summary: Fix NAG, CESM scripts with Frankfurt/PGI, serial CAM-SE
Purpose of changes:
- Fix the NAG port, since the RRTMG changes were temporarily reverted to
fix a bug on big-endian systems (e.g. Mira).
- Pull in an updated Machines tag so that the CESM scripts can run with
PGI on Frankfurt.
- Have CAM tests use PGI 13.7 on Frankfurt, and change compiler flags, to
match the new Machines tag.
- Fix HOMME when SPMD is not defined, so that it will work with
mpi-serial.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
- The "-Mvect=nosse" option is no longer used for PGI, because SSE
vectorization has been safe for a while now, and because Machines has
made the same change.
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M SVN_EXTERNAL_DIRECTORIES
- New Machines tag with Frankfurt/PGI update.
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
- New COSP and RRTMG tags, compatible with NAG, and with a
big-endian system fix.
M models/atm/cam/bld/Makefile.in
- Remove -Mvect=nosse option from PGI builds.
M models/atm/cam/src/control/runtime_opts.F90
- Fix spelling error in abort message.
M models/atm/cam/src/dynamics/se/dyn_comp.F90
M models/atm/cam/src/dynamics/se/native_mapping.F90
M models/atm/cam/src/dynamics/se/pmgrid.F90
- Fix the SE dycore when SPMD is not defined (mpi-serial).
- Add "implicit none" to pmgrid, since "iam" and "masterproc"
were getting implicit types.
M models/atm/cam/test/system/TCB_ccsm.sh
- Put the CESM build in a directory called "bld", following
convention.
M models/atm/cam/test/system/test_driver.sh
- Change the version of PGI used on Frankfurt to 13.7.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS.
frankfurt/nag: All PASS.
frankfurt/pgi or jaguar/pgi: All PASS.
CAM tag used for the baseline comparison tests if different than previous
tag:
For NAG only, change
Summarize any changes to answers, i.e.,
- what code configurations:
CAM standalone builds only.
- what platforms/compilers:
PGI, only with optimization on.
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
Roundoff.
If bitwise differences were observed, how did you show they were no worse
than roundoff?
The difference is due to a compiler optimization change, which should be
no worse than roundoff.
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_07
Originator(s): fischer, fvitt
Date: Mon Aug 5 15:10:54 MDT 2013
One-line Summary: flbc_list namelist logic update, rrtmg reverted for mira
Purpose of changes:
. SVN external updates
- Reverted rrtmg tag to previous tag due to a big endian bug caused by
a fix to get the NAG compiler to work. This was causing large errors in
the energy balance.
- Updated homme external to fix threading issues on bluegene systems
- Updated component tags to match cesm1_3_beta01
. Update to how flbc_list is set in namelist
. Added new scenario_ghg called CHEM_LBC_FILE to indicate when values are being
set from flbc_file.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: fischer, eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
. Update homme, and reverted rrtmg tag
M SVN_EXTERNAL_DIRECTORIES
. Updated externals to match cesm1_3_beta01
M models/atm/cam/bld/namelist_files/use_cases/2000_cam4_trop_chem.xml
. Removed flbc_list since this is the default being set in build-namelist
M models/atm/cam/bld/namelist_files/namelist_definition.xml
M models/atm/cam/bld/namelist_files/use_cases/2000_cam4_trop_chem.xml
M models/atm/cam/bld/build-namelist
M models/atm/cam/src/physics/cam/chem_surfvals.F90
M models/atm/cam/src/physics/cam/physpkg.F90
M models/atm/cam/src/chemistry/utils/mo_flbc.F90
. Updates to the way flbc_list is being handled
Added new scenario_ghg called CHEM_LBC_FILE
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all passed
frankfurt/nag:
001 sm111 TSM.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 8 at Fri Aug 2 11:56:12 MDT 2013
002 er111 TER.sh e8c5t5mdm ttrac 4+5s .....................................FAIL! rc= 5 at Fri Aug 2 11:56:12 MDT 2013
003 br111 TBR.sh e8c5t5mdm ttrac 6+3s .....................................FAIL! rc= 5 at Fri Aug 2 11:58:17 MDT 2013
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 4 at Fri Aug 2 11:58:17 MDT 2013
005 mc111 TMC.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 4 at Fri Aug 2 11:58:17 MDT 2013
006 dd111 TDD.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 4 at Fri Aug 2 11:58:17 MDT 2013
007 sm112 TSM.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 8 at Fri Aug 2 12:03:02 MDT 2013
008 er112 TER.sh e8c5dm ghgrmp 4+5s .......................................FAIL! rc= 5 at Fri Aug 2 12:03:02 MDT 2013
009 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 4 at Fri Aug 2 12:03:02 MDT 2013
010 sm114 TSM.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 8 at Fri Aug 2 12:05:18 MDT 2013
011 er114 TER.sh e8c5dm co2rmp 4+5s .......................................FAIL! rc= 5 at Fri Aug 2 12:05:18 MDT 2013
012 br114 TBR.sh e8c5dm co2rmp 6+3s .......................................FAIL! rc= 5 at Fri Aug 2 12:07:30 MDT 2013
013 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 4 at Fri Aug 2 12:07:30 MDT 2013
014 sm311 TSM.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 8 at Fri Aug 2 12:12:34 MDT 2013
015 er311 TER.sh f10c5t5mdm ttrac 4+5s ....................................FAIL! rc= 5 at Fri Aug 2 12:12:34 MDT 2013
016 br311 TBR.sh f10c5t5mdm ttrac 6+3s ....................................FAIL! rc= 5 at Fri Aug 2 12:15:18 MDT 2013
017 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 4 at Fri Aug 2 12:15:19 MDT 2013
018 mc311 TMC.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 4 at Fri Aug 2 12:15:19 MDT 2013
019 eq301 TEQ.sh f10c5t5mdm ttrac f10c5t5mdm ttrac_lb1 4s .................FAIL! rc= 4 at Fri Aug 2 12:17:47 MDT 2013
020 eq302 TEQ.sh f10c5t5mdm ttrac f10c5t5mdm ttrac_lb2 4s .................FAIL! rc= 4 at Fri Aug 2 12:17:47 MDT 2013
021 eq303 TEQ.sh f10c5t5mdm ttrac f10c5t5mdm ttrac_lb3 4s .................FAIL! rc= 4 at Fri Aug 2 12:17:48 MDT 2013
022 eq304 TEQ.sh f10c5t5mdm ttrac f10c5dm no_ttrac 4s .....................FAIL! rc= 4 at Fri Aug 2 12:17:48 MDT 2013
023 sm312 TSM.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 8 at Fri Aug 2 12:23:00 MDT 2013
024 er312 TER.sh f2.5c5dm ghgrmp 4+5s .....................................FAIL! rc= 5 at Fri Aug 2 12:23:01 MDT 2013
025 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 4 at Fri Aug 2 12:23:01 MDT 2013
031 sm316 TSM.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 8 at Fri Aug 2 12:35:53 MDT 2013
032 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 4 at Fri Aug 2 12:35:53 MDT 2013
036 sc003 TSM.sh scmc5armiop scmarm 7s ....................................FAIL! rc= 8 at Fri Aug 2 12:50:32 MDT 2013
. cam5 tests are expected to fail because the rrtmg external was reverted to an earlier version that has an overflow
error that NAG catches.
frankfurt/pgi:
060 sm991 TSM_ccsm.sh f19_f19 F 2d ........................................FAIL! rc= 6 at Fri Aug 2 16:32:14 MDT 2013
061 eq991 TEQ_ccsm.sh f19_f19 F f1.9c4m fcase 2d ..........................FAIL! rc= 4 at Fri Aug 2 16:32:14 MDT 2013
. expected, CESM scripts are not working with current setup on frankfurt
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_06
Originator(s): santos, fischer
Date: 2013/07/21
One-line Summary: Gravity wave refactoring, remove WACCM_PHYS CPP macro.
Purpose of changes:
- Remove WACCM_PHYS CPP macro (and the equivalent "do_waccm_phys"
function) in favor of more specific physics options.
- Merge CAM and WACCM gravity wave modules and split the result into
smaller, mostly portable modules.
- Make WACCM modules read their own namelists.
- Move uzm and frontgf/frontga from physics_state to the physics buffer,
since they are not treated the same as other physics_state variables and
introduce unnecessary complication into physics_types.
- Remove Lahey support, since new code in this tag uses Fortran 2003
features.
Bugs fixed (include bugzilla ID):
- Due to a unit mismatch between CAM and HOMME, WACCM-SE was using a
frontogenesis function that was ~14x too large.
- Fix some namelist values in 1850_cam5_pm.xml that had yellowstone paths
hard-coded (h/t Chris Fischer).
Describe any changes made to build system:
Describe any changes made to the namelist:
- New history_waccm flag, for turning on common outputs for WACCM runs.
Implementation of this flag is incomplete and should be considered a
work in progress.
- Three new namelist variables, use_gw_oro, use_gw_front, and
use_gw_convect, are provided to turn on the three types of gravity wave
sources currently used in CAM and WACCM.
- Several previously hard-coded parameters were added to gw_drag_nl.
- New namelist variables "do_molec_press" and "molec_diff_bot_press"
determine whether molecular diffusion is on, and the bottom level where
it will take place, respectively.
- Existing namelist variables have been moved from cam_inparm to qbo_nl,
radheat_nl, and iondrag_nl.
- New namelist settings for WACCM-SE with vertically Lagrangian advection.
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
D models/atm/cam/test/system/tests_pretag_frankfurt_lahey
- Lahey support dropped. (Current NAG test list is identical.)
List all subroutines added and what they do:
A models/atm/cam/src/physics/cam/qbo.F90
- Stub version of QBO module.
A models/atm/cam/src/physics/cam/vdiff_lu_solver.F90
- Contains vd_lu_decomp and vd_lu_solve, so this code can be
shared rather than duplicated between diffusion_solver and
gw_drag. These are also modified so that the output of
vd_lu_decomp is contained in an object of type "lu_decomp",
which can then be fed back to vd_lu_solver directly.
A models/atm/cam/test/system/config_files/h30c4wgdm
A models/atm/cam/test/system/config_files/wg4c4dm
A models/atm/cam/test/system/nl_files/outfrq3s_wse
- Files to test SC-WACCM-SE, and SC-WACCM's use_case.
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/build-namelist
M models/atm/cam/bld/namelist_files/namelist_definition.xml
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- Add history_waccm namelist option.
- Add new molecular diffusion pressures to namelist.
- Add new gw_drag namelist variables.
- Move namelist variables out of cam_inparm.
- Modify SC-WACCM-SE settings (experimental settings for
vertically Lagrangian advection).
M models/atm/cam/bld/namelist_files/use_cases/1850_cam5_pm.xml
- Remove hard-coded glade paths.
M models/atm/cam/doc/ChangeLog_template
M models/atm/cam/test/system/archive_baseline.sh
M models/atm/cam/test/system/gen_test_table.sh
M models/atm/cam/bld/run-pc.csh
M models/atm/cam/bld/configure
M models/atm/cam/bld/Makefile.in
M models/atm/cam/bld/config_files/definition.xml
M models/atm/cam/bld/perl5lib/Build/ChemPreprocess.pm
M models/atm/cam/src/physics/cam/phys_grid.F90
- Remove Lahey support.
- Remove some workarounds for and mentions of Lahey.
M models/atm/cam/src/control/runtime_opts.F90
- Have qbo, radheat, and iondrag read their own namelists.
- Call waccm_forcing_readnl from within radheat_readnl.
M models/atm/cam/src/control/sat_hist.F90
- Use global save statement instead of saving an individual
variable.
M models/atm/cam/src/dynamics/fv/dp_coupling.F90
M models/atm/cam/src/dynamics/fv/dyn_comp.F90
M models/atm/cam/src/dynamics/se/gravity_waves_sources.F90
M models/atm/cam/src/dynamics/se/dp_coupling.F90
M models/atm/cam/src/dynamics/se/dyn_comp.F90
M models/atm/cam/src/dynamics/se/stepon.F90
- Set WACCM variables in the physics buffer rather than
physics_state.
- Replace "do_waccm_phys" with flags specific to the qbo and
gw_drag modules.
- Sometimes frontgf was handled before frontga, and sometimes vice
versa. This made it too easy to mix them up, so move frontgf
before frontga in all these cases.
- Fix bug where WACCM-SE was using a too-large frontogenesis
function.
M models/atm/cam/src/physics/cam/convect_deep.F90
- Remove do_waccm_phys in favor of querying for presence of pbuf
field.
- Change name of "ZMDT" to "TTEND_DP" in the physics buffer.
M models/atm/cam/src/physics/cam/diffusion_solver.F90
- Pull LU solver routines out into the portable module named
vdiff_lu_solver. This is the only dependency of
diffusion_solver, so it is effectively portable in that sense.
Portability is still limited by the fact that it must be passed
routines with very specific interfaces for molecular diffusion.
- Update interfaces for compute_molec_diff and the vd_lu_*
routines.
M models/atm/cam/src/physics/cam/iondrag.F90
- Harmonize stub module with WACCM version and make calls to its
routines non-fatal (so we don't need WACCM_PHYS to determine
whether or not to call it).
M models/atm/cam/src/physics/cam/molec_diff.F90
M models/atm/cam/src/physics/cam/ref_pres.F90
- Have ref_pres find molecular diffusion levels, rather than
molec_diff.
- Change the way ref_pres calculates levels, to guarantee that a
"top" level is not too high and a "bottom" level is not too low.
- Return information about the interpretation of kvt from
compute_molec_diff, so that diffusion_solver no longer has to
use "waccmx_is".
- Use the new lu_decomp type for compatibility with
vdiff_lu_solver.
M models/atm/cam/src/physics/cam/phys_control.F90
- Add WACCM history flag.
- Add use_gw_* flags to specify which gravity wave sources are on.
M models/atm/cam/src/physics/cam/physics_types.F90
- Remove uzm, frontgf, and frontga from physics_state objects.
- Allow ZM to use a different name from convect_deep.
M models/atm/cam/src/physics/cam/physpkg.F90
- Since exbdrift is only used directly by iondrag, move the call
to exbdrift_register into iondrag as well.
- Change gw_drag interfaces.
- Remove "#ifdef WACCM_PHYS" statements.
- Pass pbuf to qbo_relax, since UZM is now in pbuf.
- Call WACCM physics routines in the same order throughout.
M models/atm/cam/src/physics/cam/radheat.F90
- Add stub namelist read to replace defaultopts and setopts.
M models/atm/cam/src/physics/cam/vertical_diffusion.F90
- Don't advertise gravity wave drag as "coming soon" in the
comments.
- Use a utility function in ref_pres to set ntop_eddy.
- Use history_waccm to switch on add_default calls.
- Account for changes in molec_diff and diffusion_solver
interfaces.
M models/atm/cam/src/physics/cam/zm_conv_intr.F90
- Change ptend%name from "convect_deep" to "zm_conv_tend".
M models/atm/cam/src/physics/waccm/iondrag.F90
- Have iondrag handle its own namelist.
- Move in register logic from physpkg.
M models/atm/cam/src/physics/waccm/qbo.F90
- Have qbo handle its own namelist.
- Get uzm from physics buffer.
M models/atm/cam/src/physics/waccm/radheat.F90
- Have radheat handle its own namelist.
- radheat calls waccm_forcing_readnl.
M models/atm/cam/src/physics/waccm/waccm_forcing.F90
- Change null filename (no effect, done just to be consistent with
QBO).
M models/atm/cam/src/physics/waccmx/majorsp_diffusion.F90
- Get molecular diffusion levels from ref_pres, not molec_diff.
M models/atm/cam/src/physics/cam/physics_buffer.F90
M models/atm/cam/src/physics/cam/physics_buffer.F90.in
- Fix spelling error.
M models/atm/cam/test/system/input_tests_master
- Add tests for WACCM-SE and tests that exercise the
waccm_sc_2000_cam4 use_case.
M models/atm/cam/test/system/CAM_compare.sh
- Fix bug where fields with small differences would not be printed
to the log after a comparison failure.
M models/atm/cam/test/system/TCB_ccsm.sh
- Allow compiler names to be specified as part of machine names
(e.g. "frankfurt_nag").
- Turn repetitive blocks into loops.
M models/atm/cam/test/system/input_tests_master
- Add SC-WACCM tests, including with the SE dycore.
M models/atm/cam/test/system/test_driver.sh
- Change walltime limit on Frankfurt so that setting
CAM_BATCHQ="short" will work.
- Remove Lahey section.
Changes to the gravity waves:
M models/atm/cam/src/physics/cam/gw_drag.F90
D models/atm/cam/src/physics/waccm/gw_drag.F90
A models/atm/cam/src/physics/cam/gw_common.F90
A models/atm/cam/src/physics/cam/gw_convect.F90
A models/atm/cam/src/physics/cam/gw_diffusion.F90
A models/atm/cam/src/physics/cam/gw_front.F90
A models/atm/cam/src/physics/cam/gw_oro.F90
A models/atm/cam/src/physics/cam/gw_utils.F90
Highlights of the refactoring:
- Merge WACCM and CAM modules.
- Use history_waccm to switch on add_default calls.
- Eliminate dead code, including an old spectral wave source from CAM's
version of gw_drag and a buggy implementation of low level stress
divergence.
- Set previously hard-coded parameters via namelist.
- Change names of gw_inti and gw_intr to gw_init and gw_tend.
- Move the gravity wave source routines into separate modules
(gw_convect, gw_front, and gw_oro), which are turned on or off via
the phys_control namelist.
- Homogenize and simplify code for the three gravity wave sources,
to simplify the process of adding new sources in the future.
- Separate addfld/outfld calls from other code, and move gravity
wave drag profile routines into a portable layer (gw_common).
- Separate diffusion routines into a separate module from
gw_common, called gw_diffusion, using the new vdiff_lu_solver
module to share code with diffusion_solver.
- Create a module for common vector operations (gw_utils).
- Fix UTGWORO, VTGWORO, and TTGWORO to more accurately reflect the
wind and temperature tendencies from orographic gravity waves.
- Use cpairv rather than cpair (for consistency in WACCM-X).
- Reorder loops to access memory in order and contiguously, and
only calculate the LU decomposition once for diffusion due to
gravity wave breaking, as diffusion_solver already does. These
two optimizations reduce the total cost of spectral gravity
waves by >40% (Yellowstone/ifort). Most of this gain is due to
loop reordering, but it's unclear how much of the benefit was
due to vectorization vs. more efficient cache use. In loops where the
level bounds varied by column for the Beres scheme, this was handled
mostly by masking assignments (where statements).
- Change most arrays from size pcols to size ncol. This cleans up
the code slightly because array operations no longer need an
explicit bound (e.g. "src_level(:ncol)"), but there was no
significant effect on computation time.
- Purely orographic waves have taken a performance hit (roughly double
the cost), but still only represent about 0.1% of the total cost of
CAM5. This cost seems to be primarily the result of changes to
gw_drag_prof, which needs more per-column level indices for the Beres
scheme.
- Miscellaneous cleanup, e.g. limiting lines to <=80 characters, and
re-ordering arguments in interfaces.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
036 bl336 TBL.sh wm4c4h outfrq3s+waccm_1850_cam4 9s .......................FAIL! rc= 7 at Sun Jul 21 14:24:03 MDT 2013
073 bl379 TBL.sh wm1.9c4h outfrq24h+waccm_1850_cam4 2d ....................FAIL! rc= 7 at Sun Jul 21 16:35:21 MDT 2013
Changes to diagnostic output only, as UTGWORO and VTGWORO have been fixed
to match the actual wind tendencies from orographic gravity waves.
frankfurt/nag:
030 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7 at Sun Jul 21 12:55:40 MDT 2013
frankfurt/pgi or jaguar/pgi:
036 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7 at Sun Jul 21 13:44:29 MDT 2013
060 sm991 TSM_ccsm.sh f19_f19 F 2d ........................................FAIL! rc= 6 at Sun Jul 21 15:34:49 MDT 2013
061 eq991 TEQ_ccsm.sh f19_f19 F f1.9c4m fcase 2d ..........................FAIL! rc= 4 at Sun Jul 21 15:34:49 MDT 2013
The bl314 tests failed because the old files used for WACCM at 10x15 have a
different vertical grid, and so some pressure limiters are now rounded to
a different level due to changes in ref_pres. This has no impact on current
WACCM cases.
The CESM tests on Frankfurt failed because of machine issues that have
temporarily broken the CESM port.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: WACCM
- what platforms/compilers: All
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
- For WACCM-FV, larger than roundoff changes for diagnostic output only
(UTGWORO and VTGWORO). The climate and all other diagnostics should be
bit-for-bit.
- For WACCM-SE, a climate-changing bug fix for the frontogenesis
function.
- Some old WACCM files, such as the default for 10x15 resolution, use
different pressure levels than more current files. The tropospheric
cloud physics top level was changed for these cases due to the change
in ref_pres. This showed up in SC-WACCM testing but does not affect
current, supported WACCM cases, as demonstrated by a test of SC-WACCM
on FV at a resolution of 4x5 degrees.
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_05
Originator(s): fischer, cacraig, santos
Date: Tue Jul 2nd, 2013
One-line Summary: COSP build fix for CESM, CLUBB threading fix,
namelist updates
Purpose of changes:
. The COSP build was broken when using cesm scripts with the latest
cesm tags. This is now fixed.
. CLUBB was giving different answers when the number of threads are
changed. This is mostly fixed, except for when clubb_history is true.
When clubb_history is true and the number of threads is > 1, then
build-namelist exits with an error message.
. Namelist updates for flbc_file and bndtvs.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
. configure was refactored so cesm scripts build COSP in a separate library
the same way cam stand alone does. cosp_libdir is a new option for configure,
this is the directory the COSP library will be built.
Describe any changes made to the namelist:
. Updates for bndtvs and flbc_file
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: fischer
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/nl_files/outfrq3s_clubb
. Removed clubb_history from clubb test.
M models/atm/cam/test/system/input_tests_master
. Turned on threading for clubb test.
M models/atm/cam/bld/cam.buildnml.csh
. Pass cosp_libdir when configure is invoked.
M models/atm/cam/bld/configure
. Check for CLUBB threading was removed.
. cosp_libdir is a new variable that can be passed into configure
. Refactored so cesm scripts build COSP in a separate library the
same way cam stand alone does.
M models/atm/cam/bld/Makefile.in
. Add -liomp5 for clubb threading.
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. Updated fpftcon and bndtvs files to match cesm.
M models/atm/cam/bld/namelist_files/namelist_definition.xml
. Fix documentation for COSP.
M models/atm/cam/bld/namelist_files/use_cases/1850-2005_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/1850-2005_cam5.xml
. Updated bndtvs files
M models/atm/cam/bld/namelist_files/use_cases/2000_cam5_trop_strat_mam3.xml
M models/atm/cam/bld/namelist_files/use_cases/2000_cam4_trop_chem.xml
M models/atm/cam/bld/namelist_files/use_cases/2000_cam5_trop_strat_mam7.xml
M models/atm/cam/bld/namelist_files/use_cases/cam4_trop_strat_chem.xml
M models/atm/cam/bld/namelist_files/use_cases/cam5_trop_strat_chem.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1955-2005_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_cam4_nuclear_winter.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2000_cam5.xml
M models/atm/cam/bld/namelist_files/use_cases/cam4_chem_radpsv_geos5.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp45.xml
. Updated flbc_file (from santos)
M models/atm/cam/bld/build-namelist
. Add an exit if threading and clubb_history is true. This combination isn't thread
safe.
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
. Update clubb external for threading fix
M models/atm/cam/src/physics/cam/clubb_intr.F90
. CLUBB threading and cmeliq fix.
M models/atm/cam/src/physics/cam/physpkg.F90
. Add cmeliq to call the clubb_tend_cam.
M models/atm/cam/src/physics/cam/micro_mg1_0.F90
. Fixes for CLUBB threading.
M SVN_EXTERNAL_DIRECTORIES
. Updated machines tag to fix cosp cesm build.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
043 bl366 TBL.sh f1.9c5clbdh outfrq3s_clubb 9s ............................FAIL! rc= 7 at Tue Jul 2 14:07:28 MDT 2013
. New test
053 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...........................FAIL! rc= 7 at Tue Jul 2 14:07:33 MDT 2013
067 bl387 TBL.sh f1.9c4dm outfrq3s+1850-2005_cam4 9s ......................FAIL! rc= 7 at Tue Jul 2 14:07:38 MDT 2013
. New bndtvs files change answers
frankfurt/lf95: all passed
frankfurt/nag: all passed
frankfurt/pgi: all passed
===============================================================
===============================================================
Tag name: cam5_3_04
Originator(s): santos, cbardeen, jedwards, pel, goldy
Date: 2013/06/11
One-line Summary: Answer-changing fixes and cleanup, NAG port fixes
Purpose of changes:
Fix a bug that caused cases with interactive chemistry involving the CLY
and BRY families to crash within a few years.
Fix a race condition when using WACCM-SE with threading.
Collect answer-changing code cleanup in one tag:
- Remove references to "nonintrinsic" functions, which were
provided only to preserve answers during the NAG port. Unit
tests have verified that nonintrinsic and intrinsic versions
of these functions match to within roundoff error in most cases
(the exceptions involve large arguments to the gamma function,
which are not needed by CAM, or gfortran, which has a less
accurate library).
- Remove a switch in pkg_cldoptics that caused a pointless
difference between WACCM and CAM cases (the do_waccm_phys switch
and the WACCM_PHYS macro will likely be removed entirely soon).
- Change hard-coded physical constants in chemistry to use
physconst (directly, or through mo_constants).
- Remove dead code and change r4 to r8 in modal_aero_coag.
Fix regressions with the NAG port and get it working out of the box.
More bug fixes from various sources:
- Fix broken Darwin build (from Chuck Bardeen).
- Fix multi-instance CAM-SE (from Jim Edwards).
- Fix for topo_tool (from Peter Lauritzen).
- Correct physics_buffer abort messages (from Steve Goldhaber).
Update externals to make the above possible:
- csm_share: Fix for shr_spfn_gamma on Intel.
- CLM: NAG port tag.
- MCT: Fixes for NAG port, Darwin build, and Intel bounds-checking.
These fixes have not been put into a released version of MCT
yet, so for now CAM is using an MCT branch as its external.
Some externals updated to try to avoid CLM compatibility issues:
- RTM
- scripts
- Machines
Bugs fixed (include bugzilla ID):
- 1729: WACCM (and possibly strat chem) crashes within a few years
- 1744: WACCM-SE sometimes fails due to race condition with threading
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton, most chemistry changes reviewed by fvitt
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M SVN_EXTERNAL_DIRECTORIES
- Update csm_share, CLM, and MCT.
M models/atm/cam/bld/Makefile.in
- Fix the COSP build on Darwin.
- Remove some OpenMP/NAG flags, as OpenMP is not ready yet.
M models/atm/cam/bld/configure
- Fix the setting of rpath on Darwin; Darwin's ld will accept
"-rpath /foo" but not "-rpath=foo".
M models/atm/cam/bld/build-namelist
- Fix setting of srf_emis_cycle_yr for super_fast_llnl_mam3,
waccm_mozart_mam3, and trop_mam7.
M models/atm/cam/src/chemistry/modal_aero/modal_aero_coag.F90
- Remove dead code.
- Change r4 declarations to r8.
- Get physical constants from physconst.
M models/atm/cam/src/chemistry/modal_aero/modal_aero_gasaerexch.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_newnuc.F90
M models/atm/cam/src/chemistry/mozart/mo_airglow.F90
M models/atm/cam/src/chemistry/mozart/mo_aurora.F90
M models/atm/cam/src/chemistry/mozart/mo_sethet.F90
M models/atm/cam/src/chemistry/mozart/mo_setinv.F90
M models/atm/cam/src/chemistry/mozart/mo_setsoa.F90
M models/atm/cam/src/chemistry/mozart/mo_usrrxt.F90
M models/atm/cam/src/chemistry/mozart/mz_aerosols_intr.F90
- Get physical constants from from physconst and mo_constants.
(E.g. R_gas, Avogadro's number, Boltzmann's constant...)
- Remove definitions of constants that are never used.
- Fix line over 132 characters in mz_aerosols_intr.F90.
M models/atm/cam/src/chemistry/modal_aero/modal_aero_rename.F90
M models/atm/cam/src/physics/cam/micro_mg1_0.F90
M models/atm/cam/src/physics/cam/micro_mg1_5.F90
M models/atm/cam/src/physics/cam/microp_aero.F90
M models/atm/cam/src/physics/cam/ndrop.F90
M models/atm/cam/src/physics/cam/ndrop_bam.F90
M models/atm/cam/src/physics/cam/uwshcu.F90
M models/atm/cam/src/physics/carma/models/dust/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/cirrus/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/sea_salt/carma_model_mod.F90
- Replace "nonintrinsic" versions of special functions with
the regular versions (which allow shr_spfn_mod to decide if a
Fortran 2008 intrinsic is available).
M models/atm/cam/src/chemistry/utils/mo_constants.F90
- Add more constants with different units needed by the chemistry.
- Explicitly call out cgs units by naming them with a "_cgs"
suffix.
M models/atm/cam/src/control/cam_history_support.F90
- Resolve ambiguity between check_hist_coord_all and
check_hist_coord_char by making the "vlen" argument mandatory
for check_hist_coord_all.
M models/atm/cam/src/control/physconst.F90
- Add molecular weight for NH4.
M models/atm/cam/doc/ChangeLog_template
- Add NAG to compilers used to run tests.
M models/atm/cam/src/dynamics/se/dyn_comp.F90
- Fix multi-instance CAM-SE by only calling write_grid_mapping
from one index.
M models/atm/cam/src/dynamics/se/gravity_waves_sources.F90
- Fix race condition that caused a crash on Intel with DEBUG
on.
M models/atm/cam/src/physics/cam/physics_buffer.F90
M models/atm/cam/src/physics/cam/physics_buffer.F90.in
- Correct endrun messages that mentioned pbuf_get_field but were
actually in pbuf_set_field.
M models/atm/cam/src/physics/cam/physpkg.F90
- Fix bug where clybry_fam_adj would crash when given negative
constituent concentrations, by calling qneg3 beforehand.
M models/atm/cam/src/physics/cam/pkg_cldoptics.F90
- Remove unnecessary truncation of a table when do_waccm_phys()
is .false.
M models/atm/cam/test/system/TCB_ccsm.sh
M models/atm/cam/test/system/test_driver.sh
- Use a CCSM_MPILIB as well as CCSM_MACH. This is a necessary step
for NAG to run CESM tests on frankfurt right now, since only
mpich is built there right now, but the CESM scripts choose
openmpi as the default MPI library.
- Define CCSM_MACH separately for all compilers on frankfurt
except Lahey.
- Correct the spelling of "edison" in edison's CCSM_MACH.
M models/atm/cam/test/system/archive_baseline.sh
- Mention NAG in the help message.
M models/atm/cam/tools/topo_tool/cube_to_target/cube_to_target.F90
M models/atm/cam/tools/topo_tool/cube_to_target/remap.F90
- Several fixes for topo_tool from Peter Lauritzen.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
004 bl132 TBL.sh e48c4dh ghgrmp+1850_cam4 9s ..............................FAIL! rc= 7 at Mon Jun 10 17:51:45 MDT 2013
006 bl133 TBL.sh e48c4paqdm aqpgro+aquaplanet_cam4 3s .....................FAIL! rc= 7 at Mon Jun 10 17:52:35 MDT 2013
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Mon Jun 10 17:55:01 MDT 2013
014 bl322 TBL.sh fn1.9c5dh outfrq3s 9s ....................................FAIL! rc= 7 at Mon Jun 10 18:03:19 MDT 2013
016 bl330 TBL.sh f4c5paqdh aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Mon Jun 10 18:04:42 MDT 2013
020 bl331 TBL.sh f4c4dh co2rmp+1850_cam4 9s ...............................FAIL! rc= 7 at Mon Jun 10 18:09:12 MDT 2013
024 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...............................FAIL! rc= 7 at Mon Jun 10 18:18:41 MDT 2013
026 bl333 TBL.sh f4c4paqdh aqpgro+aquaplanet_cam4 3s ......................FAIL! rc= 7 at Mon Jun 10 18:19:12 MDT 2013
036 bl336 TBL.sh wm4c4h outfrq3s+waccm_1850_cam4 9s .......................FAIL! rc= 7 at Mon Jun 10 18:25:18 MDT 2013
039 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ......................................FAIL! rc= 7 at Mon Jun 10 18:31:04 MDT 2013
043 bl366 TBL.sh f1.9c5clbdm outfrq3s_clubb 9s ............................FAIL! rc= 7 at Mon Jun 10 18:46:32 MDT 2013
046 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .....................................FAIL! rc= 7 at Mon Jun 10 18:59:17 MDT 2013
051 bl373 TBL.sh fm1.9c4dh outfrq3s 9s ....................................FAIL! rc= 7 at Mon Jun 10 19:24:26 MDT 2013
053 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...........................FAIL! rc= 7 at Mon Jun 10 19:28:29 MDT 2013
055 bl376 TBL.sh fm1.9c4dh+testmech outfrq3s 9s ...........................FAIL! rc= 7 at Mon Jun 10 19:35:24 MDT 2013
058 bl374 TBL.sh fs1.9c4dh outfrq3s+1850-2005_cam4_super_fast_llnl 9s .....FAIL! rc= 7 at Mon Jun 10 19:42:59 MDT 2013
060 bl430 TBL.sh fm1.9c5dh outfrq3s_mozEOOH 9s ............................FAIL! rc= 7 at Mon Jun 10 19:52:41 MDT 2013
063 bl440 TBL.sh fsoa1.9c4dh outfrq3s_mozEOOH+soa_chem_megan_emis 9s ......FAIL! rc= 7 at Mon Jun 10 20:11:34 MDT 2013
073 bl379 TBL.sh wm1.9c4h outfrq24h+waccm_1850_cam4 2d ....................FAIL! rc= 7 at Mon Jun 10 20:40:06 MDT 2013
079 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ....................FAIL! rc= 7 at Mon Jun 10 21:06:26 MDT 2013
081 bl740 TBL.sh h30c5h outfrq3s 9s .......................................FAIL! rc= 7 at Mon Jun 10 21:09:21 MDT 2013
090 bl997 TBL_ccsm.sh ne16_g37 FC5 2d .....................................FAIL! rc= 7 at Mon Jun 10 21:47:38 MDT 2013
Expected failures due to various answer changes mentioned below.
The following tests unexpectedly passed:
067 bl387 TBL.sh f1.9c4dm outfrq3s+1850-2005_cam4 9s ......................PASS at Mon Jun 10 20:20:04 MDT 2013
070 bl388 TBL.sh f1.9c4m outfrq24h+1850_cam4 2d ...........................PASS at Mon Jun 10 20:22:31 MDT 2013
076 bl731 TBL.sh h16c3aqdh aqua+aquaplanet_cam3 9s ........................PASS at Mon Jun 10 20:46:26 MDT 2013
084 bl751 TBL.sh h30c4aqdm outfrq3s+aquaplanet_cam4 9s ....................PASS at Mon Jun 10 21:30:11 MDT 2013
087 bl992 TBL_ccsm.sh f19_g16 ETEST 2d ....................................PASS at Mon Jun 10 21:35:01 MDT 2013
These were expected to have answer changes due to the change in
pkg_cldoptics.F90. However, this change would only affect CAM3 and CAM4
non-WACCM cases, and only if temperatures fell below 180K. The most likely
explanation for the fact that these baseline tests passed, is that
temperatures never fell below 180K during the test.
frankfurt/lf95:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Mon Jun 10 16:37:50 MDT 2013
009 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Mon Jun 10 17:17:27 MDT 2013
013 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Mon Jun 10 17:28:38 MDT 2013
017 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Mon Jun 10 18:17:02 MDT 2013
025 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Mon Jun 10 21:10:10 MDT 2013
027 bl313 TBL.sh f10c3aqdm outfrq3s+aquaplanet_cam3 3s ....................FAIL! rc= 7 at Mon Jun 10 21:27:46 MDT 2013
032 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 7 at Mon Jun 10 22:42:03 MDT 2013
035 bl318 TBL.sh f10c4cdm sat_hist 9s .....................................FAIL! rc= 7 at Mon Jun 10 23:15:21 MDT 2013
Expected failures due to various answer changes mentioned below.
frankfurt/nag:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................SKIPPED* at Mon Jun 10 16:02:35 MDT 2013
009 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................SKIPPED* at Mon Jun 10 16:10:17 MDT 2013
013 bl114 TBL.sh e8c5dm co2rmp 9s .........................................SKIPPED* at Mon Jun 10 16:18:33 MDT 2013
017 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................SKIPPED* at Mon Jun 10 16:30:08 MDT 2013
025 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................SKIPPED* at Mon Jun 10 16:57:04 MDT 2013
027 bl313 TBL.sh f10c3aqdm outfrq3s+aquaplanet_cam3 3s ....................SKIPPED* at Mon Jun 10 16:59:36 MDT 2013
030 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................SKIPPED* at Mon Jun 10 17:08:04 MDT 2013
032 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................SKIPPED* at Mon Jun 10 17:10:54 MDT 2013
035 bl318 TBL.sh f10c4cdm sat_hist 9s .....................................SKIPPED* at Mon Jun 10 17:24:24 MDT 2013
Baseline tests skipped because NAG is new (and baselines would be expected
to fail anyway).
frankfurt/pgi or jaguar/pgi:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Mon Jun 10 16:26:49 MDT 2013
010 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Mon Jun 10 16:36:55 MDT 2013
012 bl113 TBL.sh e8c5paqdm aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Mon Jun 10 16:43:39 MDT 2013
016 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Mon Jun 10 16:51:32 MDT 2013
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Mon Jun 10 17:09:33 MDT 2013
031 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Mon Jun 10 17:45:32 MDT 2013
033 bl313 TBL.sh f10c3aqdm outfrq3s+aquaplanet_cam3 3s ....................FAIL! rc= 7 at Mon Jun 10 17:50:21 MDT 2013
039 bl315 TBL.sh f10c3dm outfrq3s+cam3 9s .................................FAIL! rc= 7 at Mon Jun 10 18:05:47 MDT 2013
041 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 7 at Mon Jun 10 18:12:40 MDT 2013
044 bl317 TBL.sh f10c5cdm outfrq3s 9s .....................................FAIL! rc= 7 at Mon Jun 10 18:24:04 MDT 2013
047 bl319 TBL.sh fn10c5dm outfrq3s_bam 9s .................................FAIL! rc= 7 at Mon Jun 10 18:31:46 MDT 2013
050 bl320 TBL.sh fn10c5dm rad_diag 9s .....................................FAIL! rc= 7 at Mon Jun 10 18:34:44 MDT 2013
053 bl321 TBL.sh f10c5cdm atrain 9s .......................................FAIL! rc= 7 at Mon Jun 10 18:42:30 MDT 2013
056 bl420 TBL.sh f7mode4c5dm outfrq3s 9s ..................................FAIL! rc= 7 at Mon Jun 10 19:02:11 MDT 2013
059 bl734 TBL.sh hn16c5aqdm outfrq3s_bam+aquaplanet_cam5 9s ...............FAIL! rc= 7 at Mon Jun 10 19:43:20 MDT 2013
Expected failures due to various answer changes mentioned below.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
All cases with CAM3/4 physics that are not WACCM cases, all cases with
CAM5 physics, and all cases with active chemistry. The CARMA cirrus,
dust, and sea_salt models are also affected.
Therefore, the only bit-for-bit cases will be those without typical CAM
physics (e.g. using ideal physics) and WACCM4 specified chemistry cases.
- what platforms/compilers:
All.
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
Larger than roundoff but same climate, except that changes to use
statements for shr_spfn_mod may be roundoff-level for most compilers.
If bitwise differences were observed, how did you show they were no worse
than roundoff?
For shr_spfn_mod changes, this was demonstrated during previous unit
tests of csm_share code.
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_03
Originator(s): Craig, Goldhaber
Date:
One-line Summary: Implementation of pbuf and history support for sub-columns
Purpose of changes:
- Modifications to pbuf to support sub-columns
- Modifications to history to support sub-columns
- Bug fix to cosp history
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
M models/atm/cam/src/control/cam_history_support.F90
. register_history_mdim: Removed and replaced with add_hist_coord interface
M models/atm/cam/src/physics/cam/cosp_share.F90
. Removed file (merged into cospsimulator_intr.F90 as separate file no
longer needed for cam_history.F90 use).
List all subroutines added and what they do:
M models/atm/cam/src/control/cam_history.F90
. get_field_properties: Retrieve properties from CAM history field
This function needed to provide information to non-history code
needing to manipulate history data (e.g., subcol_outfld).
M models/atm/cam/src/control/cam_history_support.F90
. check_hist_coord interface consisting of routines to verify
consistency between data passed in an add_hist_coord or
add_vert_coord call and any data already residing in
hist_coords. This interface includes subroutines for each datatype
found in the hist_coord_t derived type.
. add_hist_coord interface consisting of the following two routines
. add_hist_coord_int: New routine for registering mdim and storing all
information relevant to an mdim when the dimension values are integral
. add_hist_coord_r8: New routine for registering mdim and storing all
information relevant to an mdim when the dimension values are real
. add_vert_coord: Routine for adding the vertical mdims (lev and ilev).
. write_hist_coord_att: Write the NetCDF attributes for a single hist coord
. write_hist_coord_atts: Write the NetCDF attributes for all hist coords
. write_hist_coord_var: Write the NetCDF data for a single hist coord
. write_hist_coord_vars: Write the netCDF data for all hist coords
M models/atm/cam/src/utils/grid_flag_utils.F90
. Utilities to support grid flags
col_type - number from 0-31 - used to identify which type of field (currently: 0=grid, 1=sub-col)
grid_type - integer(bit_field_kind) which contains the appropriate fields turned on/off (internally each bit of
a grid_type flag indicates on/off status for each type of field)
* clear_grid_types - clears all the flags from provided grid_types_flag
* is_col_type_set - returns true if requested col_type flag is set in provided grid_types_flag
* set_col_type - returns a grid_types integer with the requested col_type flag set
* add_col_type - adds the requested col_type to the provided grid_types_flag
* clear_col_type - clears the requested col_type fromthe provided grid_types_flag
List all existing files that have been modified, and describe the changes:
M models/atm/cam/src/control/cam_history.F90
. Changes described in the subroutine sections
. Removed unused 'use' subroutines and unused variables
. Removed explicit output of vertical or 'mdim' coordinates. This
output is now handled in the add_hist_coord and add_vert_coord interfaces.
. Moved all register_hist_coord calls to 'owner' routines
. Moved all information about specific hist coords to 'owner' routines
. Added calls to support functions to write hist coord attributes and values
M models/atm/cam/src/control/cam_history_support.F90
. Changes described in the subroutine sections
M models/atm/cam/src/control/ncdio_atm.F90
. Added optional ncolnam input to infld_real_2dncol and to
infld_real_3dncol. This input allows infld to read a variable
defined on a different grid by specifying a name other than 'ncol'.
M models/atm/cam/src/control/sat_hist.F90
. Name change of hist_mdims to hist_coords. Also %value --> %dimsize
M models/atm/cam/src/dynamics/eul/dp_coupling.F90
. Now passing array subsection in geopotential_t call
M models/atm/cam/src/dynamics/eul/dyn_comp.F90
. Moved hycoef_init call to before addfld calls (needed because
hycoef_init now defined 'lev' and 'ilev' dimensions).
M models/atm/cam/src/dynamics/fv/dp_coupling.F90
. Now passing array subsection in geopotential_t call
M models/atm/cam/src/dynamics/fv/dyn_comp.F90
. Whitespace cleanup
M models/atm/cam/src/dynamics/se/dp_coupling.F90
. Now passing array subsection in geopotential_t call
M models/atm/cam/src/dynamics/se/dyn_comp.F90
. Removed redundant subroutine 'use' statement.
M models/atm/cam/src/dynamics/se/dp_coupling.F90
. Now passing array subsection in geopotential_t call
M models/atm/cam/src/dynamics/sld/dyn_comp.F90
. Moved hycoef_init call to before addfld calls (needed because
hycoef_init now defined 'lev' and 'ilev' dimensions).
M models/atm/cam/src/physics/cam/check_energy.F90
. Arrays in arguments are now assumed shape. Local arrays are now dimensioned state%ncol instead of pcols
M models/atm/cam/src/physics/cam/cospsimulator_intr.F90
. Merged cosp_share.F90 into file. Made as many transferred variables
as possible private.
. Replaced register_hist_mdim calls with more self-contained add_hist_coord
. Replaced 'cosp_nhtml' mdim with 'lev' (change verified by Jen Kay).
. Made dimension units CF compilant (changes verified with Brian and
Jen Kay).
M models/atm/cam/src/physics/cam/geopotential.F90
. Arrays in arguments are now assumed shape. Local arrays are now dimensioned ncol instead of pcols
M models/atm/cam/src/physics/cam/physics_buffer.F90
M models/atm/cam/src/physics/cam/physics_buffer.F90.in
. Added bfg_sc to buffer field
. pbuf_add_field - added optional grid_types_flag parameter which if present adds field(s) specified
. pbuf_get_field - added optional col_types parameter which if present specifies which col_type to
retrieve, otherwise grid field is returned
. pbuf_set_field - added optional grid_types_flag paramater which if present sets the field(s) specified,
otherwise set the grid field
M models/atm/cam/src/physics/cam/radiation.F90
. use change due to cosp_share module elimination
M models/atm/cam/src/physics/rrtmg/radiation.F90
. use change due to cosp_share module elimination
M models/atm/cam/src/utils/buffer.F90
M models/atm/cam/src/utils/buffer.F90.in
. Added buffer_field_is_alloc - checks the allocation status of buffer field passed in
M models/atm/cam/src/utils/cam_pio_utils.F90
. Name change of hist_mdims to hist_coords. Also %value --> %dimsize
M models/atm/cam/src/utils/hycoef.F90
. hyai, hyam, hybi, and hybm are now targets as they are stored as
pointers in the hist_coord_t type.
. level values for 'lev' and 'ilev' are stored as new module variables,
alev and ailev (also targets).
. hycoef_init now calls add_vert_coord for 'lev' and 'ilev'.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All Passed
frankfurt/lf95: All Passed
frankfurt/pgi:
044 bl317 TBL.sh f10c5cdm outfrq3s 9s:
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
PGI with COSP test. mdim coordinate values and attributes were incorrect
in history file.
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_02
Originator(s): fischer, eaton
Date: Thu May 9 2013
One-line Summary: Merge updates from cesm1_2_0 release branches
Purpose of changes:
. Merge updates from the release branch to match cesm1_2_0_n05_cam5_3_01.
Excluding the offline radiation driver.
. Improve how CAM standalone build handles NetCDF. Improved error handling
for configure's use of the nf-config and nc-config scripts to determine
the link arguments. Also made mods to allow the user to *not* specify
NetCDF include file or library locations. This is useful for working
with compiler wrapper scripts that already have this information from the
system module settings (e.g., when using the mpif90 wrapper script on
yellowstone).
. Swapped test 735 on frankfurt with 991 on yellowstone. Did this because
The TEQ_ccsm 991 test was failing due to different intel compiler versions
being used for cam stand alone, and cesm scripts. 735 was swapped because
it's run lenght was about the same as 991. In order to do this, the pes
layout for frankfurt had to be overridden in TCB_ccsm.sh, and CCSM_MACH
had to be set for frankfurt.
. Other testing changes were turning on debugging for the WACCM-X post-tag
test on yellowstone, and add support for testing on edison.
. Updates for testing and run script example for yellowstone to set OMP_STACKSIZE and remove
INC_NETCDF and LIB_NETCDF.
. Fix units for aerosol, basically it is all variables with the name that contain "num". Units
were "kg/m2/s", should be " 1/m2/s"
. #NO_MPI2 logic was removed, and MPI2 code was removed. The MPI2 all_gather calls were failing due
to a bug in MPI2.
. SVN externals updated to match cesm1_2_beta08
Bugs fixed (include bugzilla ID):
. OMP bug fixed for WACCM-X.
. Memory leak fix for WACCM-SE.
Describe any changes made to build system:
Describe any changes made to the namelist:
. New initial condition file for ne30np4
. New tunings for rsplit and se_split to allow 7.5 minute vertical remapping for ne120np4.
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Fischer, Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/tests_pretag_frankfurt_pgi
M models/atm/cam/test/system/tests_pretag_yellowstone
. Moved the 735 tests to yellowstone. Moved 991 from yellowstone to frankfurt
M models/atm/cam/test/system/TCB_ccsm.sh
. Add the ability to run ccsm tests on frankfurt
A models/atm/cam/test/system/config_files/wx1.9c4dh
. Created a WACCM-X debug test to be run posttag on yellowstone
M models/atm/cam/test/system/test_driver.sh
. Add test support for edison.
. Set CCSM_MACH for frankfurt to allow ccsm tests to be run on frankfrut.
. Removed INC_NETCDF and LIB_NETCDF from yellowstone testing
M models/atm/cam/test/system/input_tests_master
. Turned on debugging for the WACCM-X 399 tests.
M models/atm/cam/test/system/CAM_runcmnd.sh
A models/atm/cam/test/system/tests_pretag_edison_intel
. Added support for edison testing
M models/atm/cam/bld/configure
. Update how configure deals with the NetCDF library.
- Remove the default locations of /usr/local/include and /usr/local/lib.
The compiler wrapper scripts that are used on yellowstone supply this
information, so there's no reason for configure to supply a default
that's almost never correct.
- Remove the check that the netCDF header files and libraries exist. We
may not know where they are.
- Don't fail if .mod files not found
- If the netcdf lib path is specified, then attempt to get netcdf link
args from the nf-config or nc-config scripts. But if that fails then
set defaults based on the specified library path.
- Add a -rpath setting to the link args unless the nf-config or nc-config
scripts have already set info to be passed to linker (assuming the user
has provided netcdf lib path)
- Fix bug in test code (the create mode needed to be defined).
. removed some unused code
M models/atm/cam/bld/Makefile.in
. Remove all setting of rpath for netcdf shared libs. This responsibility
has been moved to configure.
. Remove the -d option for mkDepends -- this was causing problems for the
configure tests
. Put code to append the MOD_NETCDF seach path inside conditional logic since it
may not be set.
. Need to set -DNO_C_SIZEOF to please latest PIO tag.
M models/atm/cam/bld/config_files/definition.xml
. add nc_ldflags to save args used to link NetCDF libraries.
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. New initial condition file for ne30np4 CAM5
. Set all values of rsplit to 5, and set se_nsplit to 2 for ne120np4. This
allows 7.5 minute vertical remapping.
M models/atm/cam/bld/namelist_files/use_cases/1850-PD_cam5.xml
. Renamed emis files from ar5 to RCP45
M models/atm/cam/bld/run-yellowstone.csh
. Removed LIB_NETCDF and INC_NETCDF
. Set OMP_STACKSIZE to 256M
M models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
M models/atm/cam/src/chemistry/mozart/mo_chm_diags.F90
M models/atm/cam/src/chemistry/mozart/mz_aerosols_intr.F90
. Fix units for aerosols.
M models/atm/cam/src/dynamics/se/gravity_waves_sources.F90
. Fix memory leak for WACCM-SE
M models/atm/cam/src/dynamics/se/dyn_grid.F90
M models/atm/cam/src/utils/spmd_utils.F90
. Removed #NO_MPI2
. Removed MPI2 code because of a bug in the MPI2 libraries
M models/atm/cam/src/dynamics/fv/dp_coupling.F90
. OMP bug fix, mmrSum_O_O2_H needed to be private. Changes answer for WACCM-X
M SVN_EXTERNAL_DIRECTORIES
. Externals updated to match cesm1_2_beta08.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
036 bl336 TBL.sh wm4c4h outfrq3s+waccm_1850_cam4 9s .......................FAIL! rc= 7 at Tue May 7 18:43:36 MDT 2013
051 bl373 TBL.sh fm1.9c4dh outfrq3s 9s ....................................FAIL! rc= 7 at Tue May 7 19:42:06 MDT 2013
053 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...........................FAIL! rc= 7 at Tue May 7 19:46:08 MDT 2013
055 bl376 TBL.sh fm1.9c4dh+testmech outfrq3s 9s ...........................FAIL! rc= 7 at Tue May 7 19:52:59 MDT 2013
058 bl374 TBL.sh fs1.9c4dh outfrq3s+1850-2005_cam4_super_fast_llnl 9s .....FAIL! rc= 7 at Tue May 7 20:00:27 MDT 2013
060 bl430 TBL.sh fm1.9c5dh outfrq3s_mozEOOH 9s ............................FAIL! rc= 7 at Tue May 7 20:10:01 MDT 2013
063 bl440 TBL.sh fsoa1.9c4dh outfrq3s_mozEOOH+soa_chem_megan_emis 9s ......FAIL! rc= 7 at Tue May 7 20:28:58 MDT 2013
067 bl387 TBL.sh f1.9c4dm outfrq3s+1850-2005_cam4 9s ......................FAIL! rc= 7 at Tue May 7 20:37:23 MDT 2013
073 bl379 TBL.sh wm1.9c4h outfrq24h+waccm_1850_cam4 2d ....................FAIL! rc= 7 at Tue May 7 20:57:22 MDT 2013
. Answer changes from new clm tag
079 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ....................FAIL! rc= 7 at Tue May 7 21:23:56 MDT 2013
. New test for yellowstone
081 bl740 TBL.sh h30c5h outfrq3s 9s .......................................FAIL! rc= 7 at Tue May 7 21:27:01 MDT 2013
. Expected to fail because of new initial condition file
frankfurt/lf95: All passed
frankfurt/pgi: All passed
===============================================================
===============================================================
Tag name: cam5_3_01
Originator(s): santos cbardeen hannay
Date: Apr 25, 2013
One-line Summary: Fixes for CAM-CHEM, WACCM, and CARMA issues (posttag tests).
Purpose of changes:
- Update driver and csm_share to fix some SCAM cases. (Also brings in a
driver fix for NAG builds.)
- Get CARMA working with debug flags turned on (Lahey, PGI, Intel).
- Restore WACCM-X hydrogen limiters for calls of physics_update where
hydrogen (atomic and molecular) is not being modified directly via
ptend.
- Fix issues with WACCM and CAM-CHEM debug cases.
- Fix from Chuck Bardeen for meteor_smoke and pmc CARMA models with
specified dynamics cases.
- Fix from Cecile Hannay for the units of NUMLIQ and NUMICE.
Bugs fixed (include bugzilla ID):
- SCAM cases that were over ocean, but not aquaplanet cases, were setting
lnd_present = .false., but not sno_present = .false.
- CARMA debug cases did not run.
- WACCM and some CAM-CHEM debug cases failed state_debug_check.
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
D models/atm/cam/bld/namelist_files/use_cases/arm95_scam.xml
- Replaced by scam_arm95.xml, which was added in the previous tag.
List all subroutines added and what they do:
A models/atm/cam/test/system/config_files/f1.9c5carmdusdm
- CARMA dust model with debug flags enabled.
List all existing files that have been modified, and describe the changes:
M SVN_EXTERNAL_DIRECTORIES
- Pull in update to driver and csm_share with SCAM fix.
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
- Pull in CARMA with fixes for DEBUG mode.
(See the CARMA log for details about these changes.)
M models/atm/cam/src/physics/cam/micro_mg_cam.F90
- Output different units for mass mixing ratios and number
concentrations (CLDLIQ/CLDICE vs. NUMLIQ/NUMICE).
- Change units of "#" to the CF-compliant "1".
M models/atm/cam/src/physics/cam/modal_aer_opt.F90
- Split long lines to stay under 132 character limit.
M models/atm/cam/src/physics/cam/physics_types.F90
- Remove conditionals that caused an accidental answer change for
WACCM-X in cam5_2_21. In isolation, this change restores answers
that are bit-for-bit with cam5_2_20. However, WACCM-X will not
be bit-for-bit between this tag and any previous tag, because
of other changes affecting WACCM chemistry.
M models/atm/cam/src/physics/cam/physpkg.F90
- Move clybry_fam_adj call before QNEG3, so that QNEG3 enforces
the qmin limit on the result, and the subsequent call to
physics_state_check can pass.
M models/atm/cam/src/physics/carma/cam/carma_intr.F90
- Use carma_do_thermo to prevent access of unallocated array in
the "CRTT" outfld call.
- The calls to "set_dry_to_wet" and "set_wet_to_dry" did not
respect qmin. For now, cause CARMA to operate on a temporary
copy of the state. CARMA does not deliberately modify the state;
it only produces a roundoff error due to these two conversions.
- Change ptend names to differentiate between CARMA tendencies
originating in different routines.
M models/atm/cam/src/physics/carma/models/cirrus/carma_cloudfraction.F90
- Remove unnecessary use of physics_ptend_init.
M models/atm/cam/src/physics/carma/models/meteor_smoke/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/pmc/carma_model_mod.F90
- Force internal calculations to use 365 day years, to deal with
mismatches between year lengths in CARMA and specified dynamics
files.
M models/atm/cam/test/system/input_tests_master
- Change CARMA dust test to a debug test.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
036 bl336 TBL.sh wm4c4h outfrq3s+waccm_1850_cam4 9s .......................FAIL! rc= 7 at Wed Apr 24 19:04:38 MDT 2013
073 bl379 TBL.sh wm1.9c4h outfrq24h+waccm_1850_cam4 2d ....................FAIL! rc= 7 at Wed Apr 24 21:18:15 MDT 2013
Expected small change due to moving clybry_fam_adj.
084 eq991 TEQ_ccsm.sh f19_f19 F f1.9c4m fcase 2d ..........................FAIL! rc= 7 at Wed Apr 24 21:53:28 MDT 2013
Expected failure due to mismatch between CESM compiler version and
yellowstone default.
frankfurt/lf95: All PASS.
frankfurt/pgi or jaguar/pgi: All PASS.
(There is a WACCM case on frankfurt, but it is specified chemistry, and
thus the baseline is not expected to fail.)
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
WACCM and CAM-CHEM - Small (but above roundoff) change due to reordering
of clybry_fam_adj and qneg3.
CARMA - Change due to machine roundoff in divide/multiply that is no longer
executed on the shared physics_state object.
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_3_00
Originator(s): fischer, bsingh, eaton, hannay, jet, taylor
Date: Apr 23, 2013
One-line Summary: Improve rad_diag calcs; add modal aerosol optics diagnostics
scam updates, SE bug fix
Purpose of changes:
. Remove restriction that radiation diagnostic calculations reuse the water
uptake and wet radius values calculated for the climate affecting modes.
These quantities are now recomputed for the diagnostic modes.
. Add new diagnostics for modal aerosol optics.
. Make species burdens consistent with mode burdens, i.e., only include
contributions from daylight columns.
. Add new IOP datasets, use cases, and run script for scam. Along with a
date bug fix for scam.
. Minor bug fix and external update for SE from Taylor
. Update svn externals to cesm1_2_beta06, except for pio tag. PIO has
an mpi-serial bug.
. inic files moved back to homme from se.
Bugs fixed (include bugzilla ID):
. Fix bug in configure script for SE
Describe any changes made to build system:
Describe any changes made to the namelist:
. Added a new namelist variable, history_aero_optics, to add modal aerosol
optics diagnostics to the default history fields. The existing
history_aerosol variable turns on diagnostics related to the aerosol
production and removal tendencies.
A models/atm/cam/bld/namelist_files/use_cases/scam_arm95.xml
A models/atm/cam/bld/namelist_files/use_cases/scam_arm97.xml
A models/atm/cam/bld/namelist_files/use_cases/scam_gateIII.xml
A models/atm/cam/bld/namelist_files/use_cases/scam_twp06.xml
A models/atm/cam/bld/namelist_files/use_cases/scam_togaII.xml
A models/atm/cam/bld/namelist_files/use_cases/scam_sparticus.xml
A models/atm/cam/bld/namelist_files/use_cases/scam_mpace.xml
. New use cases for scam
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: fischer
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/TBR.sh
M models/atm/cam/test/system/TER.sh
M models/atm/cam/test/system/TSM.sh
M models/atm/cam/bld/build-namelist
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
M models/atm/cam/bld/namelist_files/namelist_definition.xml
M models/atm/cam/src/physics/cam/phys_control.F90
. add namelist variable history_aero_optics
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. add namelist variable history_aero_optics
. Move inic files back to homme from se.
M models/atm/cam/bld/config_files/horiz_grid.xml
. add ncol values for se dycore -- this is only used by standalone builds
to provide info needed for the cice decomp
M models/atm/cam/bld/configure
. change set_horiz_grid to use ncol from the horiz_grid.xml file rather
than nlon and nlat when dycore is SE. A little less confusing.
M models/atm/cam/src/chemistry/utils/modal_aero_calcsize.F90
. remove restriction that dry radius is being calculated for the modes
affecting the climate. The calculation for the diagnostic modes is
returned via an optional argument rather than put in the physics buffer.
M models/atm/cam/src/chemistry/utils/modal_aero_wateruptake.F90
. remove restriction that water uptake and wet radius are being calculated
for the modes affecting the climate. The calculations for the diagnostic
modes are returned via optional arguments rather than put in the physics
buffer.
M models/atm/cam/src/physics/cam/aer_rad_props.F90
. use history_aero_optics rather than history_aerosol to add default
diagnostics to the history output
M models/atm/cam/src/physics/cam/modal_aer_opt.F90
. remove restriction that diagnostic calculations will reuse the water
uptake and wet radius values computed for the climate affecting modes.
. add diagnostic output
M models/atm/cam/src/physics/rrtmg/radconstants.F90
. add public parameters to identify nir and uv sw bands
M models/atm/cam/src/physics/cam/radconstants.F90
. add public parameters to identify nir and uv sw bands. This is only for
interface consistency; this version of radconstants shouldn't be used
when modal aerosols are in use.
M models/atm/cam/test/system/tests_pretag_yellowstone
M models/atm/cam/test/system/input_tests_master
A models/atm/cam/test/system/config_files/h30c5h
. New non aquaplanet SE test
M models/atm/cam/bld/run-scam.csh
M models/atm/cam/src/dynamics/eul/iop.F90
. Updates for scam from jet
M models/atm/cam/bld/run-yellowstone.csh
. Remove dos carriage returns
M models/atm/cam/src/dynamics/se/dyn_comp.F90
. Bug fix from Mark Taylor
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
. Update SE tag to 1_3_36 for bug fix for Taylor
M SVN_EXTERNAL_DIRECTORIES
. Update externals to match cesm1_2_beta06, except for pio, which has an
mpi-serial bug
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Tue Apr 23 11:10:56 MDT 2013
014 bl322 TBL.sh fn1.9c5dh outfrq3s 9s ....................................FAIL! rc= 7 at Tue Apr 23 11:11:00 MDT 2013
024 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...............................FAIL! rc= 7 at Tue Apr 23 11:11:03 MDT 2013
039 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ......................................FAIL! rc= 7 at Tue Apr 23 11:11:05 MDT 2013
043 bl366 TBL.sh f1.9c5clbdm outfrq3s_clubb 9s ............................FAIL! rc= 7 at Tue Apr 23 11:11:13 MDT 2013
046 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .....................................FAIL! rc= 7 at Tue Apr 23 11:11:16 MDT 2013
053 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...........................FAIL! rc= 7 at Tue Apr 23 11:11:20 MDT 2013
060 bl430 TBL.sh fm1.9c5dh outfrq3s_mozEOOH 9s ............................FAIL! rc= 7 at Tue Apr 23 11:11:28 MDT 2013
090 bl997 TBL_ccsm.sh ne16_g37 FC5 2d .....................................FAIL! rc= 7 at Tue Apr 23 11:11:51 MDT 2013
. Baseline failures are due to a change in how the diagnostic output of
species burdens (BURDENBC BURDENDUST BURDENPOM BURDENSEASALT BURDENSO4
BURDENSOA) have been redefined to be consistent with the mode burden
diagnostics which only contain contributions from daylight columns.
076 bl731 TBL.sh h16c3aqdh aqua+aquaplanet_cam3 9s ........................FAIL! rc= 7 at Tue Apr 23 11:11:33 MDT 2013
078 bl740 TBL.sh h30c5h outfrq3s 9s .......................................FAIL! rc= 7 at Tue Apr 23 11:11:33 MDT 2013
081 bl751 TBL.sh h30c4aqdm outfrq3s+aquaplanet_cam4 9s ....................FAIL! rc= 7 at Tue Apr 23 11:11:44 MDT 2013
. Bug fix for SE cause minor answer changes
078 bl740 TBL.sh h30c5h outfrq3s 9s .......................................FAIL! rc= 7 at Tue Apr 23 11:11:33 MDT 2013
. New test, no baselines
084 eq991 TEQ_ccsm.sh f19_f19 F f1.9c4m fcase 2d ..........................FAIL! rc= 7 at Tue Apr 23 11:11:47 MDT 2013
. expected, cam tests use Intel 12, cesm test use Intel 13, which have answer differences
frankfurt/lf95:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Mon Apr 22 15:12:39 MDT 2013
009 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Mon Apr 22 15:12:41 MDT 2013
013 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Mon Apr 22 15:12:42 MDT 2013
017 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Mon Apr 22 15:12:46 MDT 2013
025 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Mon Apr 22 15:13:11 MDT 2013
. Baseline failures are due to a change in how the diagnostic output of
species burdens (BURDENBC BURDENDUST BURDENPOM BURDENSEASALT BURDENSO4
BURDENSOA) have been redefined to be consistent with the mode burden
diagnostics which only contain contributions from daylight columns.
frankfurt/pgi or jaguar/pgi:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Tue Apr 23 11:17:48 MDT 2013
010 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Tue Apr 23 11:17:49 MDT 2013
016 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Tue Apr 23 11:17:52 MDT 2013
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Tue Apr 23 11:17:57 MDT 2013
031 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Tue Apr 23 11:18:22 MDT 2013
044 bl317 TBL.sh f10c5cdm outfrq3s 9s .....................................FAIL! rc= 7 at Tue Apr 23 11:18:27 MDT 2013
053 bl321 TBL.sh f10c5cdm atrain 9s .......................................FAIL! rc= 7 at Tue Apr 23 11:18:30 MDT 2013
056 bl420 TBL.sh f7mode4c5dm outfrq3s 9s ..................................FAIL! rc= 7 at Tue Apr 23 11:18:41 MDT 2013
. Baseline failures are due to a change in how the diagnostic output of
species burdens (BURDENBC BURDENDUST BURDENPOM BURDENSEASALT BURDENSO4
BURDENSOA) have been redefined to be consistent with the mode burden
diagnostics which only contain contributions from daylight columns.
===============================================================
===============================================================
Tag name: cam5_2_21
Originator(s): santos
Date: Apr 17, 2013
One-line Summary: Add debug checks to physics_update, some chores
Purpose of changes:
- Add a routine to validate physics_state objects, and provide a
namelist option to run this routine every time physics_update is
called.
- Change ptend%name for some routines to be more informative when
used in these messages.
- Fix bug in CARMA netCDF file reads.
- "endrun" is now a wrapper for "shr_sys_abort" from csm_share. This
means that ENDRUN will no longer be printed to log files; ERROR
will be printed instead.
- For CESM cases, do not re-run configure after a build is complete.
- Update mkDepends to match Machines version.
- Standardize checking of error strings with handle_errmsg.
- Move cprnc to $CESMROOT/tools, matching CESM's layout.
- This also pulls in a copy of genf90; it's convenient to have
an updated copy of genf90.pl in the same place in both CESM
and CAM tags (e.g. to auto-regenerate source code in unit
tests).
- The location of cprnc does not seem to be specified anywhere
in the CAM scripts, so this should be a "harmless" change
except in any user scripts that build and use cprnc from the
CAM source.
- Remove some "#ifdef WACCM_PHYS" in cases where it can be replaced
with a cheap runtime conditional.
- Change outputs for CAM-CHEM cases.
Bugs fixed (include bugzilla ID):
- Fixed issue with CARMA and a wrap_nf change introduced in
cam5_2_19.
Describe any changes made to build system:
Describe any changes made to the namelist:
- Add "state_debug_checks" to phys_control namelist.
- .true. turns on these checks.
- Default is .true. for cases passing "-debug" to configure,
.false. otherwise.
- Add new use_case (1850-PD_cam5.xml). This is similar to
1850-2005_cam5.xml but has data from both observations and
projections, to allow cases to be run longer.
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
A models/atm/cam/bld/namelist_files/use_cases/1850-PD_cam5.xml
- New use_case for new compset (similar to F_AMIP_CAM5 but
with added data from observations and projections).
In models/atm/cam/src/control/error_messages.F90:
handle_errmsg: Checks a string to make sure that it is empty.
If it is not empty, it is assumed to be an
error message and used to call endrun.
In models/atm/cam/src/physics/cam/physics_types.F90:
physics_state_check: Checks for invalid data in a physics_state
object. Currently limited to checking for NaN
and finiteness, in most cases.
List all existing files that have been modified, and describe the changes:
M SVN_EXTERNAL_DIRECTORIES
- Add tools/cprnc.
- Update csm_share external to get shr_assert_mod.
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
- Remove CAM copy of tools/cprnc.
M models/atm/cam/bld/Makefile.in
- Use the "-d" option of mkDepends to ensure MCT is always
built first.
- Add netCDF to RPATH for Intel builds.
M models/atm/cam/bld/build-namelist
M models/atm/cam/bld/namelist_files/namelist_definition.xml
M models/atm/cam/src/physics/cam/phys_control.F90
- Add state_debug_checks option (on by default for -debug
builds).
M models/atm/cam/bld/namelist_files/use_cases/2000_cam5_trop_moz_mam3.xml
M models/atm/cam/bld/namelist_files/use_cases/2000_cam4_trop_strat_soa.xml
M models/atm/cam/bld/namelist_files/use_cases/2000_cam5_trop_strat_mam3.xml
M models/atm/cam/bld/namelist_files/use_cases/2000_cam5_trop_strat_mam7.xml
M models/atm/cam/bld/namelist_files/use_cases/soa_chem_megan_emis.xml
- Change fincl1 output variables.
M models/atm/cam/bld/cam.buildnml.csh
- Skip configure if build is complete (but error if
config_cache.xml is missing).
M models/atm/cam/bld/mkDepends
- Updated with copy from Machines.
M models/atm/cam/src/chemistry/bulk_aero/aerosol_intr.F90
M models/atm/cam/src/physics/cam/clubb_intr.F90
M models/atm/cam/src/physics/cam/convect_deep.F90
M models/atm/cam/src/physics/cam/convect_shallow.F90
M models/atm/cam/src/physics/cam/eddy_diff.F90
M models/atm/cam/src/physics/cam/macrop_driver.F90
M models/atm/cam/src/physics/cam/micro_mg_cam.F90
M models/atm/cam/src/physics/cam/pkg_cldoptics.F90
M models/atm/cam/src/physics/waccm/iondrag.F90
M models/atm/cam/src/physics/waccmx/majorsp_diffusion.F90
- Changed handling of ptend%name so that routines no longer
need to alias themselves (e.g. MG's ptend%name was cldwat).
- Changed/removed settings of ptend%name which were redundant
or would tend to hide the true source of a change.
- Error strings coming from portable code now use
handle_errmsg.
- Replace compile-time "#ifdef WACCM_PHYS" with run-time
"if (do_waccm_phys)".
M models/atm/cam/src/control/error_messages.F90
- Add "handle_errmsg".
M models/atm/cam/src/dynamics/fv/dp_coupling.F90
M models/atm/cam/src/dynamics/se/dp_coupling.F90
- State variables initialized by dycore are now actually
*allocated* by the dycore, if supported.
- Remove temperature check that is now redundant with a check
in physpkg.
M models/atm/cam/src/dynamics/se/gravity_waves_sources.F90
M models/atm/cam/src/dynamics/se/pmgrid.F90
M models/atm/cam/src/dynamics/se/stepon.F90
- Add save statement needed by XLF on Intrepid.
M models/atm/cam/src/physics/cam/physics_types.F90
- Add new state_debug_checks option and routine that
implements it.
- Added separate routine to allocate variables needed by
WACCM and initialized by dycore (uzm, frontgf, and
frontga).
- If allocated, uzm, frontgf, and frontga are now copied
in physics_state_copy.
- Generalized pergro logic slightly so that microphysics and
macrophysics routines no longer have to set ptend%name
to cldwat.
- Simplify some constituent limiters in physics_update.
- physics_update now flushes iulog.
M models/atm/cam/src/physics/cam/physpkg.F90
- Call physics_state_check at beginning of tphysac and
tphysbc.
- Since uzm is part of the state, don't pass it as a separate
argument to qbo_relax.
- Remove/shrink some "#ifdef WACCM_PHYS" blocks.
M models/atm/cam/src/physics/cam/vertical_diffusion.F90
- Uses temporary arrays instead of ptend components to hold
local state.
- Uses handle_errmsg to check errstring.
- Replace "#ifdef WACCM_PHYS" with "if (do_waccm_phys)".
M models/atm/cam/src/physics/cam/wv_saturation.F90
- Use shr_assert_in_domain and handle_errmsg for some checks.
M models/atm/cam/src/physics/carma/cam/carma_intr.F90
M models/atm/cam/src/physics/carma/models/dust/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/meteor_smoke/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/pmc/carma_model_mod.F90
- Change CARMA netCDF reads to use netcdf module and new
wrap_nf interface.
- This is a bit hackish; wrap_nf should really just be
removed.
M models/atm/cam/src/physics/waccm/gw_drag.F90
- Check that state fields have been allocated by dycore in
gw_intr.
- Minor cleanup in preparation of some strings used as field
names in outfld (done while debugging a memory issue that
was actually caused by code elsewhere).
M models/atm/cam/src/physics/waccm/qbo.F90
- Use state%uzm directly instead of requiring it as a
separate argument to qbo_relax, and check that it is
allocated first.
M models/atm/cam/src/utils/abortutils.F90
- endrun is now just a wrapper for shr_sys_abort.
M models/atm/cam/src/physics/cam/zm_conv_intr.F90
- Comment change.
M models/atm/cam/test/system/CAM_runcmnd.sh
M models/atm/cam/test/system/archive_baseline.sh
M models/atm/cam/test/system/test_driver.sh
- Remove bluefire.
- test_driver.sh will set the wallclock limit to 2:00 if
using the "small" (debug) queue on yellowstone.
- Intel uses -rpath (like other compilers) rather than adding
$LIB_NETCDF to $LD_LIBRARY_PATH.
M models/atm/cam/test/system/config_files/fst7mode1.9c5h
M models/atm/cam/test/system/config_files/fsoa1.9c4dh
M models/atm/cam/test/system/config_files/fsoa4c4dm
M models/atm/cam/test/system/config_files/fst7mode4c5dm
- Add option -age_of_air_trcs (now required by use_cases).
M models/atm/cam/tools/README
- Remove description of cprnc.
Additionally, a number of files with the "svn:executable" property
were not actually executables. This property was removed from files
that were not scripts.
Having svn:mergeinfo on subdirectories can cause merge problems, so
this property was removed from all subdirectories.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All PASS.
Note: Intel compiler version was set to 13 to get the following CESM
comparison test to pass:
082 eq991 TEQ_ccsm.sh f19_f19 F f1.9c4m fcase 2d
frankfurt/lf95: All PASS.
frankfurt/pgi or jaguar/pgi: All PASS.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
This tag is bit-for-bit except:
cam5_2_19 introduced a roundoff-level change in "igamma" for compilers
that provide a gamma intrinsic function. This only affects the CARMA
dust and seasalt models, and was exposed in this tag (since CARMA
cases did not work in cam5_2_19 and cam5_2_20).
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_2_20
Originator(s): fischer
Date: Fri Apr 12 09:42:31 MDT 2013
One-line Summary: rename homme to se, cice decomp fix
Purpose of changes:
. HOMME is now referred to SE. The code is changed to reflect this.
If -dyn homme is used, you'll get a warning message and the dynamics
will be automatically switched to se.
. Fixes for calls to generate_cice_decomp.pl and cice configure
to handel changes with a new cice tag
. Update scripts tag to fix T42_T42 grid
. Couple minor fixes to test scripts
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/TCB.sh
. Remove cice decomp calls from test scripts, let bld/configure handel the cice decomp.
M models/atm/cam/test/system/test_driver.sh
. Changed path logic for submit scripts
M models/atm/cam/test/system/TSC.sh
. Bug fix to return error code from second call to TSM.sh
M models/atm/cam/test/system/TR8.sh
M models/atm/cam/test/system/config_files/h30c4aqdm
M models/atm/cam/test/system/config_files/hn16c5aqdm
M models/atm/cam/test/system/config_files/h16c3aqdh
M models/atm/cam/test/system/config_files/h16c5aqdm
M models/atm/cam/tools/interpic_new/README
M models/atm/cam/bld/config_files/horiz_grid.xml
M models/atm/cam/bld/config_files/definition.xml
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
M models/atm/cam/bld/namelist_files/namelist_definition.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_sc_2000_cam4.xml
M models/atm/cam/bld/build-namelist
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
M models/atm/cam/src/utils/cam_pio_utils.F90
M models/atm/cam/src/physics/cam/physpkg.F90
M models/atm/cam/src/chemistry/mozart/short_lived_species.F90
. rename homme to se
M models/atm/cam/bld/configure
. Rename homme to se. If homme dynamics is selected, print warning message and set to se.
. Fix cice decomp for cam stand alone scripts.
D models/atm/cam/bld/config_files/defaults_homme.xml
A models/atm/cam/bld/config_files/defaults_se.xml
. Rename file with homme to se
D models/atm/cam/src/dynamics/homme/pmgrid.F90
D models/atm/cam/src/dynamics/homme/dycore.F90
D models/atm/cam/src/dynamics/homme/trunc.F90
D models/atm/cam/src/dynamics/homme/gravity_waves_sources.F90
D models/atm/cam/src/dynamics/homme/nctopo_util_mod.F90
D models/atm/cam/src/dynamics/homme/dyn_grid.F90
D models/atm/cam/src/dynamics/homme/interp_mod.F90
D models/atm/cam/src/dynamics/homme/README
D models/atm/cam/src/dynamics/homme/native_mapping.F90
D models/atm/cam/src/dynamics/homme/initcom.F90
D models/atm/cam/src/dynamics/homme/dp_coupling.F90
D models/atm/cam/src/dynamics/homme/stepon.F90
D models/atm/cam/src/dynamics/homme/inidat.F90
D models/atm/cam/src/dynamics/homme/dyn_comp.F90
D models/atm/cam/src/dynamics/homme/restart_dynamics.F90
D models/atm/cam/src/dynamics/homme/spmd_dyn.F90
D models/atm/cam/src/dynamics/homme/inital.F90
A models/atm/cam/src/dynamics/se/dycore.F90
A models/atm/cam/src/dynamics/se/nctopo_util_mod.F90
A models/atm/cam/src/dynamics/se/README
A models/atm/cam/src/dynamics/se/interp_mod.F90
A models/atm/cam/src/dynamics/se/native_mapping.F90
A models/atm/cam/src/dynamics/se/dp_coupling.F90
A models/atm/cam/src/dynamics/se/dyn_comp.F90
A models/atm/cam/src/dynamics/se/stepon.F90
A models/atm/cam/src/dynamics/se/restart_dynamics.F90
A models/atm/cam/src/dynamics/se/spmd_dyn.F90
A models/atm/cam/src/dynamics/se/inital.F90
. Move homme directory to se, and rename homme to se
M SVN_EXTERNAL_DIRECTORIES
. Update scripts to fix T42_T42 grid
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all passed
frankfurt/lf95: all passed
frankfurt/pgi: all passed
===============================================================
===============================================================
Tag name: cam5_2_19
Originator(s): santos
Date: Apr 9, 2013
One-line Summary: NAG port
Purpose of changes:
This allows CAM to compile with the NAG Fortran compiler. The
build will not quite work out of the box; as of this tag, CLM
and CICE tags still must be updated, and MCT requires a patch
to remove "intent" attributes from the MPI interfaces in
m_mpif90.F90.
Not all configurations are ported; the following are still
untested or works in progress:
- OpenMP threading
- WACCM-X
- CARMA
- CLUBB
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
- Add NAG support.
- Get netCDF linking flags from nc-config/nf-config.
- This should also allow the use of a shared library.
- Add more debug flags to CAM Makefile for DEBUG cases:
- Add "-g" to C code (most compilers).
- Configure MCT with debugging enabled.
- Update most externals to NAG-compatible versions.
- All externals are between their cesm1_2_beta05
and cesm1_2_beta06 versions.
- This required the addition of a wave component (swav).
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
D models/atm/cam/src/utils/error_function.F90
D models/atm/cam/src/utils/gamma_function.F90
- Functionality moved to shr_spfn_mod (csm_share module).
find_group_name (models/atm/cam/src/utils/namelist_utils.F90)
- Moved to shr_nl_mod (csm_share module).
List all subroutines added and what they do:
A models/atm/cam/test/system/tests_pretag_frankfurt_nag
- NAG pretag test list. Duplicate of Lahey pretag tests.
A models/atm/cam/bld/get_nxny.pl
- Script to get nx and ny values from "config_grid.xml" in
the CESM scripts. These values are needed because they
must be passed to CICE configure scripts.
List all existing files that have been modified, and describe the changes:
M SVN_EXTERNAL_DIRECTORIES
- Update externals to very recent versions (between cesm1_2_beta05
and cesm1_2_beta06).
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
- Update externals to NAG-compatible versions.
M models/atm/cam/bld/Makefile.in
M models/atm/cam/bld/config_files/definition.xml
M models/atm/cam/bld/configure
- Add nagfor to compilers.
- Use netCDF config script to get linking flags.
- Add -g to CFLAGS for DEBUG builds.
- Add --enable-debugging option to MCT for DEBUG builds.
- For NAG, set MCT CFLAGS (override defaults), since otherwise
there will be binary incompatibilities and/or spurious test
failures in MCT's configure.
- Update configure for standalone build with new wav component
and CICE configure changes.
=====================
There are a number of short changes that had to be made to large numbers of
files. The following is a list of these changes, followed by the list of
all files that had only these changes made. These changes may have also
affected other files.
Changes regarding syntax and Fortran standard compliance:
- Where a module has both a save statement, and variables declared with the
save attribute individually, remove the individual save and leave only
the module-scope save.
- Additionally, add save statement to some modules where it was missing.
- Additionally, move include statements of mpif.h above any save
statements in a module, since mpif.h may include variables with
the "save" attribute specified individually.
- Where a procedure gets two arguments, and one is a dimension of the other,
the arguments may be in either order in the interface, but the dimension
must be declared first.
- Break up source code lines longer than the standard length of 132
characters (72 for fixed-form source).
- Change non-standard format specifiers to standard equivalents.
- Always pass reals to procedures requiring a real argument; do not assume
integers will be converted to real automatically.
- Pass whole arrays or appropriate slices, rather than passing the first
element of an array under the assumption that a pointer to the correct
array location will be passed.
- Correct some "intent" attributes.
Changes to pointer use:
- Default-initialize pointers to null() where NAG had detected at runtime
that they were being used uninitialized.
- Add "target" attribute to objects for which persistent pointers are
needed.
- In cases where pointer dummy arguments must remain associated with some
memory upon entry to a procedure, declare the dummy arguments with
"intent(in)" or "intent(inout)", not "intent(out)", regardless of how
that memory is used.
- Generally, this does not require changes to pointer dummy arguments,
since Fortran 95 (and therefore the Lahey compiler) does not allow
any intent to be specified on pointer dummy arguments. However, some
derived types have pointer components, and dummy arguments of these
types must have the correct intent.
Changes to use of libraries and intrinsics:
- Get F2008 intrinsics that are not present on all compilers (e.g. gamma,
erfc) from shr_spfn_mod, rather than assuming that all compilers do or
do not have them.
- When setting IEEE NaN or Infinity, use the new version of shr_infnan_mod,
which requires the operator "assignment(=)" to be used.
- "bigint" was also replaced with huge(1).
- In some cases, "inf" had to be changed to "posinf" to deal with a
bug in PGI on Frankfurt.
- Where the netCDF F77 bindings are used, change to the netCDF F90
bindings, and also replace any uses of wrap_nf in those modules.
- Change non-standard complex number intrinsics to generic standard
intrinsics (e.g. imag to aimag, dcmplx to cmplx).
- Use CAM's "endrun", not "abort" or "exit" (non-standard extensions).
Other changes:
- Where required by the above changes, change dimensions of arrays.
(Usually, from explicit or assumed size to assumed shape.)
- Add some changes that would have allowed problems described above to
be detected on other compilers (initialize pointer components to null,
add explicit intent on procedures).
- Remove unnecessary code that would otherwise require one of the above
changes to be made.
*A quick note on external libraries in Fortran:
It is technically not standard-conforming to use generic functions without
explicit interfaces, as in the netCDF and MPI F77 bindings. For netCDF, it
is easy to switch to the Fortran 90 bindings. For MPI, there are some
potential issues with using the MPI module, so instead the "-wmismatch"
argument is used to turn off some of nagfor's compile-time checks.
Files:
M models/atm/cam/src/chemistry/bulk_aero/dust_intr.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_gasaerexch.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_rename.F90
M models/atm/cam/src/chemistry/mozart/cfc11star.F90
M models/atm/cam/src/chemistry/mozart/chemistry.F90
M models/atm/cam/src/chemistry/mozart/linoz_data.F90
M models/atm/cam/src/chemistry/mozart/mo_ghg_chem.F90
M models/atm/cam/src/chemistry/mozart/mo_jlong.F90
M models/atm/cam/src/chemistry/mozart/mo_mass_xforms.F90
M models/atm/cam/src/chemistry/mozart/mo_photo.F90
M models/atm/cam/src/chemistry/mozart/mo_setinv.F90
M models/atm/cam/src/chemistry/mozart/mo_sulf.F90
M models/atm/cam/src/chemistry/mozart/mo_usrrxt.F90
M models/atm/cam/src/chemistry/mozart/short_lived_species.F90
M models/atm/cam/src/chemistry/mozart/tracer_cnst.F90
M models/atm/cam/src/chemistry/mozart/tracer_srcs.F90
M models/atm/cam/src/chemistry/mozart/upper_bc.F90
M models/atm/cam/src/chemistry/utils/modal_aero_calcsize.F90
M models/atm/cam/src/chemistry/utils/modal_aero_wateruptake.F90
M models/atm/cam/src/chemistry/utils/msise00.F90
M models/atm/cam/src/chemistry/utils/solar_data.F90
M models/atm/cam/src/control/cam_comp.F90
M models/atm/cam/src/control/cam_history.F90
M models/atm/cam/src/control/cam_restart.F90
M models/atm/cam/src/control/camsrfexch.F90
M models/atm/cam/src/control/history_defaults.F90
M models/atm/cam/src/control/physconst.F90
M models/atm/cam/src/control/rgrid.F90
M models/atm/cam/src/control/sat_hist.F90
M models/atm/cam/src/control/scamMod.F90
M models/atm/cam/src/control/wrap_nf.F90
M models/atm/cam/src/cpl_mct/atm_comp_mct.F90
M models/atm/cam/src/dynamics/eul/comspe.F90
M models/atm/cam/src/dynamics/eul/courlim.F90
M models/atm/cam/src/dynamics/eul/dyndrv.F90
M models/atm/cam/src/dynamics/eul/eul_control_mod.F90
M models/atm/cam/src/dynamics/eul/forecast.F90
M models/atm/cam/src/dynamics/eul/getinterpnetcdfdata.F90
M models/atm/cam/src/dynamics/eul/iop.F90
M models/atm/cam/src/dynamics/eul/prognostics.F90
M models/atm/cam/src/dynamics/eul/restart_dynamics.F90
M models/atm/cam/src/dynamics/eul/scanslt.F90
M models/atm/cam/src/dynamics/eul/spmd_dyn.F90
M models/atm/cam/src/dynamics/eul/stepon.F90
M models/atm/cam/src/dynamics/fv/ctem.F90
M models/atm/cam/src/dynamics/fv/dp_coupling.F90
M models/atm/cam/src/dynamics/fv/dyn_comp.F90
M models/atm/cam/src/dynamics/fv/inidat.F90
M models/atm/cam/src/dynamics/fv/metdata.F90
M models/atm/cam/src/dynamics/fv/par_xsum.F90
M models/atm/cam/src/dynamics/fv/spmd_dyn.F90
M models/atm/cam/src/dynamics/fv/stepon.F90
M models/atm/cam/src/dynamics/fv/sw_core.F90
M models/atm/cam/src/dynamics/homme/dp_coupling.F90
M models/atm/cam/src/dynamics/homme/dyn_grid.F90
M models/atm/cam/src/dynamics/homme/interp_mod.F90
M models/atm/cam/src/dynamics/homme/native_mapping.F90
M models/atm/cam/src/dynamics/sld/comspe.F90
M models/atm/cam/src/dynamics/sld/courlim.F90
M models/atm/cam/src/dynamics/sld/inidat.F90
M models/atm/cam/src/dynamics/sld/prognostics.F90
M models/atm/cam/src/dynamics/sld/restart_dynamics.F90
M models/atm/cam/src/dynamics/sld/scan2.F90
M models/atm/cam/src/dynamics/sld/scanslt.F90
M models/atm/cam/src/dynamics/sld/spmd_dyn.F90
M models/atm/cam/src/dynamics/sld/stepon.F90
M models/atm/cam/src/physics/cam/aer_rad_props.F90
M models/atm/cam/src/physics/cam/boundarydata.F90
M models/atm/cam/src/physics/cam/cam3_aero_data.F90
M models/atm/cam/src/physics/cam/cam_diagnostics.F90
M models/atm/cam/src/physics/cam/chem_surfvals.F90
M models/atm/cam/src/physics/cam/cldwat2m_macro.F90
M models/atm/cam/src/physics/cam/cloud_diagnostics.F90
M models/atm/cam/src/physics/cam/comsrf.F90
M models/atm/cam/src/physics/cam/convect_shallow.F90
M models/atm/cam/src/physics/cam/cospsimulator_intr.F90
M models/atm/cam/src/physics/cam/diffusion_solver.F90
M models/atm/cam/src/physics/cam/eddy_diff.F90
M models/atm/cam/src/physics/cam/macrop_driver.F90
M models/atm/cam/src/physics/cam/micro_mg1_0.F90
M models/atm/cam/src/physics/cam/micro_mg1_5.F90
M models/atm/cam/src/physics/cam/micro_mg_cam.F90
M models/atm/cam/src/physics/cam/microp_aero.F90
M models/atm/cam/src/physics/cam/modal_aer_opt.F90
M models/atm/cam/src/physics/cam/molec_diff.F90
M models/atm/cam/src/physics/cam/ndrop.F90
M models/atm/cam/src/physics/cam/ndrop_bam.F90
M models/atm/cam/src/physics/cam/physics_types.F90
M models/atm/cam/src/physics/cam/physpkg.F90
M models/atm/cam/src/physics/cam/radae.F90
M models/atm/cam/src/physics/cam/radiation.F90
M models/atm/cam/src/physics/cam/radiation_data.F90
M models/atm/cam/src/physics/cam/tropopause.F90
M models/atm/cam/src/physics/cam/uwshcu.F90
M models/atm/cam/src/physics/carma/cam/carma_intr.F90
M models/atm/cam/src/physics/carma/models/cirrus/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/dust/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/sea_salt/carma_model_mod.F90
M models/atm/cam/src/physics/rrtmg/radsw.F90
M models/atm/cam/src/physics/rrtmg/rrtmg_state.F90
M models/atm/cam/src/physics/waccm/gw_drag.F90
M models/atm/cam/src/physics/waccm/iondrag.F90
M models/atm/cam/src/physics/waccm/qbo.F90
M models/atm/cam/src/physics/waccm/waccm_forcing.F90
M models/atm/cam/src/physics/waccmx/ionosphere.F90
M models/atm/cam/src/utils/abortutils.F90
M models/atm/cam/src/utils/cam_dom/ocn_spmd.F90
M models/atm/cam/src/utils/cam_dom/ocn_time_manager.F90
M models/atm/cam/src/utils/cam_dom/sst_data.F90
M models/atm/cam/src/utils/pilgrim/decompmodule.F90
M models/atm/cam/src/utils/pilgrim/mod_comm.F90
M models/atm/cam/src/utils/spmd_utils.F90
M models/atm/cam/src/utils/time_manager.F90
=====================
M models/atm/cam/src/chemistry/mozart/mo_drydep.F90
- Change leading dimension of dummy argument rh to "ncol", to
match the dimensions of the actual argument.
M models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
- Ensure that extfrc and reaction_rates always have non-zero
size. (When extcnt or rxntot are 0, the arrays will have
a trailing dimension of 1, but will not actually be used.)
- Size zero arrays are not a problem with NAG, but these
arrays have default values assigned, and this cannot be
done for arrays with no elements to assign.
M models/atm/cam/src/chemistry/mozart/mo_setsox.F90
- Fix a case where a variable that was only set for 7-mode MAM
was being referenced in all MAM cases.
- It appears that only 7-mode MAM would actually enter the
block where this reference occurred, so this caused no
apparent problems and passed runtime checks on other
compilers.
M models/atm/cam/src/chemistry/mozart/wei96.F90
- Remove unnecessary common block that impaired readability
during debugging.
M models/atm/cam/src/chemistry/utils/tracer_data.F90
- Implemented workaround for a (still poorly understood) bug
with the netCDF builds on Frankfurt, where netCDF stores
a pointer to output dimension IDs and then overwrites them.
- Change "write(*,*)" to "write(iulog,*)".
- Nullify pointer in specify_fields if it will not be
allocated.
M models/atm/cam/src/control/infnan.F90
- Internal variables replaced with references to shr_infnan_mod.
- bigint removed; integers that were set to bigint are
now set to huge(1).
M models/atm/cam/src/physics/cam/physics_buffer.F90
M models/atm/cam/src/physics/cam/physics_buffer.F90.in
M models/atm/cam/src/utils/buffer.F90
M models/atm/cam/src/utils/buffer.F90.in
- Cast physics buffer pointers back to their original type (and
thus correct size) before deallocating.
M models/atm/cam/src/physics/cam/rad_constituents.F90
- Correct string with subroutine name.
M models/atm/cam/src/physics/cam/restart_physics.F90
- Randomly shuffle use statements to avoid bug with PGI on
Frankfurt.
M models/atm/cam/src/physics/cam/vertical_diffusion.F90
- Remove assumption that constituents 2 and 3 are cldliq and
cldice.
M models/atm/cam/src/utils/cam_dom/ocn_comp.F90
- Close I/O unit before attempting to reuse it to open a new
file.
M models/atm/cam/src/utils/cam_pio_utils.F90
- Fix integer overflow in find_iodesc.
M models/atm/cam/src/utils/namelist_utils.F90
- Now only a wrapper for shr_nl_mod.
M models/atm/cam/test/system/CAM_runcmnd.sh
M models/atm/cam/test/system/TBR.sh
M models/atm/cam/test/system/TER.sh
M models/atm/cam/test/system/TSM.sh
M models/atm/cam/test/system/test_driver.sh
- Add NAG to regression test compilers on Frankfurt.
- For some cases, NAG requires ocn_modelio.nml to be written.
- Set "ulimit -c unlimited" for Frankfurt, since this is generally
desirable if a regression test fails, and does not seem to be
inherited from the user's environment.
M models/atm/cam/test/system/TCB.sh
M models/atm/cam/test/system/TCB_ccsm.sh
- Changes necessary to perform standalone build with new CICE,
wav component, and scripts.
Some files with purely whitespace or comment changes:
M models/atm/cam/src/chemistry/modal_aero/modal_aero_coag.F90
M models/atm/cam/tools/interpic_new/interpolate_data.F90
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
082 eq991 TEQ_ccsm.sh f19_f19 F f1.9c4m fcase 2d ..........................FAIL! rc= 7 at Tue Apr 9 05:22:41 MDT 2013
085 bl992 TBL_ccsm.sh f19_g16 ETEST 2d ....................................FAIL! rc= 7 at Tue Apr 9 05:27:33 MDT 2013
The updated version of Machines has a new version of the Intel compilers on
yellowstone. It was therefore expected that some TBL_ccsm tests might fail.
Additionally, TEQ_ccsm tests will fail until either version 13 of the Intel
compiler becomes the default on yellowstone, or test_driver.sh is updated
to specify version 13.
The TEQ_ccsm test did in fact pass with the compiler version bumped:
082 eq991 TEQ_ccsm.sh f19_f19 F f1.9c4m fcase 2d ..........................PASS at Tue Apr 9 11:09:14 MDT 2013
frankfurt/lf95: All PASS.
frankfurt/pgi or jaguar/pgi: All PASS.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
All CAM configurations are bit-for-bit, excepting CESM cases with changes
due to updated Machines.
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_2_18
Originator(s): santos, mmills, bardeenc, hannay
Date: Apr 8, 2013
One-line Summary: Update externals and namelist for CARMA and HOMME
Purpose of changes:
- Update CARMA external with several bug fixes. These are described in the
CARMA ChangeLog at:
models/atm/cam/src/physics/carma/base/ChangeLog
- Add use cases for new WACCM/CARMA sulfur and nuclear winter compsets.
- Source code updates for CARMA sulfate model to interact with chemistry.
- Update HOMME with NAG-compatible version and bug fix for vector
Laplacian operation.
- Updates to HOMME namelist (see below). The most important change is that
the vertically Lagrangian advection method is now the default.
- Update namelist_definition.xml to describe HOMME namelist settings in
more detail (some descriptions and assistance from Mark Taylor).
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
- For HOMME cases, we use the flag "-DHAVE_F2003_PTR_BND_REMAP", as F2003
pointer bounds remapping appears to be supported on all current
compilers (except Lahey, which we no longer test with HOMME). If this
causes problems, this flag will have to be set on a per-compiler basis
in CAM's Makefile.in and Machines.
Describe any changes made to the namelist:
- New CARMA sulfate model flag "carma_do_hetchem" toggles the use of CARMA
sulfate surface area density in heterogeneous chemistry rate
calculations.
- SO2 forcing is no longer automatically added for waccm_mozart_sulfur.
- Many HOMME changes:
- Switch from Eulerian to Lagrangian vertical advection method.
- Turn on energy fixer.
- Correct nu_p value for ne30np4.
- Correct both nu_p and nu_div for ne16np4 (necessary for stability).
- Set default nu_q to -1.D0, which automatically sets nu_q = nu within
HOMME.
- ne120np4 with vertically Lagrangian code has experimental settings
from John Truesdale.
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
A models/atm/cam/bld/namelist_files/use_cases/sd_waccm_sulfur.xml
A models/atm/cam/bld/namelist_files/use_cases/waccm_cam4_nuclear_winter.xml
- Use cases for SD-WACCM sulfur and nuclear winter compsets.
List all existing files that have been modified, and describe the changes:
M SVN_EXTERNAL_DIRECTORIES
- Update csm_share external to get new shr_spfn_mod module. This is
necessary for the NAG port update, including the HOMME external
pulled in in this tag.
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
- Update CARMA & HOMME externals.
M models/atm/cam/bld/Makefile.in
- Define CPRLAHEY and CPRGNU for the respective compilers; used by
the new csm_share.
M models/atm/cam/bld/build-namelist
- Add carma_do_hetchem to CARMA sulfate model.
- Remove logic that always adds an SO2 forcing file for
waccm_mozart_sulfur.
M models/atm/cam/bld/configure
- Add "-DHAVE_F2003_PTR_BND_REMAP" to HOMME flags.
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- Changes to HOMME namelist.
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- Add carma_do_hetchem.
- Update HOMME namelist documentation.
M models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
M models/atm/cam/src/chemistry/mozart/mo_usrrxt.F90
M models/atm/cam/src/physics/cam/carma_flags_mod.F90
M models/atm/cam/src/physics/carma/models/sulfate/carma_model_mod.F90
- If carma_do_hetchem is .true., use CARMA sulfate surface area
density in mozart chemistry.
M models/atm/cam/src/physics/carma/cam/carma_intr.F90
- Minor tweaks (e.g. radiation_scheme is now discovered through
phys_getopts rather than guessing based on the number of bands).
M models/atm/cam/src/physics/carma/models/bc_strat/carma_model_mod.F90
- Remove comment that should have only been in the cirrus model.
M models/atm/cam/src/physics/carma/models/dust/carma_model_mod.F90
- Fix comment typo.
D models/atm/cam/test/system/config_files/f1.9c4carmsulm
A models/atm/cam/test/system/config_files/fsd1.9c4wcarmsulm
D models/atm/cam/test/system/config_files/f1.9c5carmbc_m
A models/atm/cam/test/system/config_files/f1.9c4wcarmbc_m
M models/atm/cam/test/system/input_tests_master
D models/atm/cam/test/system/nl_files/carma24h
A models/atm/cam/test/system/nl_files/outfrq24h_carma
- CARMA test changes:
- Remove old sulfate and bc_strat tests and replace them with
WACCM tests that more closely match the new CESM compsets
(including the new use cases).
- Rename carma24h to outfrq24h_carma (similar to outfrq3s_*).
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
076 bl731 TBL.sh h16c3aqdh aqua+aquaplanet_cam3 9s ........................FAIL! rc= 7 at Sun Apr 7 22:31:44 MDT 2013
079 bl751 TBL.sh h30c4aqdm outfrq3s+aquaplanet_cam4 9s ....................FAIL! rc= 7 at Sun Apr 7 22:52:24 MDT 2013
088 bl997 TBL_ccsm.sh ne16_g37 FC5 2d .....................................FAIL! rc= 7 at Sun Apr 7 23:13:39 MDT 2013
frankfurt/lf95: All tests PASS
frankfurt/pgi or jaguar/pgi:
059 bl734 TBL.sh hn16c5aqdm outfrq3s_bam+aquaplanet_cam5 9s ...............FAIL! rc= 7 at Sun Apr 7 21:28:48 MDT 2013
062 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ....................FAIL! rc= 7 at Sun Apr 7 22:43:14 MDT 2013
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- All HOMME cases and CARMA cirrus and sulfate cases.
- what platforms/compilers: All
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
- Above roundoff, but same climate.
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_2_17
Originator(s): fischer, bsingh, eaton,
Date: April 5, 2013
One-line Summary: Prescribed MAM update
Purpose of changes:
. Merge in latest changes for prescribed MAM.
. Add default namelist support for prescribed MAM.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
A models/atm/cam/bld/namelist_files/use_cases/1850_cam5_pm.xml
. new use_case for 1850 prescribed MAM
A models/atm/cam/test/system/config_files/fn1.9c5dh
M models/atm/cam/test/system/tests_pretag_yellowstone
M models/atm/cam/test/system/input_tests_master
. new prescribed MAM test
M models/atm/cam/test/system/test_driver.sh
. Fixed frankfurt intel tests
M models/atm/cam/bld/Makefile.in
. Removed -DNO_MPI2 rule for dyn_grid.F90
M models/atm/cam/src/dynamics/homme/dyn_grid.F90
M models/atm/cam/src/utils/spmd_utils.F90
. Add #define DNO_MPI2 1
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. Add checks for prescribed bulk/modal aerosol variables
M models/atm/cam/bld/build-namelist
. Updated prescribed mam variable list
. Fixed logic to allow prescribed bulk and modal aerosol namelist
varibles to be set to defaults
M models/atm/cam/bld/perl5lib/Build/Namelist.pm
. Modify Namelist.pm to recognize comment lines in namelist input files.
The previous code only recognized end of line comments.
M models/atm/cam/src/physics/cam/convect_shallow.F90
. Changes to option name for physics_ptend_init
M models/atm/cam/src/physics/cam/gw_drag.F90
. Changes to option name for physics_ptend_init
M models/atm/cam/src/chemistry/utils/tracer_data.F90
. Change scalar file%in_pbuf to an array.
M models/atm/cam/src/chemistry/utils/prescribed_aero.F90
. Merge in changes from cam1/branch_tags/aerosol_tags/aerosol002_cam5_2_10
. Allocate file%in_pbuf and initialize
M models/atm/cam/src/physics/waccm/waccm_forcing.F90
M models/atm/cam/src/chemistry/utils/aerodep_flx.F90
M models/atm/cam/src/chemistry/utils/prescribed_ozone.F90
M models/atm/cam/src/chemistry/utils/prescribed_ghg.F90
M models/atm/cam/src/chemistry/utils/aircraft_emit.F90
M models/atm/cam/src/chemistry/utils/tracer_data.F90
M models/atm/cam/src/chemistry/utils/prescribed_volcaero.F90
M models/atm/cam/src/chemistry/mozart/mo_tgcm_ubc.F90
M models/atm/cam/src/chemistry/mozart/mo_sulf.F90
M models/atm/cam/src/chemistry/mozart/tracer_cnst.F90
M models/atm/cam/src/chemistry/mozart/tracer_srcs.F90
M models/atm/cam/src/chemistry/mozart/mo_extfrc.F90
M models/atm/cam/src/chemistry/mozart/linoz_data.F90
M models/atm/cam/src/chemistry/mozart/mo_srf_emissions.F90
. Allocate file%in_pbuf and initialize
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
014 bl322 TBL.sh fn1.9c5dh outfrq3s 9s ....................................FAIL! rc= 7 at Thu Apr 4 21:56:30 MDT 2013
. New test for prescribed MAM
frankfurt/lf95: all passed
frankfurt/pgi: all passed
===============================================================
===============================================================
Tag name: cam5_2_16
Originator(s): Cheryl Craig, Pete Bogenschutz and Sean Santos
Date: April 1, 2013
One-line Summary: CLUBB tunings for scientifically validated release
Purpose of changes:
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
- Introduced sol_facti_cloud_borne to allow tuning, default value is 1.0
- Changed tuning of dust_emis_fact_dyn for CLUBB
- Changed tuning of sol_facti_cloud_borne for CLUBB
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
- Utilizing modified CLUBB library
M models/atm/cam/src/chemistry/mozart/mz_aerosols_intr.F90
M models/atm/cam/src/chemistry/mozart/chemistry.F90
- Introduced sol_facti_cloud_borne tuning parameter
M models/atm/cam/src/physics/cam/clubb_intr.F90
M models/atm/cam/src/physics/cam/micro_mg1_0.F90
M models/atm/cam/src/physics/cam/micro_mg_cam.F90
- CLUBB tunings
M models/atm/cam/src/physics/cam/microp_aero.F90
- Temporary CLUBB tuning
M models/atm/cam/src/physics/cam/micro_mg1_5.F90
- CLUBB tunings and revised routine from Sean
M models/atm/cam/src/physics/cam/convect_deep.F90
- Moved initializations of prec and snow to after they are acquired
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: All BFB except CLUBB, which is expected to fail due to new tunings
040 bl366 TBL.sh f1.9c5clbdm outfrq3s_clubb 9s ............................FAIL! rc= 7 at Sat Mar 30 01:45:03 MDT 2013
frankfurt/lf95: All BFB
frankfurt/pgi or jaguar/pgi: All BFB
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_2_15
Originator(s): fvitt, tilmes
Date: 29 Mar 2013
One-line Summary: CAM-Chem and WACCM updates
Purpose of changes:
- call endrun when lower boundary conditions are specified
in more than one manner via bndtvghg, flbc_file, or *vmr namelist
variables to prevent multiple specification of lower boundary conditions
- remove deprecated waccm_mozart_v1 chemistry package
- replace chemistry package trop_strat_bam_v1 with trop_strat_soa
- update chemistry in packages:
trop_mozart_mam3
trop_mozart_mam7
trop_mozart_soa
- set default nspltvrm to 2 for waccm_phys ( waccm and waccmx )
- corrections to surface area calculations of modal aerosols
- correction to units of the SOA*_PROD diagnostics
- restrict message logging to masterproc in many of the chemistry modules
- update to MEGAN factors input file
- update to solar protons input file
- update to SAD input file
- update to sulfate file
- new build-namelist use cases:
2000_cam5_trop_moz_mam3
2000_cam4_trop_strat_soa
2000_cam5_trop_strat_mam3
2000_cam5_trop_strat_mam7
Bugs fixed (include bugzilla ID):
Bug #: 1639
Summary: reduce stdout from cam
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
D models/atm/cam/src/chemistry/pp_waccm_mozart_v1
- deprecated chemistry package removed
D models/atm/cam/src/chemistry/pp_trop_strat_bam_v1
- mechanism placed by pp_trop_strat_soa
D models/atm/cam/src/chemistry/utils/mo_regrider.F90
- removed this deprecated module that only works on regular lat/lon grids
List all subroutines added and what they do:
A models/atm/cam/test/system/nl_files/outfrq3s_mozEOOH
- added for mechanisms that have EOOH chemical tracer
A models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_lu_solve.F90
A models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_setrxt.F90
A models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_adjrxt.F90
A models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_rxt_rates_conv.F90
A models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_imp_sol.F90
A models/atm/cam/src/chemistry/pp_trop_strat_soa/chem_mods.F90
A models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_prod_loss.F90
A models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_lin_matrix.F90
A models/atm/cam/src/chemistry/pp_trop_strat_soa/m_rxt_id.F90
A models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_phtadj.F90
A models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_nln_matrix.F90
A models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_lu_factor.F90
A models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_sim_dat.F90
A models/atm/cam/src/chemistry/pp_trop_strat_soa/mo_indprd.F90
A models/atm/cam/src/chemistry/pp_trop_strat_soa/m_spc_id.F90
A models/atm/cam/src/chemistry/pp_trop_strat_soa/chem_mech.doc
A models/atm/cam/src/chemistry/pp_trop_strat_soa/chem_mech.in
A models/atm/cam/src/chemistry/pp_trop_strat_soa
- chemical mechanism added to replace trop_strat_bam_v1
A models/atm/cam/bld/namelist_files/use_cases/2000_cam5_trop_moz_mam3.xml
A models/atm/cam/bld/namelist_files/use_cases/2000_cam4_trop_strat_soa.xml
A models/atm/cam/bld/namelist_files/use_cases/2000_cam5_trop_strat_mam3.xml
A models/atm/cam/bld/namelist_files/use_cases/2000_cam5_trop_strat_mam7.xml
- new build-namelist use case files
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/config_files/wm1.9c4h
- changed "waccm_mozart_v1" to "waccm_mozart"
M models/atm/cam/test/system/config_files/fst1.9c4dh
- changed "trop_strat_bam_v1" to "trop_strat_soa"
M models/atm/cam/test/system/config_files/fsoa4c4dm
- correction to -nosmp config option
M models/atm/cam/test/system/tests_chem_hybrid
- moved SOA tests to be first
M models/atm/cam/test/system/test_driver.sh
- set yellowstone queue to $CAM_BATCHQ
M models/atm/cam/test/system/input_tests_master
- adjustments for outfrq3s_mozEOOH namelist options
M models/atm/cam/bld/configure
- removed chemistry package waccm_mozart_v1
- replaced chemistry package trop_strat_bam_v1 with trop_strat_soa
- adjusted number of advected tracers in chemistry packages
trop_strat_mam3
trop_strat_mam7
trop_mozart_mam3
M models/atm/cam/bld/config_files/definition.xml
- removed chemistry package waccm_mozart_v1
- replaced chemistry package trop_strat_bam_v1 with trop_strat_soa
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- waccm_mozart_v1 defaults removed
- update to megan_factors_file
- update to default SAD file -- default for WACCMX also
- update to sulfate file
- update to photolysis cross sections
- set defaults of nspltvrm, nspltac and nsplit when waccm_phys is used
M models/atm/cam/bld/namelist_files/master_gas_wetdep_list.xml
M models/atm/cam/bld/namelist_files/master_drydep_list.xml
- updates to species lists
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- increase in megan_specifier string (to 1024 characters)
M models/atm/cam/bld/namelist_files/use_cases/sd_waccm_geos5.xml
M models/atm/cam/bld/namelist_files/use_cases/sd_waccm_tslt_geos5.xml
- update to solar proton input file
- removed bndtvghg -- this is redundant specification of lower boundary with flbc_file
M models/atm/cam/bld/namelist_files/use_cases/waccm_1955-2005_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp85.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1850-2005_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp26.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp45.xml
M models/atm/cam/bld/namelist_files/use_cases/1850-2005_cam4_super_fast_llnl.xml
- removed bndtvghg -- this is redundant specification of lower boundary with flbc_file
- update SPE file
M models/atm/cam/bld/namelist_files/use_cases/mozart_megan_emis.xml
M models/atm/cam/bld/namelist_files/use_cases/soa_chem_megan_emis.xml
- update to megan_factors_file
- corrected emissions from files
- set megan_mapped_emisfctrs to .false.
- included a fincl1 list
- include IC files
M models/atm/cam/bld/build-namelist
- removed waccm_mozart_v1 setting
- replaced "trop_strat_bam_v1" with "trop_strat_soa"
- set default nsplit and nspltrac for waccm_phys
- set default srf_emis_cycle_yr to 2000 for waccm_mozart_mam3 chemistry
M models/atm/cam/bld/perl5lib/Build/ChemNamelist.pm
- removed kludge to set wet deposition list for waccm_mozart_v1
M models/atm/cam/src/physics/cam/chem_surfvals.F90
- call endrun if both bndtvghg and flbc_file are specified
M models/atm/cam/src/chemistry/utils/mo_flbc.F90
- call endrun if lower boundary is specified in more than one manner
M models/atm/cam/src/chemistry/mozart/mo_neu_wetdep.F90
- map EOOH to CH2O
M models/atm/cam/src/physics/waccm/gw_drag.F90
M models/atm/cam/src/chemistry/utils/mo_solar_parms.F90
M models/atm/cam/src/chemistry/bulk_aero/progseasalts_intr.F90
M models/atm/cam/src/chemistry/mozart/mo_jlong.F90
M models/atm/cam/src/chemistry/mozart/upper_bc.F90
M models/atm/cam/src/chemistry/mozart/wei96.F90
M models/atm/cam/src/chemistry/mozart/mo_jshort.F90
M models/atm/cam/src/chemistry/mozart/mo_chemini.F90
M models/atm/cam/src/chemistry/mozart/mo_aurora.F90
M models/atm/cam/src/chemistry/mozart/mo_apex.F90
M models/atm/cam/src/chemistry/mozart/mo_fstrat.F90
M models/atm/cam/src/chemistry/mozart/mo_sad.F90
M models/atm/cam/src/chemistry/mozart/efield.F90
- restrict message logging to masterproc
M models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
- interface changes to mo_sulf and mo_drydep
- zero out topospheric sulfate above tropopause and
stratospheric sulfate below the tropopause
M models/atm/cam/src/chemistry/mozart/chemistry.F90
- interface changes to mo_sulf
M models/atm/cam/src/chemistry/mozart/mo_lightning.F90
M models/atm/cam/src/chemistry/mozart/mo_strato_sad.F90
- removed use mo_regrider statement -- mo_regrider not used in these modules
M models/atm/cam/src/chemistry/mozart/mo_sulf.F90
- replaced the use of mo_regrider with tracer_data utility module
M models/atm/cam/src/chemistry/mozart/mo_drydep.F90
- replaced the use of mo_regrider with lininterp_data module
which will regrid data on unstructure grids
- depostition velocity of EOOH mapped to CH2O
M models/atm/cam/src/chemistry/mozart/linoz_data.F90
- misc code clean up
M models/atm/cam/src/chemistry/mozart/mo_setsoa.F90
- correction to units of the SOA*_PROD diagnostics
- correction for changing soa* tag names
M models/atm/cam/src/chemistry/mozart/mo_usrrxt.F90
- corrections to surface area calculations of modal aerosols
M models/atm/cam/src/chemistry/pp_trop_bam/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_ghg/mo_sim_dat.F90
- preprocessor re-ran on these mechanisms to give updated data
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_rxt_rates_conv.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/m_spc_id.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/chem_mech.in
- mechanism updates
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_setrxt.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_rxt_rates_conv.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/m_spc_id.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/chem_mech.in
- mechanism updates
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_setrxt.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_rxt_rates_conv.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/m_spc_id.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/chem_mech.in
- mechanism updates
M models/atm/cam/src/chemistry/pp_waccm_ghg/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_sulfur/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_bam/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_ghg/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_imp_sol.F90
- correction for renaming the soa1 tag name
M SVN_EXTERNAL_DIRECTORIES
- chem proprocessor updates
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
033 bl336 TBL.sh wm4c4h outfrq3s+waccm_1850_cam4 9s .......................FAIL! rc= 7 at Wed Mar 27 19:14:26 MDT 2013
- expected failure due to default setting of nspltvrm
070 bl379 TBL.sh wm1.9c4h outfrq24h+waccm_1850_cam4 2d ....................FAIL! rc= 7 at Wed Mar 27 21:22:55 MDT 2013
- expected failure due to default setting of nspltvrm
- changed chemistry mechanism from waccm_mozart_v1 to waccm_mozart
048 bl373 TBL.sh fm1.9c4dh outfrq3s 9s ....................................FAIL! rc= 7 at Wed Mar 27 20:11:43 MDT 2013
052 bl376 TBL.sh fm1.9c4dh+testmech outfrq3s 9s ...........................FAIL! rc= 7 at Wed Mar 27 20:22:14 MDT 2013
055 bl374 TBL.sh fs1.9c4dh outfrq3s+1850-2005_cam4_super_fast_llnl 9s .....FAIL! rc= 7 at Wed Mar 27 20:29:51 MDT 2013
- changes in input files
057 bl430 TBL.sh fm1.9c5dh outfrq3s_mozEOOH 9s ............................FAIL! rc= 7 at Wed Mar 27 20:38:59 MDT 2013
- changes in chemistry mechanism and input files
060 bl440 TBL.sh fsoa1.9c4dh outfrq3s_mozEOOH+soa_chem_megan_emis 9s ......FAIL! rc= 7 at Wed Mar 27 20:55:25 MDT 2013
frankfurt/lf95:
030 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7 at Wed Mar 27 19:27:30 MDT 2013
- expected failure due to default setting of nspltvrm
frankfurt/pgi:
036 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7 at Wed Mar 27 15:31:43 MDT 2013
- expected failure due to default setting of nspltvrm
056 bl420 TBL.sh f7mode4c5dm outfrq3s 9s ..................................FAIL! rc= 7 at Wed Mar 27 16:38:04 MDT 2013
- expected failure due to changes in photolysis inputs
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_2_14
Originator(s): fischer, hannay
Date: Fri Mar 15 11:09:40 MDT 2013
One-line Summary: Add namelist variables to control history file output, generate
pre cam5_2_13 baselines, waccm sc ic file
Purpose of changes:
. Namelist variables history_amwg, history_aerosol, history_eddy, and history_budget
are setup to control the history file outputs. These are boolean flags that are
used to control the add_default calls. history_amwg is the only one that defaults to
true.
. A -b argument was added to test_driver.sh, incase there is a need to generate baselines
for pre cam5_2_13 tags. Usage would be test_driver.sh -f -b
. New better logic in test_driver.sh to handle command line arguments
. New ne30 initial condition file for waccm_sc_2000_cam4.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
M models/atm/cam/bld/namelist_files/use_cases/waccm_sc_2000_cam4.xml
. Changed ne30np4 ic from cami_ne30np4_L66_SC_WACCM_c121127.nc to
f.e11.FWSC.ne30_ne30.wm_SE.003.cam.i.0006-01-01-00000.nc
. Add history_amwg, history_aerosol, history_budget, and history_eddy to control
variables being saved to history files. The default is history_amwg is true,
the rest are false.
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: fischer, eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/archive_baseline.sh
. Change group write permission
M models/atm/cam/test/system/TCB.sh
M models/atm/cam/test/system/TBR.sh
M models/atm/cam/test/system/TER.sh
M models/atm/cam/test/system/TSM.sh
. Turn on extra history output for cam3/4/5
M models/atm/cam/test/system/TBL_ccsm.sh
M models/atm/cam/test/system/TBL.sh
. Add logic to be able to create pre cam5_2_13 baselines
M models/atm/cam/test/system/test_driver.sh
. Improve logic for handling command line arguments
. Add -b argument to be able to generate baselines for pre cam5_2_13 tags.
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. Remove history_microphysics variable
. Set defaults for history_amwg(true), history_aerosol(false),
history_eddy(false), and history_budget(false)
M models/atm/cam/bld/namelist_files/namelist_definition.xml
. Add history_amwg and history_eddy, removed history_microphyics
M models/atm/cam/bld/build-namelist
. Add history_amwg, history_aerosol, history_budget, and history_eddy to namelists
Remove history_microphysics
M models/atm/cam/src/physics/cam/cloud_diagnostics.F90
M models/atm/cam/src/physics/cam/microp_aero.F90
M models/atm/cam/src/physics/cam/tropopause.F90
M models/atm/cam/src/physics/cam/radiation.F90
M models/atm/cam/src/physics/cam/cloud_cover_diags.F90
M models/atm/cam/src/physics/cam/micro_mg_cam.F90
M models/atm/cam/src/physics/cam/aer_rad_props.F90
M models/atm/cam/src/physics/cam/phys_control.F90
M models/atm/cam/src/physics/cam/cam_diagnostics.F90
M models/atm/cam/src/physics/cam/stratiform.F90
M models/atm/cam/src/physics/cam/modal_aer_opt.F90
M models/atm/cam/src/physics/cam/ndrop.F90
M models/atm/cam/src/physics/cam/convect_shallow.F90
M models/atm/cam/src/physics/cam/macrop_driver.F90
M models/atm/cam/src/physics/cam/gw_drag.F90
M models/atm/cam/src/physics/cam/zm_conv_intr.F90
M models/atm/cam/src/physics/cam/vertical_diffusion.F90
M models/atm/cam/src/physics/rrtmg/radiation.F90
M models/atm/cam/src/chemistry/utils/modal_aero_wateruptake.F90
M models/atm/cam/src/chemistry/bulk_aero/progseasalts_intr.F90
M models/atm/cam/src/chemistry/bulk_aero/dust_intr.F90
M models/atm/cam/src/chemistry/mozart/mo_chm_diags.F90
M models/atm/cam/src/dynamics/fv/dyn_comp.F90
. Move add_default into conditionals based on history_amwg, history_aerosol,
history_budget, and history_eddy. Remove history_microphysics.
M models/atm/cam/bld/namelist_files/use_cases/waccm_sc_2000_cam4.xml
. Change ic file for ne30np4
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel: all passed
frankfurt/lf95: all passed
frankfurt/pgi: all passed
===============================================================
===============================================================
Tag name: cam5_2_13
Originator(s): fischer, eaton, cacraig, cecila, jedwards, Mark Taylor
Date: 3-11-2013
One-line Summary: test script improvements, homme update, external updates, bug fixes,
change solar namelist defaults
Purpose of changes:
. The default for the solar data namelist variables was being set
differently for the configurations "-phys cam5 -chem trop_mam3" and
"-phys cam5 -chem none". This was changed to make the value for
prescribed aerosols be the same as for prognostic aerosols. Also added
a check to build-namelist to disallow setting both solar_const and
solar_data_file. It only makes sense to set one of these variables.
. Made changes to test scripts to improve testing on yellowstone.
- -j option was added to be able to specify number of of jobs for gmake.
The default setting of -j was unlimited, which was producing slower builds
than specifying -j 4, -j 8, or -j 16.
- Fix issues with builds on yellowstone being repeated done when it
fails to build the first time.
- Fix issues with builds being tried on yellowstone compute nodes.
- Add CAM_RBOPTIONS to give more flexibility on how builds are done.
Valued values are "build_only", "run_only" and "run_and_build"
CAM_RBOPTIONS is passed to the other test scripts through the argument
list. The effect of this change is this tag cannot make a baseline for
previous tags.
- Option for pgi testing was added, but not working yet.
. SVN externals were updated to match cesm1_2_beta02.
- Issues with PIO forced the use of a later PIO tag.
- Also, the Machines tag wasn't updated do to failed tests with intel/13.0.1
. CAM-SE (homme) external was update to homme1_3_17
Bugs fixed (include bugzilla ID):
. Fixed long name for SNOWHICE. Bugzilla #945
. Fix memory leak by adding missing deallocate. Bugzilla #1629
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: fischer, fvitt
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/tests_pretag_yellowstone
M models/atm/cam/test/system/tests_posttag_yellowstone
. Move sm415 and bl415(fst1.9c5dh) from pretag to posttag test.
. Replaced sm993(ne30_g16), er993, and bl993 with sm997(ne16_g37),er997, and bl997
M models/atm/cam/test/system/input_tests_master
. Add an ne16_g37 cesm test sm997, er997, and bl997
D models/atm/cam/test/system/TBL_cb.sh
D models/atm/cam/test/system/TNE_cb_ccsm.sh
D models/atm/cam/test/system/TBL_cb_ccsm.sh
D models/atm/cam/test/system/TSC_cb.sh
. These test scripts are no longer needed.
M models/atm/cam/test/system/test_driver.sh
. Changed pes layouts for yellowstone and frankfurt
. Removed bluefire logic
. Add -j command line option to set the number of jobs for gmake
. Changed logic for yellowstone builds, added new enviroment variable
CAM_RBOPTIONS, valide values are 'build_only', 'run_only', and 'run and build'
. Started adding pgi support for yellowstone, currently not working.
M models/atm/cam/test/system/TCB.sh
M models/atm/cam/test/system/TDD.sh
M models/atm/cam/test/system/TCB_ccsm.sh
M models/atm/cam/test/system/TBL.sh
M models/atm/cam/test/system/TBR.sh
M models/atm/cam/test/system/TPF.sh
M models/atm/cam/test/system/TER.sh
M models/atm/cam/test/system/TEQ_ccsm.sh
M models/atm/cam/test/system/TER_ccsm.sh
M models/atm/cam/test/system/TSM_ccsm.sh
M models/atm/cam/test/system/TMC.sh
M models/atm/cam/test/system/TBL_ccsm.sh
M models/atm/cam/test/system/TSC.sh
M models/atm/cam/test/system/TEQ.sh
M models/atm/cam/test/system/TNE_ccsm.sh
M models/atm/cam/test/system/TSM.sh
. Pass CAM_RBOPTIONS to test scripts through argument list
. Add logic to handle CAM_RBTIONS
M models/atm/cam/bld/configure
. remove a debug print
. Updated build filepaths for changes in the clm directory structure to
handle multiple versions of clm (clm4.0, clm4.5)
M models/atm/cam/bld/namelist_files/use_cases/cam4_trop_strat_chem.xml
. remove extraneous setting of solar_const
M models/atm/cam/bld/build-namelist
. add check to disallow setting both solar_const and solar_data_file
. change how solar namelist defaults are set so that for cam5 physics the
defaults don't depend on which chemistry package is used. The defaults
for cam4 physics continue to depend on whether chemistry is used or not.
M models/atm/cam/bld/perl5lib/Build/Namelist.pm
. add delete_variable method
M models/atm/cam/src/physics/cam/cam_diagnostics.F90
. Fixed long name for SNOWHICE. Bugzilla #945
M models/atm/cam/src/utils/cam_pio_utils.F90
. Fix memory leak by adding missing deallocate. Bugzilla #1629
M models/atm/cam/src/chemistry/mozart/short_lived_species.F90
. Bug fix to pbuf_set_field
M models/atm/cam/src/dynamics/homme/dyn_comp.F90
. Bug fix for dynamics timestep when using lagrangian code.
M models/atm/cam/src/dynamics/fv/te_map.F90
. Abort if vertical levels cross. This prevents some unstable runs in
various FV CAM and WACCM simulations.
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
. Update homme(SE) external
M SVN_EXTERNAL_DIRECTORIES
. Update svn external to match cesm1_2_beta02. Issues with pio forced the
use of a later version of pio.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
. All passed
frankfurt/lf95:
035 bl318 TBL.sh f10c4cdm sat_hist 9s .....................................FAIL! rc= 7 at Thu Mar 7 17:07:33 MST 2013
. Changed ntasks from 8 to 16, causes answer changes in COSP due to known bugs in COSP code
frankfurt/pgi:
044 bl317 TBL.sh f10c5cdm outfrq3s 9s .....................................FAIL! rc= 7 at Thu Mar 7 12:39:38 MST 2013
. Changed ntasks from 8 to 16, causes answer changes in COSP due to known bugs in COSP code
047 bl319 TBL.sh fn10c5dm outfrq3s_bam 9s .................................FAIL! rc= 7 at Thu Mar 7 12:46:18 MST 2013
050 bl320 TBL.sh fn10c5dm rad_diag 9s .....................................FAIL! rc= 7 at Thu Mar 7 12:49:13 MST 2013
. The change to solar input defaults change answers for configurations with
the args "-phys cam5 -chem none".
CAM tag used for the baseline comparison tests if different than previous
tag:
===============================================================
===============================================================
Tag name: cam5_2_12
Originator(s): fvitt
Date: 01 Mar 2012
One-line Summary: Corrections to super_fast_llnl_mam3 chemistry package
and misc updates and fixes
Purpose of changes:
- correct super_fast_llnl_mam3 chemistry package
- fix in modal_aero_data to avoid index out of range error
- change history output of SOA* aerosols from VMR to MMR
- add build-namelist use case for MOZSOA compset
so that 'NEU' wet deposition scheme is used
- update solar input files for SD-WACCM
- update default MEGAN emissions factors input file
- make chemistry preprocessor more robust
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
A models/atm/cam/bld/namelist_files/use_cases/soa_chem_megan_emis.xml
- new use case file for MOZSOA compsets -- this uses 'NEU' wet dep method
A models/atm/cam/test/system/config_files/fsoa4c4dm
- new SOA chem regestion test
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/tests_chem_hybrid
M models/atm/cam/test/system/input_tests_master
M models/atm/cam/test/system/tests_chem_mpi
- adjustments to chem tests
M models/atm/cam/bld/configure
- minor cleanup associated with customizing chemistry (preprocessor)
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- corrections to prescribed ozone input for SC-WACCM
- added default IC file for waccm5 at 4x5 resolution
- update to default MEGAN emissions factors
M models/atm/cam/bld/build-namelist
- correction in super_fast_llnl_mam3 linoz and chlorine loading inputs
M ... /use_cases/.. sd_waccm_geos5 ???
- updates to solar_data_file and solar_parms_file
M models/atm/cam/bld/perl5lib/Build/ChemPreprocess.pm
- changes to make preprocessor more robust
. check if need to re-run preprocessor -- checks changes to input
. attempt to store executable in a public place so that every user
does not need to compile every time the preprocessor is needed
. errors in preprocessor input more apparent
M models/atm/cam/src/chemistry/mozart/mo_mass_xforms.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_data.F90
- avoid index out of range error when index is greater the number of
registered constituents which can occur when we have chemical tracers
that are not registered (e.g., H2O in WACCM)
M models/atm/cam/src/chemistry/pp_super_fast_llnl/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/chem_mech.doc
M models/atm/cam/src/chemistry/pp_super_fast_llnl/chem_mech.in
- change in tag name of O3 + hv photo reaction to be compatable with
tag name check in llnl_O1D_to_2OH_adj
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_setrxt.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_rxt_rates_conv.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/chem_mods.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_phtadj.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/m_spc_id.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/chem_mech.doc
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/chem_mech.in
- LLNL corrections to super_fast_llnl_mam3 chem mechanism
M models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
M models/atm/cam/src/chemistry/mozart/mo_setz.F90
M models/atm/cam/src/chemistry/mozart/mo_pchem.F90
M models/atm/cam/src/chemistry/mozart/llnl_O1D_to_2OH_adj.F90
M models/atm/cam/src/chemistry/mozart/chemistry.F90
M models/atm/cam/src/chemistry/mozart/mo_xsections.F90
- changes for super_fast_llnl_mam3 -- checks for j2oh rxt tag
M models/atm/cam/src/chemistry/mozart/mo_chm_diags.F90
- output SOA aerosol species in units of kg/kg
M models/atm/cam/src/chemistry/mozart/mo_setsoa.F90
- added diagnostics
M models/atm/cam/src/chemistry/mozart/mo_neu_wetdep.F90
- map SOG species to H2O2
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
062 bl440 TBL.sh fsoa1.9c4dh outfrq3s+soa_chem_megan_emis 9s ..............FAIL! rc= 7 at Thu Feb 28 14:41:12 MST 2013
- failed due to change in build-namelist use case
frankfurt/lf95:
030 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7 at Thu Feb 28 14:58:28 MST 2013
- failed due to change in prescribed ozone input file
frankfurt/pgi:
036 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7 at Thu Feb 28 10:52:11 MST 2013
- failed due to change in prescribed ozone input file
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_2_11
Originator(s): santos
Date: Feb 25, 2013
One-line Summary: Separate MAM top level, code simplifications
Purpose of changes:
- Separate MAM top level from tropospheric cloud physics
top level, so that modal aerosols can sediment above the
clouds in WACCM5.
- Make "tend" an optional argument to physics_update.
- Make interpic_new always produce netCDF with 64-bit offset.
- Cleanup regarding gw_drag.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
- Add "clim_modal_aero_top_press", the new top level for
modal aerosol microphysics. By default this is high enough
that modal aerosol calculations are done throughout the
WACCM levels, but below most WACCM-X levels.
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
M models/atm/cam/bld/namelist_files/namelist_definition.xml
M models/atm/cam/bld/build-namelist
- Add clim_modal_aero_top_lev option with default 1.e-4 Pa.
M models/atm/cam/src/chemistry/utils/modal_aero_wateruptake.F90
M models/atm/cam/src/chemistry/utils/modal_aero_calcsize.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_coag.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_gasaerexch.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_newnuc.F90
M models/atm/cam/src/chemistry/mozart/mz_aerosols_intr.F90
M models/atm/cam/src/chemistry/mozart/mo_usrrxt.F90
M models/atm/cam/src/physics/cam/aer_rad_props.F90
M models/atm/cam/src/physics/cam/modal_aer_opt.F90
- Change top level from trop_cloud_top_lev to
clim_modal_aero_top_lev.
M models/atm/cam/src/physics/cam/physics_types.F90
- Move the physics_tend argument to the end of
physics_update's argument list, and make it
optional.
M models/atm/cam/src/physics/cam/micro_mg_cam.F90
M models/atm/cam/src/physics/cam/stratiform.F90
M models/atm/cam/src/physics/cam/clubb_intr.F90
M models/atm/cam/src/physics/cam/convect_shallow.F90
M models/atm/cam/src/physics/cam/macrop_driver.F90
M models/atm/cam/src/physics/cam/zm_conv_intr.F90
- Remove dummy physics_tend argument that is no
longer necessary to call physics_update.
M models/atm/cam/src/physics/cam/physpkg.F90
M models/atm/cam/src/physics/cam/tphysidl.F90
- Since both versions of gw_drag have a register
routine now, move gw_drag_register out of
#ifdef WACCM_PHYS.
- Move "tend" argument to the end of every
physics_update call.
M models/atm/cam/src/physics/cam/gw_drag.F90
- Add empty subroutine gw_drag_register, so that
the interface for this module matches the other
gw_drag.
M models/atm/cam/src/physics/cam/ref_pres.F90
- Add clim_modal_aero_top_lev, analogous to
trop_cloud_top_lev, and force the MAM top to
be above the cloud top.
M models/atm/cam/src/physics/cam/microp_aero.F90
M models/atm/cam/src/physics/cam/ndrop.F90
M models/atm/cam/src/physics/waccm/gw_drag.F90
- Changes with no effect (e.g. white space).
M models/atm/cam/tools/interpic_new/fmain.F90
- Remove -64 option, as 64-bit offset is now
always on.
WARNING: This will break scripts that specify
"-64" if they are used on interpic built from
this tag (or later).
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
WACCM5 posttag baseline will fail. WACCM5 is bit-for-bit if
clim_modal_aero_top_press is set to 100 Pa in the new code,
but the new default is 1.e-4 Pa. Part of ongoing development
of WACCM5.
yellowstone/intel:
All PASS.
frankfurt/lf95:
All PASS.
frankfurt/pgi or jaguar/pgi:
All PASS.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- WACCM5
- what platforms/compilers:
- All
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
- Affects climate, but can be made bit-for-bit through namelist.
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_2_10
Originator(s): santos
Date: Feb 8, 2013
One-line Summary: saturation vapor pressure fixes, WACCM5 fixes
Purpose of changes:
- Fix problems with saturation vapor pressure schemes
discovered during development of cam5_2_02.
- Fix outstanding problems with WACCM5 that cause it to
crash, by enforcing trop_cloud_top_lev in more modules
that require this limit.
- Make treatment of WACCM4 and WACCM5 more consistent. Both
now set the same top_lev limiter, but some modules perform
additional checks for CAM5 features to decide if they want
to use it.
- Add "rsplit" namelist option for HOMME (courtesy Cecile
Hannay).
Bugs fixed (include bugzilla ID):
"Meta"bugs:
1588: Incorrect calculation of saturation specific humidity at
low pressures throughout CAM
Bugs:
1583: zm_conv generates NaN and negative values in saturation mmr
calculation at low pressure
1584: radae bypasses checks for low pressure when calculating
saturation specific humidity
1585: modal_aero_wateruptake bypasses checks for low pressure in
saturation specific humidity
1589: Findsp does not conserve enthalpy for very low temperatures
Enhancements:
1587: Unify dqsdt calculations in CAM
Describe any changes made to build system:
Describe any changes made to the namelist:
- Add rsplit option for HOMME.
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/build-namelist
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- Add rsplit namelist option for HOMME.
M models/atm/cam/src/chemistry/bulk_aero/aerosol_intr.F90
- Set all fields added for clim_modal_aero to 0, to try to
reduce the chance of uninitialized data use.
M models/atm/cam/src/chemistry/modal_aero/modal_aero_gasaerexch.F90
M models/atm/cam/src/chemistry/mozart/mz_aerosols_intr.F90
M models/atm/cam/src/chemistry/utils/modal_aero_calcsize.F90
M models/atm/cam/src/chemistry/utils/modal_aero_wateruptake.F90
M models/atm/cam/src/physics/cam/aer_rad_props.F90
M models/atm/cam/src/physics/cam/cldwat2m_macro.F90
M models/atm/cam/src/physics/cam/cloud_fraction.F90
M models/atm/cam/src/physics/cam/macrop_driver.F90
M models/atm/cam/src/physics/cam/microp_driver.F90
M models/atm/cam/src/physics/cam/microp_aero.F90
- Add top_lev to more loops and array slices.
- Remove unnecessary/unused logic.
- The following miscellaneous changes:
- modal_aero_wateruptake: Change qsat_water call to use
new version.
- aer_rad_props: Since trop_cloud_top_lev is now set in WACCM4,
use it only "if (prog_modal_aero)".
- cldwat2m_macro: Rename "qsat" to "qs" to avoid confusion with
wv_saturation intrinsic (and "esat" to "es" to match).
- cloud_fraction: cldfrc_fice moved from cldwat, and top_lev
only used if macrop_scheme /= "rk".
- Microphysics/macrophysics modules: Change indexes to ensure
that output variables do not contain uninitialized data above
top_lev.
M models/atm/cam/src/chemistry/utils/tracer_data.F90
M models/atm/cam/src/physics/cam/uwshcu.F90
- Comment changes only.
M models/atm/cam/src/physics/cam/cldwat.F90
- Replace k1mb with top_lev.
- Update calculation of dqsdt to use new code (and remove check for
low pressure that is now done in wv_sat_methods).
- Because cldwat_fice is shared code between CAM4 and CAM5, it
has been moved to cloud_fraction. The resulting code is only
used by RK microphysics and thus no longer checks microp_scheme.
M models/atm/cam/src/physics/cam/micro_mg1_0.F90
M models/atm/cam/src/physics/cam/micro_mg_cam.F90
M models/atm/cam/src/physics/cam/radae.F90
M models/atm/cam/src/physics/cam/radlw.F90
M models/atm/cam/src/physics/cam/zm_conv.F90
- Update code that did SVP calculations without proper limits for
low pressure, and/or used a method other than the CAM default.
- ntopcld moved, and is now max(trop_cloud_top_lev,2).
- SVP conversion from Pa to hPa has been moved back into zm_conv,
since the requirement is specific to this module and not used
by any other CAM module.
M models/atm/cam/src/physics/cam/physpkg.F90
M models/atm/cam/src/physics/cam/stratiform.F90
- Moved "inimc" call from physpkg to stratiform, since cldwat
is not used at all unless RK microphysics is on.
- Slight simplifications to stratiform constituent logic.
M models/atm/cam/src/physics/cam/wv_sat_methods.F90
- Do proper pressure limiting in qsat/qmmr functions.
- Changed wv_sat_svp_trans to work correctly for ttrice == 0._r8.
- Bolton water scheme now outputs in Pa, like the other schemes.
M models/atm/cam/src/physics/cam/wv_saturation.F90
- Remove "legacy" functions and special function for ZM.
- Remove multiple copies of tmin; SVP calculations are now
always performed down to 127.16K.
- Fix findsp enthalpy checking at low temperatures, and generally
improve consistency/integration with wv_saturation. Enthalpy is
now always checked to ensure conservation.
- Note: Findsp used to ENDRUN with a non-convergence error for
very low temperatures. This (sometimes beneficial) checking
behavior was unintentional, and only triggered in certain edge
cases. Now findsp will abort the run only if a problem occurs
while it is running, and not due to bad input. If necessary,
temperatures should be checked elsewhere.
- Cleanup changes:
- trinv removed.
- estblf reorganized, and uses kind=r8 in aint.
- Pull out enthalpy formula for findsp into a function
(tq_enthalpy).
- Enthalpy at saturation can now be calculated as an optional
argument to the qsat routines.
- Temperature derivative optional outputs from qsat_* routines
are calculated in a separate (shared) function.
M models/atm/cam/src/physics/waccm/gw_drag.F90
- Prevent divide by 0 if wind happens to be exactly 0 at certain
points.
- Prevent segfault from certain bad winds by avoiding use of "abs"
to limit integer indices. This is just to avoid a "red herring"
crash in gw_drag, when the real problem is elsewhere.
D models/atm/cam/test/system/config_files/wm1.9c5h
A models/atm/cam/test/system/config_files/wm1.9c5dh
M models/atm/cam/test/system/input_tests_master
- WACCM5 test is now a debug test, to catch some common errors
that occurred during development.
- Change "outfrq3s" to "outfrq3s_clubb" in CLUBB baseline test.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
004 bl132 TBL.sh e48c4dh ghgrmp+1850_cam4 9s ..............................FAIL! rc= 7 at Fri Feb 8 15:44:21 MST 2013
006 bl133 TBL.sh e48c4paqdm aqpgro+aquaplanet_cam4 3s .....................FAIL! rc= 7 at Fri Feb 8 15:44:22 MST 2013
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Fri Feb 8 15:44:27 MST 2013
013 bl330 TBL.sh f4c5paqdh aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Fri Feb 8 15:44:28 MST 2013
017 bl331 TBL.sh f4c4dh co2rmp+1850_cam4 9s ...............................FAIL! rc= 7 at Fri Feb 8 15:44:33 MST 2013
021 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...............................FAIL! rc= 7 at Fri Feb 8 15:44:39 MST 2013
023 bl333 TBL.sh f4c4paqdh aqpgro+aquaplanet_cam4 3s ......................FAIL! rc= 7 at Fri Feb 8 15:44:40 MST 2013
033 bl336 TBL.sh wm4c4h outfrq3s+waccm_1850_cam4 9s .......................FAIL! rc= 7 at Fri Feb 8 15:45:08 MST 2013
036 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ......................................FAIL! rc= 7 at Fri Feb 8 15:45:12 MST 2013
040 bl366 TBL.sh f1.9c5clbdm outfrq3s_clubb 9s ............................FAIL! rc= 7 at Fri Feb 8 15:45:12 MST 2013
043 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .....................................FAIL! rc= 7 at Fri Feb 8 15:45:25 MST 2013
048 bl373 TBL.sh fm1.9c4dh outfrq3s 9s ....................................FAIL! rc= 7 at Fri Feb 8 15:45:47 MST 2013
050 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...........................FAIL! rc= 7 at Fri Feb 8 15:45:56 MST 2013
052 bl376 TBL.sh fm1.9c4dh+testmech outfrq3s 9s ...........................FAIL! rc= 7 at Fri Feb 8 15:46:19 MST 2013
055 bl374 TBL.sh fs1.9c4dh outfrq3s+1850-2005_cam4_super_fast_llnl 9s .....FAIL! rc= 7 at Fri Feb 8 15:46:31 MST 2013
057 bl415 TBL.sh fst1.9c5dh outfrq3s+cam5_trop_strat_chem 9s ..............FAIL! rc= 7 at Fri Feb 8 15:47:08 MST 2013
059 bl430 TBL.sh fm1.9c5dh outfrq3s 9s ....................................FAIL! rc= 7 at Fri Feb 8 15:47:35 MST 2013
062 bl440 TBL.sh fsoa1.9c4dh outfrq3s+mozart_megan_emis 9s ................FAIL! rc= 7 at Fri Feb 8 15:48:00 MST 2013
066 bl387 TBL.sh f1.9c4dm outfrq3s+1850-2005_cam4 9s ......................FAIL! rc= 7 at Fri Feb 8 15:48:07 MST 2013
069 bl388 TBL.sh f1.9c4m outfrq24h+1850_cam4 2d ...........................FAIL! rc= 7 at Fri Feb 8 15:48:12 MST 2013
072 bl379 TBL.sh wm1.9c4h outfrq24h+waccm_1850_cam4 2d ....................FAIL! rc= 7 at Fri Feb 8 15:49:16 MST 2013
075 bl731 TBL.sh h16c3aqdh aqua+aquaplanet_cam3 9s ........................FAIL! rc= 7 at Fri Feb 8 15:49:21 MST 2013
078 bl751 TBL.sh h30c4aqdm outfrq3s+aquaplanet_cam4 9s ....................FAIL! rc= 7 at Fri Feb 8 15:49:37 MST 2013
084 bl992 TBL_ccsm.sh f19_g16 ETEST 2d ....................................FAIL! rc= 7 at Fri Feb 8 15:49:41 MST 2013
087 bl993 TBL_ccsm.sh ne30_g16 FC5 2d .....................................FAIL! rc= 7 at Fri Feb 8 15:50:11 MST 2013
frankfurt/lf95:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Fri Feb 8 14:44:16 MST 2013
009 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Fri Feb 8 14:44:19 MST 2013
013 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Fri Feb 8 14:44:22 MST 2013
017 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Fri Feb 8 14:44:31 MST 2013
025 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Fri Feb 8 14:45:18 MST 2013
027 bl313 TBL.sh f10c3aqdm outfrq3s+aquaplanet_cam3 3s ....................FAIL! rc= 7 at Fri Feb 8 14:45:19 MST 2013
030 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7 at Fri Feb 8 14:45:27 MST 2013
032 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 7 at Fri Feb 8 14:45:28 MST 2013
035 bl318 TBL.sh f10c4cdm sat_hist 9s .....................................FAIL! rc= 7 at Fri Feb 8 14:45:32 MST 2013
frankfurt/pgi or jaguar/pgi:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Fri Feb 8 14:46:31 MST 2013
010 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Fri Feb 8 14:46:35 MST 2013
012 bl113 TBL.sh e8c5paqdm aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Fri Feb 8 14:46:35 MST 2013
016 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Fri Feb 8 14:46:38 MST 2013
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Fri Feb 8 14:46:44 MST 2013
031 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Fri Feb 8 14:47:30 MST 2013
033 bl313 TBL.sh f10c3aqdm outfrq3s+aquaplanet_cam3 3s ....................FAIL! rc= 7 at Fri Feb 8 14:47:31 MST 2013
036 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7 at Fri Feb 8 14:47:41 MST 2013
039 bl315 TBL.sh f10c3dm outfrq3s+cam3 9s .................................FAIL! rc= 7 at Fri Feb 8 14:47:43 MST 2013
041 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 7 at Fri Feb 8 14:47:43 MST 2013
044 bl317 TBL.sh f10c5cdm outfrq3s 9s .....................................FAIL! rc= 7 at Fri Feb 8 14:47:51 MST 2013
047 bl319 TBL.sh fn10c5dm outfrq3s_bam 9s .................................FAIL! rc= 7 at Fri Feb 8 14:47:56 MST 2013
050 bl320 TBL.sh fn10c5dm rad_diag 9s .....................................FAIL! rc= 7 at Fri Feb 8 14:48:00 MST 2013
053 bl321 TBL.sh f10c5cdm atrain 9s .......................................FAIL! rc= 7 at Fri Feb 8 14:48:05 MST 2013
056 bl420 TBL.sh f7mode4c5dm outfrq3s 9s ..................................FAIL! rc= 7 at Fri Feb 8 14:48:31 MST 2013
059 bl734 TBL.sh hn16c5aqdm outfrq3s_bam+aquaplanet_cam5 9s ...............FAIL! rc= 7 at Fri Feb 8 14:49:20 MST 2013
062 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ....................FAIL! rc= 7 at Fri Feb 8 14:50:25 MST 2013
**All CAM physics cases fail baseline tests due to wv_saturation bug fixes.**
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- All CAM physics cases (everything but ideal and adiabatic).
- what platforms/compilers:
- All.
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
- Above roundoff but same climate as previous version and CESM1.1.1
This was demonstrated with:
- A CAM5 present day case (FC5, 1 degree), with code based on
cam5_2_03, run for 5 years.
- A fully-coupled 1850 WACCM case (B1850WCN, 2 degree), with code
based on cam5_2_06, run for 25 years.
- This run is in the experiment DB with ID #1635.
- The case name is: b.e11.B1850WCN.f19_g16.test_lim.001
- A fully-coupled 1850 CAM5 case (B1850C5CN, 2 degree) is also
in the run database, but has not had its climate compared with
CESM 1.1.1.
- ID #1634
- Case name is b.e11.B1850C5CN.f19_g16.test_lim.001
- Finite-volume cases are bit-for-bit between cam5_2_06 and
cam5_2_09, except that there is a roundoff-level difference due
to changes to shr_reprosum_mod in csm_share.
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_2_09
Originator(s): Cheryl Craig
Date: 1/28/13
One-line Summary: Dynamic state,tend,ptend
Purpose of changes:
- Introduce dynamic state,tend, ptend structures and supporting routines for use with sub-columns
- eliminated MODAL_AERO ifdef and replaced with rad_constituent interface inside ndrop.F90 (Eaton modification)
- Removed CLUBB_core from repository and the directory models/atm/cam/src/physics/clubb is now an external library
- Added Yellowstone pretag test for CLUBB
- Add module load mkl for CLUBB library
- renamed sub_column to microp_uniform to better reflect its purpose inside microp routines
Bugs fixed (include bugzilla ID):
- Corrected flag for the intel compiler to ignore temp array creation
Describe any changes made to build system:
- Introduced new configure parameter "psubcols". Default is 1 = no subcolumns.
Describe any changes made to the namelist:
- Introduced use_subcol_microp - Switch which will be used to control sub-columns within microphysics
Namelist flag is accessed in phys_control, but it is not used inside CAM yet
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
D models/atm/cam/src/physics/CLUBB_core
D models/atm/cam/src/physics/CLUBB_core/constants_clubb.F90
D models/atm/cam/src/physics/CLUBB_core/variables_prognostic_module.F90
D models/atm/cam/src/physics/CLUBB_core/interpolation.F90
D models/atm/cam/src/physics/CLUBB_core/clip_explicit.F90
D models/atm/cam/src/physics/CLUBB_core/saturation.F90
D models/atm/cam/src/physics/CLUBB_core/mono_flux_limiter.F90
D models/atm/cam/src/physics/CLUBB_core/mixing_length.F90
D models/atm/cam/src/physics/CLUBB_core/stats_zm.F90
D models/atm/cam/src/physics/CLUBB_core/file_functions.F90
D models/atm/cam/src/physics/CLUBB_core/mean_adv.F90
D models/atm/cam/src/physics/CLUBB_core/grid_class.F90
D models/atm/cam/src/physics/CLUBB_core/pdf_closure_module.F90
D models/atm/cam/src/physics/CLUBB_core/stats_zt.F90
D models/atm/cam/src/physics/CLUBB_core/parameter_indices.F90
D models/atm/cam/src/physics/CLUBB_core/hyper_diffusion_4th_ord.F90
D models/atm/cam/src/physics/CLUBB_core/csr_matrix_class_3array.F90
D models/atm/cam/src/physics/CLUBB_core/input_reader.F90
D models/atm/cam/src/physics/CLUBB_core/parameters_model.F90
D models/atm/cam/src/physics/CLUBB_core/error_code.F90
D models/atm/cam/src/physics/CLUBB_core/lapack_wrap.F90
D models/atm/cam/src/physics/CLUBB_core/stats_rad_zm.F90
D models/atm/cam/src/physics/CLUBB_core/parameters_radiation.F90
D models/atm/cam/src/physics/CLUBB_core/mt95.f90
D models/atm/cam/src/physics/CLUBB_core/endian.F90
D models/atm/cam/src/physics/CLUBB_core/recl.inc
D models/atm/cam/src/physics/CLUBB_core/parameters_microphys.F90
D models/atm/cam/src/physics/CLUBB_core/gmres_cache.F90
D models/atm/cam/src/physics/CLUBB_core/stats_rad_zt.F90
D models/atm/cam/src/physics/CLUBB_core/sigma_sqd_w_module.F90
D models/atm/cam/src/physics/CLUBB_core/hydrostatic_module.F90
D models/atm/cam/src/physics/CLUBB_core/array_index.F90
D models/atm/cam/src/physics/CLUBB_core/anl_erf.F90
D models/atm/cam/src/physics/CLUBB_core/T_in_K_module.F90
D models/atm/cam/src/physics/CLUBB_core/calendar.F90
D models/atm/cam/src/physics/CLUBB_core/model_flags.F90
D models/atm/cam/src/physics/CLUBB_core/advance_windm_edsclrm_module.F90
D models/atm/cam/src/physics/CLUBB_core/pos_definite_module.F90
D models/atm/cam/src/physics/CLUBB_core/pdf_parameter_module.F90
D models/atm/cam/src/physics/CLUBB_core/fill_holes.F90
D models/atm/cam/src/physics/CLUBB_core/sponge_layer_damping.F90
D models/atm/cam/src/physics/CLUBB_core/stats_variables.F90
D models/atm/cam/src/physics/CLUBB_core/stats_subs.F90
D models/atm/cam/src/physics/CLUBB_core/variables_radiation_module.F90
D models/atm/cam/src/physics/CLUBB_core/numerical_check.F90
D models/atm/cam/src/physics/CLUBB_core/stat_file_module.F90
D models/atm/cam/src/physics/CLUBB_core/input_names.F90
D models/atm/cam/src/physics/CLUBB_core/output_netcdf.F90
D models/atm/cam/src/physics/CLUBB_core/diffusion.F90
D models/atm/cam/src/physics/CLUBB_core/advance_wp2_wp3_module.F90
D models/atm/cam/src/physics/CLUBB_core/advance_xm_wpxp_module.F90
D models/atm/cam/src/physics/CLUBB_core/extrapolation.F90
D models/atm/cam/src/physics/CLUBB_core/stats_sfc.F90
D models/atm/cam/src/physics/CLUBB_core/surface_varnce_module.F90
D models/atm/cam/src/physics/CLUBB_core/clip_semi_implicit.F90
D models/atm/cam/src/physics/CLUBB_core/clubb_core.F90
D models/atm/cam/src/physics/CLUBB_core/stats_type.F90
D models/atm/cam/src/physics/CLUBB_core/clubb_precision.F90
D models/atm/cam/src/physics/CLUBB_core/advance_helper_module.F90
D models/atm/cam/src/physics/CLUBB_core/gmres_wrap.F90
D models/atm/cam/src/physics/CLUBB_core/Skw_module.F90
D models/atm/cam/src/physics/CLUBB_core/parameters_tunable.F90
D models/atm/cam/src/physics/CLUBB_core/advance_xp2_xpyp_module.F90
D models/atm/cam/src/physics/CLUBB_core/output_grads.F90
D models/atm/cam/src/physics/CLUBB_core/variables_diagnostic_module.F90
- Removed CLUBB_core from repository and the directory
models/atm/cam/src/physics/clubb is now an external library
List all subroutines added and what they do:
A + models/atm/cam/test/system/config_files/f1.9c5clbdm
A + models/atm/cam/test/system/nl_files/outfrq3s_clubb
- Added Yellowstone pretag test for CLUBB
A models/atm/cam/bld/run-yellowstone.csh
- Example build and run script for yellowstone
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/test_driver.sh
- Add module load mkl for CLUBB library
M models/atm/cam/test/system/tests_pretag_yellowstone
M models/atm/cam/test/system/input_tests_master
- Added Yellowstone pretag test for CLUBB
MM models/atm/cam/bld/configure
MM models/atm/cam/bld/namelist_files/namelist_definition.xml
M models/atm/cam/src/physics/cam/ppgrid.F90
- Added psubcols. Changed CLUBB_core directory to clubb.
MM models/atm/cam/bld/Makefile.in
- Corrected flag for the intel compiler to ignore temp array creation
MM models/atm/cam/bld/config_files/definition.xml
MM models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
MM models/atm/cam/bld/build-namelist
M models/atm/cam/src/physics/cam/phys_control.F90
- Added use_subcol_microp switch to control sub-columns in microphysics
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
- CLUBB code is now in an external library
M models/atm/cam/src/physics/cam/microp_aero.F90
- last microphysics fields which were in calling lists were added to pbuf
MM models/atm/cam/src/physics/cam/micro_mg_cam.F90
- "sub_column" renamed "microp_uniform" to better describe variable
- modifications using new dynamically dimensioned state/tend/ptend
M models/atm/cam/src/physics/cam/physics_types.F90
- modifications for dynamically dimensioned state/tend/ptend
- added new routines physics_state_alloc, physics_tend_alloc, physics_ptend_alloc,
physics_state_dealloc, physics_tend_dealloc, and physics_ptend_dealloc
- state now has:
ngrdcol -- Grid -- number of active columns (on the grid)
nsubcol(pcols)-- Sub-columns -- number of active sub-columns in each grid column
psetcols -- -- max number of columns set - if subcols = pcols*psubcols, else = pcols
ncol -- -- sum of nsubcol for all ngrdcols - number of active columns
indcol(pcols*psubcols) -- indices for mapping from subcols to grid cols
- ptend and tend now has:
psetcols -- -- max number of columns set - if subcols = pcols*psubcols, else = pcols
- new routine physics_ptend_copy_subcol - copies ptend on grid to ptend on subcols
M models/atm/cam/src/physics/cam/ndrop.F90
- eliminated MODAL_AERO ifdef and replaced with rad_constituent interface (Eaton modification)
M models/atm/cam/src/physics/cam/clubb_intr.F90
- modifications using new dynamically dimensioned state/tend/ptend
- eliminated MODAL_AERO ifdef and now uses rad_constituent interface
- corrected bug found with Intel compiler in rho calculation
M models/atm/cam/src/physics/cam/physpkg.F90
M models/atm/cam/src/physics/cam/microp_driver.F90
- microp_aero_run and summing with microp ptend now called in tphysbc instead of inside microp_driver_tend
- modifications using new dynamically dimensioned state/tend/ptend
MM models/atm/cam/src/physics/cam/micro_mg1_0.F90
M models/atm/cam/src/physics/cam/micro_mg1_5.F90
- renamed sub_column to microp_uniform to better reflect its purpose
M models/atm/cam/src/physics/cam/tphysidl.F90
M models/atm/cam/src/physics/cam/rayleigh_friction.F90
M models/atm/cam/src/physics/cam/tracers.F90
M models/atm/cam/src/physics/cam/radheat.F90
M models/atm/cam/src/physics/cam/check_energy.F90
M models/atm/cam/src/physics/cam/stratiform.F90
M models/atm/cam/src/physics/cam/carma_intr.F90
M models/atm/cam/src/physics/cam/convect_shallow.F90
M models/atm/cam/src/physics/cam/macrop_driver.F90
M models/atm/cam/src/physics/cam/convect_deep.F90
M models/atm/cam/src/physics/cam/gw_drag.F90
M models/atm/cam/src/physics/cam/iondrag.F90
M models/atm/cam/src/physics/cam/aoa_tracers.F90
M models/atm/cam/src/physics/cam/zm_conv_intr.F90
M models/atm/cam/src/physics/cam/vertical_diffusion.F90
M models/atm/cam/src/physics/carma/cam/carma_intr.F90
M models/atm/cam/src/physics/carma/models/dust/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/pmc/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/bc_strat/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/test_tracers/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/sulfate/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/test_swelling/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/test_detrain/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/cirrus/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/test_radiative/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/meteor_smoke/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/test_growth/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/test_passive/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/sea_salt/carma_model_mod.F90
M models/atm/cam/src/physics/waccm/iondrag.F90
M models/atm/cam/src/physics/waccm/radheat.F90
M models/atm/cam/src/physics/waccm/gw_drag.F90
M models/atm/cam/src/physics/waccm/qbo.F90
M models/atm/cam/src/chemistry/pp_none/chemistry.F90
M models/atm/cam/src/chemistry/bulk_aero/progseasalts_intr.F90
M models/atm/cam/src/chemistry/bulk_aero/dust_intr.F90
M models/atm/cam/src/chemistry/bulk_aero/aerosol_intr.F90
M models/atm/cam/src/chemistry/mozart/chemistry.F90
M models/atm/cam/src/dynamics/sld/dp_coupling.F90
M models/atm/cam/src/dynamics/sld/stepon.F90
M models/atm/cam/src/dynamics/eul/dp_coupling.F90
M models/atm/cam/src/dynamics/eul/stepon.F90
M models/atm/cam/src/dynamics/homme/dp_coupling.F90
M models/atm/cam/src/dynamics/homme/stepon.F90
M models/atm/cam/src/dynamics/fv/dp_coupling.F90
M models/atm/cam/src/dynamics/fv/stepon.F90
- modifications using new dynamically dimensioned state/tend/ptend
M models/atm/cam/doc/ChangeLog
M models/atm/cam/test/system/tests_pretag_frankfurt_pgi
M models/atm/cam/test/system/tests_posttag_frankfurt
M models/atm/cam/test/system/tests_pretag_frankfurt_lahey
M models/atm/cam/bld/namelist_files
M models/atm/cam/src/control/camsrfexch.F90
M models/atm/cam/src/cpl_share/cam_cpl_indices.F90
M models/atm/cam/src/cpl_mct/atm_comp_mct.F90
M models/atm/cam/src/physics/cam/comsrf.F90
M models/atm/cam/src/physics/cam/wv_sat_methods.F90
M models/atm/cam/src/physics/cam/uwshcu.F90
M models/atm/cam/src/dynamics
M models/atm/cam/src/cpl_esmf/atm_comp_esmf.F90
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
040 bl366 TBL.sh f1.9c5clbdm outfrq3s 9s ..................................FAIL! rc= 7 at Wed Jan 30 16:31:23 MST 2013
- This is a new regression test for CLUBB. CLUBB was tested offline by both Cheryl Craig and Pete Bogenschutz.
frankfurt/lf95: all passed
frankfurt/pgi: all passed
Summarize any changes to answers, i.e.,: none
===============================================================
===============================================================
Tag name: cam5_2_08
Originator(s): fischer
Date: Jan 25th, 2013
One-line Summary: Remove SRFRAD, new diagnostics, chem preprocessor build,
update externals, fix dust_emis_fact
Purpose of changes:
-Remove unneeded SRFRAD from the monthly averaged output
AMWG diagnostics uses SRFRAD to compute flwds, but FLDS can
used instead.
-New diagnostics added for aerosol burdens and weed speeds.
-Update to the chemistry preprocessor to recognize the intel
compiler on yellowstone.
-SVN externals updated to match cesm1_2_beta01 externals.
-dust_emis_fact was being set to .35 for CAM-SE5, should be set to .55
Bugs fixed (include bugzilla ID):
-Bugzilla #1618, remove misleading SRFRAD variable
Describe any changes made to build system:
-New information about the fc_type is being passed to the chemistry
preprocessor build scripts. This is to correct a problem with the
chem preprocessor not recognizing the intel compiler on yellowstone.
Describe any changes made to the namelist:
-dust_emis_fact was being set to .35 instead of .55 when dyn="homme", tms="1", and
phys="cam5". The problem was there was a better match for the .35 value
than the .55 value. This was fixed by added tests for tms and phys to the
.55 value, making it the better match.
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/archive_baseline.sh
. Set read permissions for archived files
M models/atm/cam/test/system/tests_pretag_yellowstone
. Returned sm376, bl376, and ne996 test for the chemistry preprocessor
M models/atm/cam/bld/Makefile.in
. Add flag for the intel compiler to ignore temp array creation
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. Fixed bug where dust_emis_fact was being set to .35 instead of .55 for
CAM-SE 5
M models/atm/cam/bld/namelist_files/use_cases/sd_waccm_geos5.xml
M models/atm/cam/bld/namelist_files/use_cases/waccmx_2000_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/sd_waccm_tslt_geos5.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_refb1.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_sc_2000_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1850-2005_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2000_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2000_cam5.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp26.xml
M models/atm/cam/bld/namelist_files/use_cases/cam4_trop_strat_chem.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp45.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1955-2005_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp85.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1974-2005_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1850_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1850_cam5.xml
M models/atm/cam/bld/namelist_files/use_cases/waccmx_1996_cam4.xml
. Removed SRFRAD from namelist
M models/atm/cam/src/control/camsrfexch.F90
M models/atm/cam/src/physics/cam/radiation.F90
M models/atm/cam/src/physics/rrtmg/radiation.F90
. Removed srfrad. Moved calculation for netsw from camsrfexch to
the radiation_trend subroutines.
M models/atm/cam/bld/configure
M models/atm/cam/bld/perl5lib/Build/ChemPreprocess.pm
. Add new argument fc_type to be passed to ChemPreprocess.pm. This is needed to
build on yellowstone using intel.
M models/atm/cam/src/physics/cam/cam_diagnostics.F90
M models/atm/cam/src/physics/cam/modal_aer_opt.F90
. Add new diagnostics variables for aerosol burdens and wind speeds.
M SVN_EXTERNAL_DIRECTORIES
. Update externals to match cesm1_2_beta01
clm tag updated to branch_tags/nmlfix_tags/nmlfix01_clm4_0_58
pio tag updated to trunk_tags/pio1_5_7
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
049 bl376 TBL.sh fm1.9c4dh+testmech outfrq3s 9s ...........................FAIL! rc= 7 at Fri Jan 25 09:54:03 MST 2013
. expected since +testmech was broken on the previous tag
084 bl993 TBL_ccsm.sh ne30_g16 FC5 2d .....................................FAIL! rc= 7 at Fri Jan 25 10:32:33 MST 2013
. answer changes expected for CAM_SE5
frankfurt/lf95: all passed
frankfurt/pgi: all passed
Summarize any changes to answers, i.e.,: all bfb, except CAM-SE using cam5
physics.
===============================================================
===============================================================
Tag name: cam5_2_07
Originator(s): santos
Date: Jan 23, 2012
One-line Summary: HOMME update, move repro_sum_mod, WACCM-SE
Purpose of changes:
- Update CESM externals.
- Note: "setup" has been renamed to "cesm_setup", and
"env_pesetup.xml" has been renamed back to "env_mach_pes.xml"
- csm_share updated for to get working (and recently
improved!) shr_reprosum_mod.
- During debugging, updated the driver, Machines, and scripts.
While this was unnecessary, these were verified BFB, so
were updated anyway.
- RTM had to be updated for the driver update. build_namelist
had to be updated due to RTM namelist changes. Note that
standalone CAM has RTM off by default, so no answer changes.
- Update HOMME external (and make changes to interface layer
to accomodate this).
- Add changes that were waiting on this update:
- repro_sum_mod is deleted, and modules now use
shr_reprosum_mod from csm_share
- New routine and first namelist options for WACCM-SE
(WACCM+HOMME, still under development).
- Fix WACCM5 restart test that fails since cam5_2_05 (diagnostic
variables were not being initialized above top_lev, so random
unused garbage above that level was in the output files).
- Fix HOMME settings for ne120np4.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
- Replace do_rtm = .false. with rtm_mode = "NULL", for RTM update.
- Standalone CAM still allows modification of reprosum module
options, but these are now written to a driver namelist.
- New HOMME settings for ne120np4.
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
D models/atm/cam/src/utils/repro_sum_x86.c
D models/atm/cam/src/utils/repro_sum_mod.F90
- Eliminated in favor of shr_reprosum_mod in csm_share.
List all subroutines added and what they do:
A models/atm/cam/src/dynamics/homme/gravity_waves_sources.F90
- New module with HOMME-compatible frontogenesis calculation
(for WACCM).
List all existing files that have been modified, and describe the changes:
M SVN_EXTERNAL_DIRECTORIES
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
- Update driver, RTM, csm_share, scripts, Machines, and HOMME.
M models/atm/cam/bld/build-namelist
- Update for RTM with CAM standalone.
M models/atm/cam/bld/configure
- Allow WACCM+HOMME configurations.
- Define NC=4 for new HOMME.
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2000_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_sc_2000_cam4.xml
- WACCM-SE defaults for ne30np4.
- Specify which defaults in the use_cases are for fv only.
- Add ne30np4 data set for FWSC compset.
- Fix default parameters for ne120np4 resolution.
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- Move RTM and reprosum options (see above).
M models/atm/cam/src/chemistry/modal_aero/modal_aero_rename.F90
- Remove use of shr_kind_r4 (cleanup).
M models/atm/cam/src/chemistry/mozart/mo_lightning.F90
M models/atm/cam/src/dynamics/fv/d2a3dikj.F90
M models/atm/cam/src/dynamics/fv/fv_prints.F90
M models/atm/cam/src/dynamics/fv/mean_module.F90
M models/atm/cam/src/dynamics/fv/p_d_adjust.F90
M models/atm/cam/src/dynamics/fv/par_xsum.F90
M models/atm/cam/src/physics/cam/phys_gmean.F90
M models/atm/cam/src/physics/cam/polar_avg.F90
- Change use of repro_sum_mod to shr_reprosum_mod.
M models/atm/cam/src/control/cam_comp.F90
- Nullify pointers in module scope.
M models/atm/cam/src/control/cam_restart.F90
- Change dyn_in/dyn_out declarations to "intent(inout)"
(to preserve allocation of pointer components).
M models/atm/cam/src/control/runtime_opts.F90
- Remove processing of repro_sum_mod options.
M models/atm/cam/src/dynamics/fv/ctem.F90
- Put explicit limits on arrays in par_xsum call (leftover
cleanup from debugging, will be necessary for NAG).
M models/atm/cam/src/dynamics/fv/dp_coupling.F90
M models/atm/cam/src/physics/cam/phys_control.F90
M models/atm/cam/src/physics/cam/physics_types.F90
- Remove WACCM_PHYS "ifdef" statements in favor of equivalent
runtime switch in phys_control.
- When dynamic state changes arrive, this will also be used
to change allocation of frontgf/frontga.
D models/atm/cam/src/physics/waccm/gravity_waves_sources.F90
A models/atm/cam/src/dynamics/fv/gravity_waves_sources.F90
- Moved FV-specific gravity_waves_sources from WACCM, since
this module is dycore-specific.
M models/atm/cam/src/dynamics/homme/dp_coupling.F90
M models/atm/cam/src/dynamics/homme/stepon.F90
- Add call to gravity_waves_sources for WACCM.
- HOMME update changes (see below).
- Check for negative temperatures in d_p_coupling and abort
with a debug message if any are found.
M models/atm/cam/src/dynamics/homme/dyn_comp.F90
M models/atm/cam/src/dynamics/homme/dyn_grid.F90
M models/atm/cam/src/dynamics/homme/inidat.F90
M models/atm/cam/src/dynamics/homme/restart_dynamics.F90
- Change interface to HOMME for the updated version.
- Remove timelevel dimension from elem%state%Q.
- Add calls to Timelevel_Qdp to get Qdp-specific timelevel
index.
- Rename references to "cslam" to "fvm".
- Remove logic for deprecated or removed options.
- "leapfrog" scheme not supported.
- tracer_advection_formulation is now always
TRACERADV_TOTAL_DIVERGENCE.
- dyn_import_t and dyn_export_t are now created with pointer
components nullified. This means that they usually must be
"intent(inout)" arguments to maintain those pointers'
association. (Using "intent(out)" was already incorrect
Fortran, but still worked with most compilers.)
M models/atm/cam/src/physics/cam/geopotential.F90
- Changed arrays to assumed shape in order to prevent
creation of temporary copies (flagged by ifort).
M models/atm/cam/src/physics/cam/micro_mg1_5.F90
- Change mistaken use of qin to qi (bug fix for
sub_column=.true., which is not yet turned on).
- Correct spelling of "heterogeneous".
M models/atm/cam/src/physics/cam/micro_mg_cam.F90
- Initialize diagnostic output above top_lev (prevents
garbage from being output above top_lev in WACCM5).
M models/atm/cam/src/physics/cam/pkg_cld_sediment.F90
M models/atm/cam/src/physics/cam/vertical_diffusion.F90
- Pure cleanup changes.
M models/atm/cam/src/physics/waccm/qbo.F90
- Removed unused FV-specific dependencies. QBO forcing does
not currently work with HOMME, but removing these dependencies
is necessary for the module to be compiled in a HOMME case.
M models/atm/cam/test/system/archive_baseline.sh
- New script for archiving CAM regression test baselines on
current testing machines.
M models/atm/cam/test/system/TCB_ccsm.sh
- Update for CESM scripts.
M models/atm/cam/test/system/test_driver.sh
- Remove "impi" module on yellowstone, as MPICH is supposedly
faster.
M models/atm/cam/tools/interpic_new/Makefile
- Add ifort and make it default compiler (so that interpic_new
can be used on yellowstone out-of-the-box).
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
- All baselines using the HOMME dycore are expected to fail due to
the updated external.
- All baselines for cases that used repro_sum_mod are expected to
fail due to updates to the csm_share version.
- In practice, this means HOMME cases (which would fail baseline
tests anyway), and most FV cases.
yellowstone/intel:
013 bl330 TBL.sh f4c5paqdh aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Fri Jan 11 03:02:02 MST 2013
017 bl331 TBL.sh f4c4dh co2rmp+1850_cam4 9s ...............................FAIL! rc= 7 at Fri Jan 11 03:05:43 MST 2013
021 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...............................FAIL! rc= 7 at Fri Jan 11 03:20:21 MST 2013
026 bl334 TBL.sh f4adh fv2d_8tsk 9s .......................................FAIL! rc= 7 at Fri Jan 11 03:21:51 MST 2013
030 bl335 TBL.sh f4idm idphys 9s ..........................................FAIL! rc= 7 at Fri Jan 11 03:23:22 MST 2013
033 bl336 TBL.sh wm4c4h outfrq3s+waccm_1850_cam4 9s .......................FAIL! rc= 7 at Fri Jan 11 03:26:27 MST 2013
036 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ......................................FAIL! rc= 7 at Fri Jan 11 03:34:40 MST 2013
040 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .....................................FAIL! rc= 7 at Fri Jan 11 04:04:58 MST 2013
045 bl373 TBL.sh fm1.9c4dh outfrq3s 9s ....................................FAIL! rc= 7 at Fri Jan 11 05:03:22 MST 2013
047 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...........................FAIL! rc= 7 at Fri Jan 11 05:07:13 MST 2013
050 bl374 TBL.sh fs1.9c4dh outfrq3s+1850-2005_cam4_super_fast_llnl 9s .....FAIL! rc= 7 at Fri Jan 11 05:18:56 MST 2013
052 bl415 TBL.sh fst1.9c5dh outfrq3s+cam5_trop_strat_chem 9s ..............FAIL! rc= 7 at Fri Jan 11 05:55:46 MST 2013
054 bl430 TBL.sh fm1.9c5dh outfrq3s 9s ....................................FAIL! rc= 7 at Fri Jan 11 06:24:36 MST 2013
057 bl440 TBL.sh fsoa1.9c4dh outfrq3s+mozart_megan_emis 9s ................FAIL! rc= 7 at Fri Jan 11 07:01:50 MST 2013
061 bl387 TBL.sh f1.9c4dm outfrq3s+1850-2005_cam4 9s ......................FAIL! rc= 7 at Fri Jan 11 07:10:08 MST 2013
064 bl388 TBL.sh f1.9c4m outfrq24h+1850_cam4 2d ...........................FAIL! rc= 7 at Fri Jan 11 07:12:28 MST 2013
067 bl379 TBL.sh wm1.9c4h outfrq24h+waccm_1850_cam4 2d ....................FAIL! rc= 7 at Fri Jan 11 07:50:50 MST 2013
070 bl731 TBL.sh h16c3aqdh aqua+aquaplanet_cam3 9s ........................FAIL! rc= 7 at Fri Jan 11 08:00:33 MST 2013
073 bl751 TBL.sh h30c4aqdm outfrq3s+aquaplanet_cam4 9s ....................FAIL! rc= 7 at Fri Jan 11 08:18:07 MST 2013
079 bl992 TBL_ccsm.sh f19_g16 ETEST 2d ....................................FAIL! rc= 7 at Fri Jan 11 08:27:19 MST 2013
082 bl993 TBL_ccsm.sh ne30_g16 FC5 2d .....................................FAIL! rc= 7 at Fri Jan 11 08:59:27 MST 2013
bluefire: Most tests run with identical results to yellowstone, but machine
is being retired
frankfurt/lf95:
017 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Thu Jan 10 19:06:24 MST 2013
025 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Thu Jan 10 21:55:34 MST 2013
027 bl313 TBL.sh f10c3aqdm outfrq3s+aquaplanet_cam3 3s ....................FAIL! rc= 7 at Thu Jan 10 22:07:33 MST 2013
030 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7 at Thu Jan 10 22:37:52 MST 2013
032 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 7 at Thu Jan 10 23:01:07 MST 2013
035 bl318 TBL.sh f10c4cdm sat_hist 9s .....................................FAIL! rc= 7 at Thu Jan 10 23:23:40 MST 2013
frankfurt/pgi:
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Thu Jan 10 18:03:55 MST 2013
031 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Thu Jan 10 18:32:55 MST 2013
033 bl313 TBL.sh f10c3aqdm outfrq3s+aquaplanet_cam3 3s ....................FAIL! rc= 7 at Thu Jan 10 18:37:10 MST 2013
036 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7 at Thu Jan 10 18:44:21 MST 2013
039 bl315 TBL.sh f10c3dm outfrq3s+cam3 9s .................................FAIL! rc= 7 at Thu Jan 10 18:49:26 MST 2013
041 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 7 at Thu Jan 10 18:54:43 MST 2013
044 bl317 TBL.sh f10c5cdm outfrq3s 9s .....................................FAIL! rc= 7 at Thu Jan 10 19:02:42 MST 2013
047 bl319 TBL.sh fn10c5dm outfrq3s_bam 9s .................................FAIL! rc= 7 at Thu Jan 10 19:08:34 MST 2013
050 bl320 TBL.sh fn10c5dm rad_diag 9s .....................................FAIL! rc= 7 at Thu Jan 10 19:11:07 MST 2013
053 bl321 TBL.sh f10c5cdm atrain 9s .......................................FAIL! rc= 7 at Thu Jan 10 19:14:42 MST 2013
056 bl420 TBL.sh f7mode4c5dm outfrq3s 9s ..................................FAIL! rc= 7 at Thu Jan 10 19:30:33 MST 2013
059 bl734 TBL.sh hn16c5aqdm outfrq3s_bam+aquaplanet_cam5 9s ...............FAIL! rc= 7 at Thu Jan 10 20:02:34 MST 2013
062 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ....................FAIL! rc= 7 at Thu Jan 10 21:02:17 MST 2013
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: HOMME and FV dycores
- what platforms/compilers: all
- nature of change (roundoff; larger than roundoff but same climate; new
climate): roundoff (due to shr_reprosum_mod and HOMME)
If bitwise differences were observed, how did you show they were no worse
than roundoff?
- Testing of shr_reprosum_mod and HOMME output outside of CAM.
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_2_06
Originator(s): fischer
Date: Dec 10 2012
One-line Summary: bug fix for yellowstone
Purpose of changes:
. Use cam erfc function in modal_aero_rename.F90 instead
of intrinsic erfc function. This is to get around an error
on yellowstone.
. Change the queue name for the yellowstone test build
script from caldera go gpgpu. This will give you a
dedicated node on caldera, instead of sharing a node.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: fischer
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/test_driver.sh
. Changed caldera queue to gpgpu to get a dedicate node on caldera to do
the pre builds.
M models/atm/cam/src/chemistry/modal_aero/modal_aero_rename.F90
. Use cam erfc function instead of intrinsic erfc. Will cause answer
changes for cam5.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Sat Dec 8 00:08:12 MST 2012
013 bl330 TBL.sh f4c5paqdh aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Sat Dec 8 00:12:07 MST 2012
021 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...............................FAIL! rc= 7 at Sat Dec 8 00:36:50 MST 2012
036 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ......................................FAIL! rc= 7 at Sat Dec 8 00:59:45 MST 2012
040 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .....................................FAIL! rc= 7 at Sat Dec 8 01:43:28 MST 2012
047 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...........................FAIL! rc= 7 at Sat Dec 8 03:51:51 MST 2012
052 bl415 TBL.sh fst1.9c5dh outfrq3s+cam5_trop_strat_chem 9s ..............FAIL! rc= 7 at Sat Dec 8 05:48:46 MST 2012
054 bl430 TBL.sh fm1.9c5dh outfrq3s 9s ....................................FAIL! rc= 7 at Sat Dec 8 07:04:02 MST 2012
082 bl993 TBL_ccsm.sh ne30_g16 FC5 2d .....................................FAIL! rc= 7 at Sat Dec 8 11:29:06 MST 2012
. cam5 baselines expected to fail
bluefire:
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Mon Dec 10 08:29:36 MST 2012
013 bl330 TBL.sh f4c5paqdh aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Mon Dec 10 08:29:39 MST 2012
021 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...............................FAIL! rc= 7 at Mon Dec 10 08:29:51 MST 2012
036 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ......................................FAIL! rc= 7 at Mon Dec 10 08:30:00 MST 2012
040 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .....................................FAIL! rc= 7 at Mon Dec 10 08:30:26 MST 2012
047 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...........................FAIL! rc= 7 at Mon Dec 10 08:30:45 MST 2012
054 bl415 TBL.sh fst1.9c5dh outfrq3s+cam5_trop_strat_chem 9s ..............FAIL! rc= 7 at Mon Dec 10 09:54:35 MST 2012
056 bl430 TBL.sh fm1.9c5dh outfrq3s 9s ....................................FAIL! rc= 7 at Mon Dec 10 10:29:45 MST 2012
084 bl993 TBL_ccsm.sh ne30_g16 FC5 2d .....................................FAIL! rc= 7 at Mon Dec 10 11:32:58 MST 2012
. cam5 baselines expected to fail
frankfurt/lf95:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 5 at Mon Dec 10 08:32:42 MST 2012
009 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 5 at Mon Dec 10 08:32:57 MST 2012
013 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 5 at Mon Dec 10 08:32:57 MST 2012
017 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 5 at Mon Dec 10 08:33:11 MST 2012
025 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 5 at Mon Dec 10 08:33:25 MST 2012
032 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 5 at Mon Dec 10 08:33:39 MST 2012
. cam5 baselines expected to fail
frankfurt/pgi:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 5 at Fri Dec 7 14:51:27 MST 2012
010 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 5 at Fri Dec 7 14:57:45 MST 2012
012 bl113 TBL.sh e8c5paqdm aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 5 at Fri Dec 7 15:01:36 MST 2012
016 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 5 at Fri Dec 7 15:05:51 MST 2012
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 5 at Fri Dec 7 15:21:43 MST 2012
031 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 5 at Fri Dec 7 15:50:54 MST 2012
041 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 5 at Fri Dec 7 16:21:03 MST 2012
044 bl317 TBL.sh f10c5cdm outfrq3s 9s .....................................FAIL! rc= 5 at Fri Dec 7 16:28:08 MST 2012
047 bl319 TBL.sh fn10c5dm outfrq3s_bam 9s .................................FAIL! rc= 5 at Fri Dec 7 16:33:12 MST 2012
050 bl320 TBL.sh fn10c5dm rad_diag 9s .....................................FAIL! rc= 5 at Fri Dec 7 16:35:52 MST 2012
053 bl321 TBL.sh f10c5cdm atrain 9s .......................................FAIL! rc= 5 at Fri Dec 7 16:40:07 MST 2012
056 bl420 TBL.sh f7mode4c5dm outfrq3s 9s ..................................FAIL! rc= 5 at Fri Dec 7 16:55:40 MST 2012
059 bl734 TBL.sh hn16c5aqdm outfrq3s_bam+aquaplanet_cam5 9s ...............FAIL! rc= 5 at Fri Dec 7 17:26:16 MST 2012
062 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ....................FAIL! rc= 5 at Fri Dec 7 18:26:50 MST 2012
. cam5 baselines expected to fail
===============================================================
===============================================================
Tag name: cam5_2_05
Originator(s): andrew, morrison, santos
Date: Dec 6, 2012
One-line Summary: New "MG1.5" microphysics scheme
Purpose of changes:
- Added new microphysics scheme 'mg1.5', the next development
version of the Morrison & Gettelman stratiform cloud microphysics.
Bugs fixed (include bugzilla ID):
- Bug 1600: CARMA cirrus model fails in build_namelist in cam5_2_02
- This bug was introduced in cam5_2_02, when the name of wvsat_scheme
was changed to wv_sat_scheme. The name was not changed in the
add_default call in build_namelist, so the cirrus model would not
build.
Describe any changes made to build system:
- Added mg1.5 option to -microphys configure option.
Describe any changes made to the namelist:
- Added micro_mg_version and micro_mg_sub_version (MG version numbers).
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
A models/atm/cam/src/physics/cam/micro_mg1_5.F90
- Alternative microphysics related to micro_mg.F90
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/build-namelist
- Add micro_mg_version and micro_mg_sub_version to
represent the version number of MG.
- Fix bug with CARMA cirrus model and wv_sat_scheme.
M models/atm/cam/bld/config_files/definition.xml
M models/atm/cam/bld/configure
- Added "mg1.5" as valid option for -microphys.
- "mg" is preserved as an alias for the old microphysics (which
is now "mg1").
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- Add version number for MG.
- Added 1.9x2.5 MG1.5 defaults for cldfrc_rhminl and uwshcu_rpen.
M models/atm/cam/src/chemistry/utils/tracer_data.F90
- Initialize pointers in derived types to null (bug fix for
gfortran, courtesy jedwards).
M models/atm/cam/src/control/history_defaults.F90
M models/atm/cam/src/dynamics/eul/inidat.F90
M models/atm/cam/src/dynamics/eul/iop.F90
M models/atm/cam/src/dynamics/eul/tfilt_massfix.F90
M models/atm/cam/src/dynamics/fv/inidat.F90
M models/atm/cam/src/dynamics/homme/inidat.F90
M models/atm/cam/src/dynamics/sld/inidat.F90
M models/atm/cam/src/physics/cam/cloud_fraction.F90
M models/atm/cam/src/physics/cam/cospsimulator_intr.F90
M models/atm/cam/src/physics/cam/macrop_driver.F90
M models/atm/cam/src/physics/rrtmg/cloud_rad_props.F90
M models/atm/cam/src/physics/rrtmg/ebert_curry.F90
M models/atm/cam/src/physics/rrtmg/oldcloud.F90
M models/atm/cam/src/physics/rrtmg/radiation.F90
M models/atm/cam/src/physics/rrtmg/slingo.F90
- Remove unused (or unnecessary) microp_scheme variable.
M models/atm/cam/src/physics/cam/cldwat.F90
M models/atm/cam/src/physics/cam/cloud_diagnostics.F90
M models/atm/cam/src/physics/cam/conv_water.F90
M models/atm/cam/src/physics/cam/convect_shallow.F90
M models/atm/cam/src/physics/cam/microp_driver.F90
M models/atm/cam/src/physics/cam/phys_control.F90
M models/atm/cam/src/physics/cam/physics_types.F90
M models/atm/cam/src/physics/cam/physpkg.F90
M models/atm/cam/src/physics/cam/pkg_cldoptics.F90
M models/atm/cam/src/physics/cam/radiation_data.F90
M models/atm/cam/src/physics/cam/radsw.F90
M models/atm/cam/src/physics/cam/vertical_diffusion.F90
- Files that did not need to be changed, but retain "cleanup"
or stylistic changes due to changes to microp_scheme in a
previous stage.
D models/atm/cam/src/physics/cam/micro_mg.F90
A models/atm/cam/src/physics/cam/micro_mg1_0.F90
- File moved to emphasize place within series.
M models/atm/cam/src/physics/cam/micro_mg_cam.F90
- Added MG1.5 call, uniform use of top_lev.
M models/atm/cam/src/physics/cam/stratiform.F90
- stratiform_implements_cnst now always returns .false.
if microp_scheme is not RK.
M models/atm/cam/src/physics/cam/wv_sat_methods.F90
- "save" attribute is now global.
- Added qsat methods (non-optional logic from wv_saturation).
M models/atm/cam/src/physics/cam/wv_saturation.F90
- Moved qsat_water/ice logic to wv_sat_methods.
A models/atm/cam/test/system/config_files/f1.0c5mg1_5m
M models/atm/cam/test/system/input_tests_master
M models/atm/cam/test/system/tests_posttag_frankfurt
- Add test for mg1.5.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Thu Dec 6 11:46:02 MST 2012
021 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...............................FAIL! rc= 7 at Thu Dec 6 12:14:45 MST 2012
036 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ......................................FAIL! rc= 7 at Thu Dec 6 12:37:43 MST 2012
040 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .....................................FAIL! rc= 7 at Thu Dec 6 13:22:20 MST 2012
047 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...........................FAIL! rc= 7 at Thu Dec 6 15:04:19 MST 2012
052 bl415 TBL.sh fst1.9c5dh outfrq3s+cam5_trop_strat_chem 9s ..............FAIL! rc= 7 at Thu Dec 6 17:01:20 MST 2012
054 bl430 TBL.sh fm1.9c5dh outfrq3s 9s ....................................FAIL! rc= 7 at Thu Dec 6 18:17:04 MST 2012
082 bl993 TBL_ccsm.sh ne30_g16 FC5 2d .....................................FAIL! rc= 7 at Thu Dec 6 22:51:13 MST 2012
bluefire:
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Thu Dec 6 16:46:46 MST 2012
021 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...............................FAIL! rc= 7 at Thu Dec 6 16:46:54 MST 2012
036 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ......................................FAIL! rc= 7 at Thu Dec 6 16:47:01 MST 2012
040 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .....................................FAIL! rc= 7 at Thu Dec 6 16:47:15 MST 2012
047 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...........................FAIL! rc= 7 at Thu Dec 6 16:47:27 MST 2012
054 bl415 TBL.sh fst1.9c5dh outfrq3s+cam5_trop_strat_chem 9s ..............FAIL! rc= 7 at Thu Dec 6 16:48:09 MST 2012
056 bl430 TBL.sh fm1.9c5dh outfrq3s 9s ....................................FAIL! rc= 7 at Thu Dec 6 16:48:39 MST 2012
084 bl993 TBL_ccsm.sh ne30_g16 FC5 2d .....................................FAIL! rc= 7 at Thu Dec 6 19:44:23 MST 2012
frankfurt/lf95:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Wed Dec 5 14:13:13 MST 2012
009 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Wed Dec 5 15:07:40 MST 2012
013 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Wed Dec 5 15:15:56 MST 2012
017 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Wed Dec 5 16:24:19 MST 2012
025 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Wed Dec 5 20:24:51 MST 2012
frankfurt/pgi:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Wed Dec 5 13:28:35 MST 2012
010 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Wed Dec 5 13:38:17 MST 2012
016 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Wed Dec 5 13:50:04 MST 2012
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Wed Dec 5 14:08:13 MST 2012
031 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Wed Dec 5 14:44:06 MST 2012
044 bl317 TBL.sh f10c5cdm outfrq3s 9s .....................................FAIL! rc= 7 at Wed Dec 5 15:25:27 MST 2012
047 bl319 TBL.sh fn10c5dm outfrq3s_bam 9s .................................FAIL! rc= 7 at Wed Dec 5 15:33:30 MST 2012
050 bl320 TBL.sh fn10c5dm rad_diag 9s .....................................FAIL! rc= 7 at Wed Dec 5 15:37:02 MST 2012
053 bl321 TBL.sh f10c5cdm atrain 9s .......................................FAIL! rc= 7 at Wed Dec 5 15:42:24 MST 2012
056 bl420 TBL.sh f7mode4c5dm outfrq3s 9s ..................................FAIL! rc= 7 at Wed Dec 5 16:06:10 MST 2012
059 bl734 TBL.sh hn16c5aqdm outfrq3s_bam+aquaplanet_cam5 9s ...............FAIL! rc= 7 at Wed Dec 5 16:50:56 MST 2012
062 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ....................FAIL! rc= 7 at Wed Dec 5 18:16:15 MST 2012
- On all systems, CAM5 baselines were expected to fail due to a change in
diagnostic output. Outside of this output, the model is bit-for-bit.
- The changes are in these variables: AREI AREL AWNC AWNI FREQI FREQL
- The only difference is that these variables are now scaled by the
cloud fraction (liquid or ice, as appropriate).
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name:cam5_2_04
Originator(s): fischer
Date: Tue Dec 5 14:54:12 MST 2012
One-line Summary: merge in changes from the release tag, yellowstone testing
support
Purpose of changes:
. Merge in changes from release branch tag
- update for mozart_megan_emis.xml
- cam-se topo tools from Peter L.
- Support for SCAM running from cesm script fixed
- Changed QNEG3 lchnk format from i5 to i7
- EMIS addfld bug fix
. Fix fincl6lonlat namelist bug
. Support to run cam's test suite on yellowstone.
- test_driver.sh first submits a build script to
caldera to build all the needed exacutables.
- The a run script is submitted to yellowstone.
- There will be 2 td.*.status files, the first
is for the builds, the second is for the runs.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: fischer
List all subroutines eliminated:
List all subroutines added and what they do:
A models/atm/cam/test/system/TNE_cb_ccsm.sh
A models/atm/cam/test/system/TBL_cb.sh
A models/atm/cam/test/system/TSC_cb.sh
A models/atm/cam/test/system/TBL_cb_ccsm.sh
- script needed to do all pre builds, only used on yellowstone at this time.
A models/atm/cam/test/system/tests_pretag_yellowstone
A models/atm/cam/test/system/tests_posttag_yellowstone
- New test lists for yellowstone, the same as bluefire test lists,
except for +testmech tests were removed
A models/atm/cam/tools/topo_tool
- topography generation tools
A models/atm/cam/bld/namelist_files/use_cases/arm95_scam.xml
- use case need to run SCAM with cesm scripts
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/TCB_ccsm.sh
- Change pes layout for yellowstone
M models/atm/cam/test/system/test_driver.sh
- Added ability to read in an existing test list and build
everthing that's needed. Only used on yellowstone.
M models/atm/cam/test/system/CAM_runcmnd.sh
- updated to run on yellowstone
M models/atm/cam/src/control/runtime_opts.F90
- Add fincl6lonlat
M models/atm/cam/bld/Makefile.in
- set rules to build dyn_grid.F90 due to mpich2 v1208 bug
- added rpath to LIB_NETCDF to build test_nc
M models/atm/cam/bld/namelist_files/use_cases/mozart_megan_emis.xml
- updated
M models/atm/cam/doc/ChangeLog_template
- Add yellowstone/intel to testing results
M models/atm/cam/src/cpl_mct/atm_comp_mct.F90
- Fixes to run SCAM with cesm scripts
M models/atm/cam/src/physics/cam/qneg3.F90
- Changed QNEG3 lchnk format from i5 to i7
M models/atm/cam/src/physics/carma/models/dust/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/pmc/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/bc_strat/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/sulfate/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/cirrus/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/meteor_smoke/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/sea_salt/carma_model_mod.F90
- Removed old comments
M SVN_EXTERNAL_DIRECTORIES
- Updated externals to match cesm1_1_beta18. Machines and scripts tags were
updated to newer version to get yellowstone support.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
yellowstone/intel:
All TBL tests failed because previous tag didn't fully support yellowstone.
bluefire:
059 bl440 TBL.sh fsoa1.9c4dh outfrq3s+mozart_megan_emis 9s ................FAIL! rc= 7 at Sat Dec 1 20:34:10 MST 2012
update to megan emis use case
frankfurt/lf95:
all passed
frankfurt/pgi:
all passed
Summarize any changes to answers, i.e.,
===============================================================
===============================================================
Tag name: cam5_2_03
Originator(s): fvitt
Date: 26 Nov 2012
One-line Summary: Chemistry change and fix for WACCM5
Purpose of changes:
- correction to trop_strat_mam3 chemistry (removed NH4NO3)
- correction in RRTMG radiation code for WACCM5
- a workaround for OpenMP problem in SC-WACCM when compiled with
intel compiler (yellowstone)
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/configure
- reduced number of advected tracers for trop_strat_mam3 chemistry package
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- changed default IC file for trop_strat_mam7 chemistry
M models/atm/cam/bld/namelist_files/use_cases/waccm_2000_cam5.xml
- added IC file for WACCM5 at 4x5 resolution
M models/atm/cam/src/physics/rrtmg/radsw.F90
- corrections WACCM5
M models/atm/cam/src/chemistry/utils/tracer_data.F90
- a workaround for OpenMP problem in SC-WACCM when compiled with intel compiler
- corrections to reading of zonal-averaged input data which applies only to
varied arrangment of data dimensions in input file
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_rxt_rates_conv.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/m_spc_id.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/chem_mech.in
- removed NH4NO3 from chemical mechanism
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
054 bl415 TBL.sh fst1.9c5dh outfrq3s+cam5_trop_strat_chem 9s ..............FAIL! rc= 7 at Thu Nov 15 16:27:37 MST 2012
expected failure due to correction to trop_strat_mam3 chemistry package
frankfurt/lf95: all pass
frankfurt/pgi: all pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_2_02
Originator(s): santos
Date: Nov 9, 2012
One-line Summary: Refactoring of wv_saturation module
Purpose of changes:
- Refactoring of wv_saturation:
- Consolidate routines in wv_saturation.
- Pull in related logic from other modules to wv_saturation.
- Create wv_sat_methods to hold SVP estimation methods separately
from interface logic.
- Update interfaces throughout CAM to match.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
- Changed wvsat_scheme in phys_control to wv_sat_scheme in
wv_saturation.
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
D models/atm/cam/src/physics/cam/esinti.F90
- Removed in favor of wv_sat_init in wv_saturation.
D models/atm/cam/src/physics/cam/gffgch.F90
- Moved to gffgch_legacy in wv_saturation, in anticipation of
it being phased out.
List all subroutines added and what they do:
A models/atm/cam/src/physics/cam/wv_sat_methods.F90
- Contains the basic formulae used to calculate SVP, and
simple lookup routines for finding them.
- Contains the basic formulae to convert SVP to saturation
specific humidity and saturation mass mixing ratio.
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- Moved wvsat_scheme to wv_sat_scheme in wv_sat_nl.
M models/atm/cam/src/chemistry/bulk_aero/progseasalts_intr.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_newnuc.F90
M models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
M models/atm/cam/src/chemistry/utils/modal_aero_wateruptake.F90
M models/atm/cam/src/physics/cam/aer_rad_props.F90
M models/atm/cam/src/physics/cam/cam_diagnostics.F90
M models/atm/cam/src/physics/cam/cloud_fraction.F90
M models/atm/cam/src/physics/cam/convect_shallow.F90
M models/atm/cam/src/physics/cam/cospsimulator_intr.F90
M models/atm/cam/src/physics/cam/hk_conv.F90
M models/atm/cam/src/physics/cam/micro_mg_cam.F90
M models/atm/cam/src/physics/cam/microp_aero.F90
M models/atm/cam/src/physics/cam/ndrop.F90
M models/atm/cam/src/physics/cam/ndrop_bam.F90
M models/atm/cam/src/physics/cam/nucleate_ice.F90
M models/atm/cam/src/physics/cam/radae.F90
M models/atm/cam/src/physics/cam/vertical_diffusion.F90
M models/atm/cam/src/physics/cam/zm_conv.F90
- Changed calls to qsat/qsat_water/qsat_ice and/or legacy svp
functions to match new interface.
M models/atm/cam/src/physics/cam/eddy_diff.F90
M models/atm/cam/src/physics/cam/micro_mg.F90
- Change svp calls to match new interface.
- Now get gamma and svp functions through use statement
rather than having them passed in.
- micro_mg.F90 uses portable layer rather than wv_saturation.
M models/atm/cam/src/chemistry/mozart/mo_aerosols.F90
- Removed unnecessary dependence on wv_saturation.
M models/atm/cam/src/control/physconst.F90
- Added the triple point of water as h2otrip (used from
csm_share, equivalent to triple point already used in
polysvp).
- Removed unused variables.
M models/atm/cam/src/control/runtime_opts.F90
- Add wv_saturation's namelist.
M models/atm/cam/src/physics/cam/cldwat.F90
M models/atm/cam/src/physics/cam/cldwat2m_macro.F90
M models/atm/cam/src/physics/cam/uwshcu.F90
- Change dependencies on wv_saturation to dependencies on
physconst.
- Move findsp to wv_saturation.
- Change interface of qsat/svp calls.
- uwshcu gets qsat through use statement rather than as
an argument.
M models/atm/cam/src/physics/cam/macrop_driver.F90
M models/atm/cam/src/physics/rrtmg/radiation.F90
M models/atm/cam/src/utils/cam_aqua/shr_flux_mod.F90
- Trivial cleanup unrelated to wv_saturation (e.g. removal of
unused dependencies, comment typos).
M models/atm/cam/src/physics/cam/phys_control.F90
- Remove wvsat_scheme.
M models/atm/cam/src/physics/cam/physpkg.F90
- Call init/final for wv_saturation module.
- Move some "use" statements around to deal with a PGI bug
for large modules.
M models/atm/cam/src/physics/cam/wv_saturation.F90
- Completely refactored. A list of public routines:
- Control routines, replacing esinti and other CAM logic:
- wv_sat_readnl
- wv_sat_init
- wv_sat_final
- SVP calculations, replacing polysvp:
- svp_water
- svp_ice
- SVP table lookup, largely unchanged:
- estblf
- SVP to saturation specific humidity conversion:
- svp_to_qsat
- SVP to saturation mass mixing ratio conversion:
- svp_to_qmmr
- Saturation specific humidity functions, replacing all
subroutines of the form [a,v]qsat[d[2]][_single]
- qsat
- qsat_water
- qsat_ice
- Wet bulb temperature solver, replacing findsp[_water]:
- findsp_vc
- "Legacy" routines, included only to preserve answers:
- gffgch_legacy, gffgch_legacy_water
- qsat_water_legacy
- qmmr_zm
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS
frankfurt/lf95: All PASS
frankfurt/pgi or jaguar/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_2_01
Originator(s): eaton
Date: Mon Oct 29 10:42:35 MDT 2012
One-line Summary: build changes for yellowstone; remove LCLOUD from default hist output
Purpose of changes:
. Made mods to configure and Makefile.in to account for new compiler
wrappers used on yellowstone.
. Made some changes to the MCT build part of CAM's configure to allow
trapping the error return from the MCT configure script.
. Modified the Makefile rules used by configure's -test option so they work
with the recent changes to build MCT as a separate lib.
. Remove LCLOUD from the default history output. It is only used by the
stratus cloud activation, and is not the same as the fraction of liquid
stratus which is available as LIQCLDF (but is not output by default).
. Move the addfld/outfld calls for AST, LIQCLDF, ICECLDF from micro_mg_cam
to macrop_driver since these fields are computed by the macrophysics.
Bugs fixed (include bugzilla ID):
. fix addfld call for EMIS in cam5 radiation (rrtmg).
Describe any changes made to build system:
. mods for yellowstone and MCT build as noted above
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/configure
. set default CC to mpicc when mpif90 is the fortran compiler
. fix the check for the return status of the MCT configure command
models/atm/cam/bld/Makefile.in
. fix the rules for building test_fc and test_nc by removing the
specification of $(LDFLAGS). Now that we build MCT as a separate lib,
these rules were failing because LDFLAGS contains -lmct which doesn't
exist yet. Similar fixes for test_mpi and test_esmf. Had to add
$(FFLAGS) to the link commands to pick up the OpenMP flags that
are in the $(LDFLAGS) macro which was removed.
models/atm/cam/src/physics/cam/microp_aero.F90
. add addfld/outfld call for LCLOUD (moved from ndrop). Modify long_name
to 'Liquid cloud fraction used in stratus activation'
models/atm/cam/src/physics/cam/ndrop.F90
. move addfld call for LCLOUD to microp_aero.F90 and remove the add_default
and outfld calls.
models/atm/cam/src/physics/cam/macrop_driver.F90
. add addfld/outfld calls for AST, LIQCLDF, ICECLDF
models/atm/cam/src/physics/cam/micro_mg_cam.F90
. remove addfld/outfld calls for AST, LIQCLDF, ICECLDF (move to
macrop_driver)
models/atm/cam/src/physics/cam/rrtmg/radiation.F90
. fix addfld call for EMIS. change number of levels from 1 to plev, and
changed default averaging from 'I' to 'A'
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS
frankfurt/lf95: All PASS
frankfurt/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_2_00
Originator(s): fischer eaton
Date: Tue Oct 9 13:36:42 MDT 2012
One-line Summary: namelist updates, bug fixes
Purpose of changes:
. In cam_diagnostics.F90, the default diagnostic that dqcond is used to
compute is only written to the history file for 1 constituent (specific
humidity). The memory for the other pcnst-1 constituents is allocated and
set but not used. In the case where the diagnostic is written for all
pcnst constituents the data structure was modified to avoid index addresses
that don't fit into 32 bit integers.
. cam's configure script needed to be fixed to pass $mpi_inc and $mpi_lib to the
MCT configure script, when mpi_inc and mpi_lib are passed in from the configure
argument list.
. update svn externals to match cesm1_1_alpha18g.
. Namelist updates for waccmx, sd-waccm and homme.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
. cam configure was updated to pass mpi_lib and mpi_inc, being
set by arguments to cam's configure, to MCT configure
Describe any changes made to the namelist:
. set dust_emis_fact to .55 for homme
. set so2_emis_file for sd_waccm use_cases
. clm_rtm_inparm group moved to clm_inparm
. sim_year for waccmx_1996_cam4 changed to 2000
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: fischer
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/test_driver.sh
. intel support on frankfurt
M models/atm/cam/test/system/CAM_runcmnd.sh
. use mvapich mpirun for franfurt intel
M models/atm/cam/bld/configure
. fix MCT configure to allow mpi_lib and mpi_inc to be set with argument to cam configure
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. set dust_emis_fact to .55 for homme
M models/atm/cam/bld/namelist_files/namelist_definition.xml
. moved clm_rtm_inparm group to clm_inparm
M models/atm/cam/bld/namelist_files/use_cases/sd_waccm_geos5.xml
M models/atm/cam/bld/namelist_files/use_cases/sd_waccm_tslt_geos5.xml
. set so2_emis_file
M models/atm/cam/bld/namelist_files/use_cases/waccmx_1996_cam4.xml
. Changed sim_year from 1996 to 2000
M models/atm/cam/src/physics/cam/cam_diagnostics.F90
. Data allocation fixed to avoid indexing addresses that don't fit into
32-bit integers.
. Avoid allocating memory that's not used.
M models/atm/cam/src/physics/cam/clubb_intr.F90
. Changes to appease the intel compiler.
M SVN_EXTERNAL_DIRECTORIES
. Changed externals to match cesm1_1_alpha18g
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: all passed
frankfurt/lf95: all passed
frankfurt/pgi: all passed
===============================================================
===============================================================
Tag name: cam5_1_44
Originator(s): fischer, hannay, eaton
Date: Mon Oct 1 10:29:58 MDT 2012
One-line Summary: SE/cam5 tuning mods; mods for SE in standalone mode; some code cleanup.
Purpose of changes:
. Add tuning mods for SE/cam5 configurations. Also add "tuned" topo
dataset for ne30np4.
. Mods to CAM standalone build to support running SE in F compset
configurations (i.e., active land, thermodynamic cice, and data ocean).
Up to now the standalone build has only worked for SE running in
adiabatic, ideal physics, or aquaplanet configurations.
. Some cleanup in the dynamics initialization layer. In particular remove
a bunch of dead code from the SE dynamics interfaces.
. The attempt to access the initial file from the dyn_init routine created
a circular dependency. Accessing the initial file in the dyn_comp module
requires it to use startup_initialconds. But startup_initialconds
depends on dyn_comp because of the initial_conds method. Resolve this by
separating the code that opens the initial and topo files (and is not
dycore dependent) from the initial_conds routine which reads the initial
file and is dycore dependent.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system:
. support for SE standalone scripts
Describe any changes made to the namelist:
. add some tuning parameters for SE
List any changes to the defaults for the boundary datasets:
. new default for ne16np4 and ne30np4 topo datasets
. new default for ne16np4 soil_erod dataset
. new defaults for SE grids for fsurdat, fatmlndfrc, bndtvs, focndomain
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
models/atm/cam/src/control/binary_io.F90
. remove unused file
models/atm/cam/src/dynamics/homme/commap.F90
. eliminate this file. move module data to dyn_grid
List all subroutines added and what they do:
models/atm/cam/src/control/cam_initfiles.F90
. new file that contains just the routines from startup_initialconds that
were responsible for open/close of initial and topo files, and provide
accessor functions for the filehandles.
- rename setup_initial -> cam_initfiles_open
- rename close_initial_file -> cam_initfiles_close
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/build-namelist
. add bndtvs_domain to docn stream txt file
. remove unused code for csim4
. add defaults for nu_p, nu_div, hypervis_subcycle, hypervis_subcycle_q
models/atm/cam/bld/configure
. mod to run mct configure in quiet mode unless cam configure is in verbose
mode
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. add defaults for SE/cam5
5.0
1.0e15
0.0
1.0e15
25.e14
2
1
-1
1
. add ne16np4 topo dataset
. update ne30np4 topo dataset
. add ne16np4 soil_erod dataset
. add se datasets (ne16,30,60,120,240) for fsurdat for 1850 and 2000
. add se datasets (ne16,30,60,120,240) for fatmlndfrc
. add se datasets (use 1x1 HadOIBl) for bndtvs for 1850, 2000, and
1850-2000
. add se dataset for bndtvs_domain
. add se datasets for focndomain
models/atm/cam/bld/namelist_files/namelist_definition.xml
. add bndtvs_domain
. add nu_p, nu_div, hypervis_subcycle, hypervis_subcycle_q
models/atm/cam/src/control/startup_initialconds.F90
. leave behind just the subroutine initial_conds. This code should
eventually move to the dycore interfaces.
models/atm/cam/src/dynamics/homme/dyn_comp.F90
. move dynamics initialization code out of cam_initial and into dyn_init1
models/atm/cam/src/dynamics/homme/dyn_grid.F90
. add coordinate/weights arrays from commap
. remove refs to commap
models/atm/cam/src/dynamics/homme/interp_mod.F90
. move reference to w from commap to dyn_grid
models/atm/cam/src/dynamics/homme/inital.F90
. move dynamics initialization stuff out of cam_initial and into dyn_init1
models/atm/cam/src/dynamics/homme/initcom.F90
. Tried to remove this file but initcom is called from cam_read_restart.
That should be moved to a dycore specific spot since initcom is dycore
specific. For now just moved the contents to dyn_init1 and left a stub.
models/atm/cam/src/dynamics/homme/spmd_dyn.F90
. remove unused routines spmd_dyn_defaultopts, spmd_dyn_setopts,
spmdinit_dyn, compute_gsfactors
. remove unused module data
models/atm/cam/src/dynamics/sld/restart_dynamics.F90
. remove unused reference to binary_io module
models/atm/cam/src/dynamics/eul/inital.F90
models/atm/cam/src/dynamics/fv/inital.F90
models/atm/cam/src/dynamics/homme/inital.F90
models/atm/cam/src/dynamics/sld/inital.F90
. remove call to setup_initial (which has been renamed cam_initfiles_open)
-- move it to cam_comp::cam_init. Opening files is not dycore specific
and shouldn't be buried in dycore specific code.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
072 bl731 TBL.sh h16c3aqdh aqua+aquaplanet_cam3 9s ........................FAIL! rc= 7 at Sun Sep 30 18:09:05 MDT 2012
075 bl751 TBL.sh h30c4aqdm outfrq3s+aquaplanet_cam4 9s ....................FAIL! rc= 7 at Sun Sep 30 18:27:18 MDT 2012
084 bl993 TBL_ccsm.sh ne30_g16 FC5 2d .....................................FAIL! rc= 7 at Mon Oct 1 08:58:46 MDT 2012
frankfurt/lf95: All PASS
frankfurt/pgi: All PASS except:
059 bl734 TBL.sh hn16c5aqdm outfrq3s_bam+aquaplanet_cam5 9s ...............FAIL! rc= 7 at Sat Sep 29 20:57:00 MDT 2012
062 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ....................FAIL! rc= 7 at Sat Sep 29 22:11:40 MDT 2012
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB except for SE configurations due to
new datasets and tuning parameters.
===============================================================
===============================================================
Tag name: cam5_1_43
Originator(s): santos, joemci, bardeenc
Date: Sept 27, 2012
One-line Summary: WACCM-X use_case, CARMA dust, RTM in standalone builds
Purpose of changes:
- Add use_case for WACCM-X solar minimum conditions.
- Add CARMA dust model.
- Allow CAM standalone builds to build RTM and new CLM.
- Add rof externals.
- Update CARMA external (bug fix).
- Update other externals (to alpha18e or higher).
- Clean up CARMA models.
- Clean up residual issues after CLUBB commit.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
- Add build of ROF in standalone CAM.
Describe any changes made to the namelist:
- Add options for CARMA dust model.
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
A models/atm/cam/bld/namelist_files/use_cases/waccmx_1996_cam4.xml
- use_case containing solar minimum (1996) conditions for WACCM-X.
A models/atm/cam/src/physics/carma/models/dust/
A models/atm/cam/src/physics/carma/models/dust/carma_model_flags_mod.F90
A models/atm/cam/src/physics/carma/models/dust/carma_model_mod.F90
M models/atm/cam/test/system/config_files/f1.9c5carmdusm
- New CARMA dust model.
List all existing files that have been modified, and describe the changes:
M SVN_EXTERNAL_DIRECTORIES
- Update externals, add rof.
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
- Update CARMA external with bug fixes.
M models/atm/cam/bld/build-namelist
M models/atm/cam/bld/config_files/definition.xml
M models/atm/cam/bld/configure
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- Add rof for CAM standalone builds.
- Add CARMA dust model options.
M models/atm/cam/src/control/camsrfexch.F90
M models/atm/cam/src/cpl_esmf/atm_comp_esmf.F90
M models/atm/cam/src/cpl_mct/atm_comp_mct.F90
M models/atm/cam/src/cpl_share/cam_cpl_indices.F90
M models/atm/cam/src/physics/cam/comsrf.F90
- Add volumetric soil water (soilw) to coupling from CLM, for CARMA dust.
M models/atm/cam/src/physics/cam/clubb_intr.F90
M models/atm/cam/src/physics/cam/convect_shallow.F90
M models/atm/cam/src/physics/cam/microp_aero.F90
- Bug fixes and cleanup from previous tag (CLUBB commit).
M models/atm/cam/src/physics/carma/models/sea_salt/carma_model_mod.F90
- Use gamma from gamma_function, move igamma to gamma_function.
M models/atm/cam/src/utils/gamma_function.F90
- Update gamma function and make it pure.
- Cleanup older code (e.g. remove goto statements).
- Add "igamma", the upper incomplete gamma function.
M models/atm/cam/test/system/input_tests_master
M models/atm/cam/test/system/tests_carma
M models/atm/cam/test/system/tests_posttag_bluefire
- Add tests for CARMA dust, new WACCM-X use_case.
- Currently testing solar min in posttag, rather than solar max.
- Change E compset to ETEST, for new docn/scripts compatibility.
M models/atm/cam/test/system/TCB_ccsm.sh
- Add rof to build.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
- All cases with chemistry impacted by dust fluxes have answer changes.
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Sat Sep 29 13:14:57 MDT 2012
021 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...............................FAIL! rc= 7 at Sat Sep 29 13:15:10 MDT 2012
036 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ......................................FAIL! rc= 7 at Sat Sep 29 13:15:19 MDT 2012
040 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .....................................FAIL! rc= 7 at Sat Sep 29 13:15:43 MDT 2012
045 bl373 TBL.sh fm1.9c4dh outfrq3s 9s ....................................FAIL! rc= 7 at Sat Sep 29 13:16:26 MDT 2012
047 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...........................FAIL! rc= 7 at Sat Sep 29 13:16:44 MDT 2012
049 bl376 TBL.sh fm1.9c4dh+testmech outfrq3s 9s ...........................FAIL! rc= 7 at Sat Sep 29 13:17:25 MDT 2012
054 bl415 TBL.sh fst1.9c5dh outfrq3s+cam5_trop_strat_chem 9s ..............FAIL! rc= 7 at Sat Sep 29 13:18:42 MDT 2012
056 bl430 TBL.sh fm1.9c5dh outfrq3s 9s ....................................FAIL! rc= 7 at Sat Sep 29 13:19:37 MDT 2012
059 bl440 TBL.sh fsoa1.9c4dh outfrq3s+mozart_megan_emis 9s ................FAIL! rc= 7 at Sat Sep 29 13:20:22 MDT 2012
081 bl992 TBL_ccsm.sh f19_g16 ETEST 2d ....................................FAIL! rc= 5 at Sat Sep 29 13:20:29 MDT 2012
084 bl993 TBL_ccsm.sh ne30_g16 FC5 2d .....................................FAIL! rc= 7 at Sat Sep 29 15:09:30 MDT 2012
frankfurt/lf95:
- All cases with chemistry impacted by dust fluxes have answer changes.
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Sat Sep 29 13:49:45 MDT 2012
009 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Sat Sep 29 13:49:51 MDT 2012
013 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Sat Sep 29 13:49:58 MDT 2012
017 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Sat Sep 29 13:50:05 MDT 2012
025 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Sat Sep 29 13:51:43 MDT 2012
frankfurt/pgi or jaguar/pgi:
- All cases with chemistry impacted by dust fluxes have answer changes.
005 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Fri Sep 28 10:44:50 MDT 2012
010 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Fri Sep 28 11:17:22 MDT 2012
016 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Fri Sep 28 11:51:00 MDT 2012
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Fri Sep 28 12:45:56 MDT 2012
031 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Fri Sep 28 13:57:40 MDT 2012
044 bl317 TBL.sh f10c5cdm outfrq3s 9s .....................................FAIL! rc= 7 at Fri Sep 28 16:05:53 MDT 2012
053 bl321 TBL.sh f10c5cdm atrain 9s .......................................FAIL! rc= 7 at Fri Sep 28 16:42:32 MDT 2012
056 bl420 TBL.sh f7mode4c5dm outfrq3s 9s ..................................FAIL! rc= 7 at Fri Sep 28 17:29:37 MDT 2012
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- Most active chemistry and CARMA cases.
- what platforms/compilers:
- All
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
- Larger than roundoff, same climate.
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_1_42
Originator(s): cacraig, bogenschutz, eaton, santos, edwards
Date: 09/26/12
One-line Summary: CLUBB and two minor changes for trunk
Purpose of changes:
- Introduction of CLUBB onto CAM trunk. CLUBB (Cloud Layers Unified by Bi-normals) is a
higher-order turbulence closure centered around an assumed double Guassian PDF. It is
a unified parameterization in that it returns tendencies due to PBL, shallow convection,
and cloud macrophysical processes. This unified moist turbulence parameterization
uses multivariate PDFs to predict moments to generate shallow and stratiform clouds
that are then passed to the microphysics.
Reference: Bogenschutz, P. A., Gettelman, A., Morrison, H., Larson, V. E., Schanen, D. P.,
Meyer, N. R., and Craig, C.: Unified parameterization of the planetary boundary layer and
shallow convection with a higher-order turbulence closure in the community atmosphere
model: single column experiments, Geosci. Model Dev. Discuss., 5, 1743-1780, doi:10.5194/gmdd-5-1743-2012
- Removed CONCLDQL from pbuf as it was only written and never read
- bug fix to sat_hist (edwards)
- addition of carma_inparm namelist (santos)
Bugs fixed (include bugzilla ID):
1547: CAM dies trying to writ out N-dimensional fields to the "satellite" history file
- fixed in sat_hist.F90
Describe any changes made to build system:
- Introduced clubb_sgs flag as a configure option to turn on CLUBB
Describe any changes made to the namelist:
- Introduced in phys_ctl_nl
macrop_scheme - character, valid values are "rk, park, clubb_sgs, none"
do_clubb_sgs - logical, if true, turns on CLUBB_SGS
- Added option CLUBB_SGS to shallow_scheme and eddy_scheme
- Introduced in cam_inparm
scm_clubb_iop_name - character, IOP name for CLUBB running in single column mode
- Introduced clubb namelist "clubb_his_nl" with variables:
clubb_history - logical, if true, output CLUBB history statistics
clubb_rad_history - logical, if true, output CLUBBs radiative history statistics
- Introduced clubb namelist "clubb_stats_nl" with variables:
clubb_vars_zt - character, same as fincl1, but for CLUBB statistics on zt grid
clubb_vars_zm - character, same as fincl1, but for CLUBB statistics on zm grid
clubb_vars_rad_zt - character, same as fincl1, but for CLUBB statistics on radiation zt grid
clubb_vars_rad_zm - character, same as fincl1, but for CLUBB statistics on radiation zm grid
clubb_vars_src - character, same as fincl1, but for CLUBB statistics on surface
clubb_iop_name - character, name of the IOP case for adjustments within CLUBB
- Moved from vert_diff_nl to phys_ctl_nl
do_tms
- Moved from vert_diff_nl to physconst_nl
tms_orocnst, tms_z0fac
- Introduced carma namelist "carma_inparm" with variables:
carma_fields - character, list of fluxes needed by the CARMA model from CLM to CAM
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do: none
A + models/atm/cam/src/physics/cam/clubb_intr.F90
- Interface module between CLUBB and CAM
A + models/atm/cam/src/physics/CLUBB_core
A + models/atm/cam/src/physics/CLUBB_core/constants_clubb.F90
A + models/atm/cam/src/physics/CLUBB_core/variables_prognostic_module.F90
A + models/atm/cam/src/physics/CLUBB_core/interpolation.F90
A + models/atm/cam/src/physics/CLUBB_core/clip_explicit.F90
A + models/atm/cam/src/physics/CLUBB_core/saturation.F90
A + models/atm/cam/src/physics/CLUBB_core/mono_flux_limiter.F90
A + models/atm/cam/src/physics/CLUBB_core/mixing_length.F90
A + models/atm/cam/src/physics/CLUBB_core/stats_zm.F90
A + models/atm/cam/src/physics/CLUBB_core/file_functions.F90
A + models/atm/cam/src/physics/CLUBB_core/mean_adv.F90
A + models/atm/cam/src/physics/CLUBB_core/grid_class.F90
A + models/atm/cam/src/physics/CLUBB_core/pdf_closure_module.F90
A + models/atm/cam/src/physics/CLUBB_core/stats_zt.F90
A + models/atm/cam/src/physics/CLUBB_core/parameter_indices.F90
A + models/atm/cam/src/physics/CLUBB_core/hyper_diffusion_4th_ord.F90
A + models/atm/cam/src/physics/CLUBB_core/csr_matrix_class_3array.F90
A + models/atm/cam/src/physics/CLUBB_core/input_reader.F90
A + models/atm/cam/src/physics/CLUBB_core/parameters_model.F90
A + models/atm/cam/src/physics/CLUBB_core/error_code.F90
A + models/atm/cam/src/physics/CLUBB_core/lapack_wrap.F90
A + models/atm/cam/src/physics/CLUBB_core/stats_rad_zm.F90
A + models/atm/cam/src/physics/CLUBB_core/parameters_radiation.F90
A + models/atm/cam/src/physics/CLUBB_core/mt95.f90
A + models/atm/cam/src/physics/CLUBB_core/endian.F90
A + models/atm/cam/src/physics/CLUBB_core/recl.inc
A + models/atm/cam/src/physics/CLUBB_core/parameters_microphys.F90
A + models/atm/cam/src/physics/CLUBB_core/gmres_cache.F90
A + models/atm/cam/src/physics/CLUBB_core/stats_rad_zt.F90
A + models/atm/cam/src/physics/CLUBB_core/sigma_sqd_w_module.F90
A + models/atm/cam/src/physics/CLUBB_core/hydrostatic_module.F90
A + models/atm/cam/src/physics/CLUBB_core/array_index.F90
A + models/atm/cam/src/physics/CLUBB_core/anl_erf.F90
A + models/atm/cam/src/physics/CLUBB_core/T_in_K_module.F90
A + models/atm/cam/src/physics/CLUBB_core/calendar.F90
A + models/atm/cam/src/physics/CLUBB_core/model_flags.F90
A + models/atm/cam/src/physics/CLUBB_core/advance_windm_edsclrm_module.F90
A + models/atm/cam/src/physics/CLUBB_core/pos_definite_module.F90
A + models/atm/cam/src/physics/CLUBB_core/pdf_parameter_module.F90
A + models/atm/cam/src/physics/CLUBB_core/fill_holes.F90
A + models/atm/cam/src/physics/CLUBB_core/sponge_layer_damping.F90
A + models/atm/cam/src/physics/CLUBB_core/stats_variables.F90
A + models/atm/cam/src/physics/CLUBB_core/stats_subs.F90
A + models/atm/cam/src/physics/CLUBB_core/variables_radiation_module.F90
A + models/atm/cam/src/physics/CLUBB_core/numerical_check.F90
A + models/atm/cam/src/physics/CLUBB_core/stat_file_module.F90
A + models/atm/cam/src/physics/CLUBB_core/input_names.F90
A + models/atm/cam/src/physics/CLUBB_core/output_netcdf.F90
A + models/atm/cam/src/physics/CLUBB_core/diffusion.F90
A + models/atm/cam/src/physics/CLUBB_core/advance_wp2_wp3_module.F90
A + models/atm/cam/src/physics/CLUBB_core/advance_xm_wpxp_module.F90
A + models/atm/cam/src/physics/CLUBB_core/extrapolation.F90
A + models/atm/cam/src/physics/CLUBB_core/stats_sfc.F90
A + models/atm/cam/src/physics/CLUBB_core/surface_varnce_module.F90
A + models/atm/cam/src/physics/CLUBB_core/clip_semi_implicit.F90
A + models/atm/cam/src/physics/CLUBB_core/clubb_core.F90
A + models/atm/cam/src/physics/CLUBB_core/stats_type.F90
A + models/atm/cam/src/physics/CLUBB_core/clubb_precision.F90
A + models/atm/cam/src/physics/CLUBB_core/advance_helper_module.F90
A + models/atm/cam/src/physics/CLUBB_core/gmres_wrap.F90
A + models/atm/cam/src/physics/CLUBB_core/Skw_module.F90
A + models/atm/cam/src/physics/CLUBB_core/parameters_tunable.F90
A + models/atm/cam/src/physics/CLUBB_core/advance_xp2_xpyp_module.F90
A + models/atm/cam/src/physics/CLUBB_core/output_grads.F90
A + models/atm/cam/src/physics/CLUBB_core/variables_diagnostic_module.F90
- CLUBB code
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/configure
M models/atm/cam/bld/config_files/definition.xml
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
M models/atm/cam/bld/namelist_files/namelist_definition.xml
MM models/atm/cam/bld/build-namelist
- namelist changes described above
M models/atm/cam/doc/ChangeLog
M models/atm/cam/src/control/runtime_opts.F90
M models/atm/cam/src/control/scamMod.F90
- introduction of scm_clubb_iop_name
M models/atm/cam/src/control/physconst.F90
- physconst_nl now has tms_orocnst and tms_z0fac
M models/atm/cam/src/control/sat_hist.F90
- bug fix described above
M models/atm/cam/src/dynamics/eul/iop.F90
- shflx and sh are now mutually exclusive
M models/atm/cam/src/physics/cam/cam_diagnostics.F90
- Modifications to allow for QCWAT, TCWAT, ICWAT
and qpert possibly not being in pbuf
M models/atm/cam/src/physics/cam/cldwat2m_macro.F90
- aist_vector is now public
M models/atm/cam/src/physics/cam/convect_deep.F90
- modified deep_scheme_does_scav_trans logic to make
it more flexible
- jctop and jcbot are now initialized to non-zero values
M models/atm/cam/src/physics/cam/convect_shallow.F90
- moved qpert as it just exists in Hack scheme
- cnt2 and cnb2 are now initialized to non-zero values
- added CLUBB_SGS to possible cases for shallow_scheme
M models/atm/cam/src/physics/cam/hb_diff.F90
- pblintd is now public
M models/atm/cam/src/physics/cam/macrop_driver.F90
M + models/atm/cam/src/physics/cam/micro_mg_cam.F90
M + models/atm/cam/src/physics/cam/microp_driver.F90
- bug fix- energy conservation was giving off false alarms - non
answer changing. Removed rliq from microp to macro to be
considered for energy conservation
M models/atm/cam/src/physics/cam/microp_aero.F90
- Introduced logic for CLUBB_SGS
M models/atm/cam/src/physics/cam/phys_control.F90
- Added to phys_ctl_nl: macrop_scheme, do_clubb_sgs and do_tms
M models/atm/cam/src/physics/cam/physpkg.F90
- Addition of CLUBB_SGS logic
- Modifications to allow for QCWAT, TCWAT, ICWAT
and qpert possibly not being in pbuf
- bug fix for energy conservation (see macrop_driver above for more details)
M models/atm/cam/src/physics/cam/vertical_diffusion.F90
- Moved do_tms to phys_ctl_nl
- Moved tms_orocnst and tms_z0fac to physconst_nl
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS
frankfurt/lf95: All PASS
frankfurt/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: All BFB
===============================================================
===============================================================
Tag name: cam5_1_41
Originator(s): cacraig, eaton
Date: Mon Sep 24 13:57:57 MDT 2012
One-line Summary: cleanup in stratiform, macrop, microp modules; new namelist
Purpose of changes:
. Remove lots of unused code left over from the original cam5 development.
This was a consequence of the cam5 macro/micro physics being developed in
the stratiform module, and then later moved to separate macrop and
micro_mg modules. The result was code that's only needed for the cam5
physics was left in the stratiform module, and code that's only needed for
cam4 physics was left in the macrop and micro_mg modules.
. Move several pbuf_add calls to the modules that are responsible for
setting the pbuf field.
. Add namelist group microp_aero_nl to the microp_aero module. Currently
just contains the variable microp_aero_bulk_scale which is a tuning
parameter for using bulk aerosols with the cam5 physics package.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: none
Describe any changes made to the namelist:
. Add namelist group microp_aero_nl to the microp_aero module. Currently
just contains the variable microp_aero_bulk_scale which is a tuning
parameter for using bulk aerosols with the cam5 physics package.
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
models/atm/cam/src/physics/cam/mcshallow.F90
. not used
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/Makefile.in
. occasionally the serial builds were failing due to trying to compile the
pio_kinds.F90 before the mct build was finished. This fails because in
the serial build the mpi-serial lib is part of mct, and that build puts
the mpif.h file in the appropriate include directory. For now have added
a rule to make pio_kind.o depend on libmct.a.
models/atm/cam/bld/namelist_files/namelist_definition.xml
. add microp_aero_bulk_scale
models/atm/cam/src/control/runtime_opts.F90
. add call for microp_aero_readnl
models/atm/cam/src/physics/cam/cloud_fraction.F90
. add cldfrc_register for SH_FRAC, DP_FRAC
models/atm/cam/src/physics/cam/conv_water.F90
. remove refs to pbuf fields ALST, AIST, QLST, QIST -- not used
. remove ref to CONCLDQL -- not used
models/atm/cam/src/physics/cam/eddy_diff.F90
. fix logfile info that was written from all procs -- only write from
masterproc
. replace 'stop' statements by calls to endrun
models/atm/cam/src/physics/cam/macrop_driver.F90
. Remove RHDFDA and RHU00 from pbuf and local vars. Only used in cam4
physics.
. Remove the second call to cldfrc which was only used for the rhdfda
calculation.
. remove addfld/outfld for CNVCLD -- no longer used
. remove calls to add SH_FRAC, DP_FRAC to pbuf. Now done in
cldfrc_register.
. remove pbuf_add calls for TINI, QINI, CLDLIQINI, CLDICEINI. Now done in
phys_register.
. remove access of QME -- not used
. remove pbuf_add for CC_* variables -- moved to micro_mg_cam. Replace
with pbuf_get_index calls for those variables.
models/atm/cam/src/physics/cam/micro_mg_cam.F90
. remove REL2, REI2, DEICONV, MUCONV, LAMBDACONV from pbuf. Not used.
. add initializer for CLDO.
. add pbuf_add for CC_* variables; also add initializers.
models/atm/cam/src/physics/cam/microp_aero.F90
. add microp_aero_readnl, and microp_aero_bulk_scale for tuning the
activation by bulk aerosol.
models/atm/cam/src/physics/cam/physpkg.F90
. call cldfrc_register from phys_register
. add pbuf_add calls for TINI, QINI, CLDLIQINI, CLDICEINI (used for physics
package diagnostics).
. move initializer for CLDO to microp_mg_cam
models/atm/cam/src/physics/cam/stratiform.F90
. remove micro_treatment parameter which isn't relevent for cam4
. Remove RHDFDA and RHU00 from pbuf and use local variables only.
rhu00 is computed in cldfrc, then used to compute rhdfda, and is
intent(in) to pcond.
rhdfda is computed then used as intent(in) arg to pcond. There is no
use of saved values.
. remove many unused local variables
. remove AIST,ALST from pbuf. Replace local use by ast.
. remove LIQCLDF, ICECLDF from hist. They are identical to AST.
. remove CLDO, CONCLD_QL, ICCWAT, REL_FN, DEICONV, MUCONV, LAMBDACONV
from pbuf. Not used.
. remove pbuf_add calls for SH_FRAC, DP_FRAC. Now done in
cldfrc_register.
. remove pbuf_add calls for TINI, QINI, CLDLIQINI, CLDICEINI. Now done in
phys_register.
. replace local var fice by pbuf FICE. get rid of fice_ql pointer
models/atm/cam/src/physics/carma/cam/carma_intr.F90
. remove pbuf_times dimension from CG*, CI*, CL*, CT
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS
frankfurt/lf95: All PASS
frankfurt/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: All BFB
===============================================================
===============================================================
Tag name: cam5_1_40
Originator(s): fischer
Date: Fri Sep 20 2012
One-line Summary: update SVN exteranals, build MCT library separately
Purpose of changes:
. There are a couple of changes with the cesm build scripts with the new
scripts tag.
. The command line command "configure -case" has been replace with "setup"
. env_configure.xml has been renamed env_pesetup.xml
. Update SVN externals to match cesm1_1_beta17, except for clm4_0_47
and pio1_5_4. These are later tags that are a bug fix for cam and
a fix for lahey
. With the new tag for MCT, the MCT configure script must be run. So
configure and Makefile.in were updated to handle these changes. A new
varaible mct_libdir was added to be able to set the location of the mct
library.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
. MCT is now built as a stand alone library. In order to build mct, changes
where made to configure and Makefile.in.
.FC, FC_TYPE, and cc are now determined in configure instead of Makefile.in.
.cam's configure script creates a directory structure in $mct_libdir and copies the MCT
makefiles over.
.MCT's configure script is called in order to set up MCT's Makefile.conf.
.CAM's Makefile.in has new rules to build the MCT library.
.$mct_libdir default is $cam_bld/mct, can be set on the command line -mct_libdir, or
enviroment variable MCT_LIBDIR.
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:eaton
List all subroutines eliminated:
List all subroutines added and what they do:
M models/atm/cam/test/system/TCB_ccsm.sh
. updated test script to reflect changes with the cesm build system
M models/atm/cam/test/system/tests_pretag_frankfurt_lahey
. removed 113 and 115 tests
M models/atm/cam/bld/configure
. Added code to create directory structure needed to build MCT.
. Moved logic for determining fc and fc_type from Makefile.in to configure.
This was done so the new call to the MCT configure script can be passed
the compiler type.
M models/atm/cam/bld/Makefile.in
. removed logic for determining fc and fc_type.
. Added make rules for MCT library build.
M models/atm/cam/bld/config_files/definition.xml
. add mct_libdir, and add xlf to fc_type defaults
M SVN_EXTERNAL_DIRECTORIES
. Update SVN externals to match cesm1_1_beta17, except for clm4_0_47
and pio1_5_4. The clm tag is a bug fix to allow cam builds using
cesm scripts. The pio tag is a fix to allow lahey to work.
List all existing files that have been modified, and describe the changes:
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: all passed
frankfurt/lf95: all passed
frankfurt/pgi: all passed
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,: all bfb
===============================================================
===============================================================
Tag name: cam5_1_39
Originator(s): fvitt
Date: 18 Sep 2012
One-line Summary: MAM7 corrections
Purpose of changes:
Correction to dust density weighting factors in microp_aero when
7-mode modal aerosol model is used.
Correction to the aerosol optics inputs for MAM7.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
D models/atm/cam/test/system/tests_posttag_edinburgh
D models/atm/cam/test/system/tests_pretag_edinburgh_lahey
D models/atm/cam/test/system/tests_pretag_edinburgh_pgi
- edinburgh is replaced by frankfurt
List all subroutines added and what they do:
A models/atm/cam/test/system/tests_pretag_frankfurt_pgi
A models/atm/cam/test/system/tests_posttag_frankfurt
A models/atm/cam/test/system/tests_pretag_frankfurt_lahey
- edinburgh is replaced by frankfurt
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/test_driver.sh
M models/atm/cam/test/system/CAM_runcmnd.sh
- edinburgh is replaced by frankfurt
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- correction to the aerosol optics inputs for MAM7
M models/atm/cam/src/physics/cam/microp_aero.F90
- corrections weighting factors of dust densities for MAM7
M models/atm/cam/src/chemistry/bulk_aero/progseasalts_intr.F90
- sea salt emission scaling factor adjusted for MAM7
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS
frankfurt/lf95: All PASS
frankfurt/pgi or jaguar/pgi:
056 bl420 TBL.sh f7mode4c5dm outfrq3s 9s ..................................FAIL! rc= 7 at Mon Sep 17 15:04:53 MDT 2012
- this is expected due to changes for MAM7
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_1_38
Originator(s): mmills, santos
Date: September 10, 2012
One-line Summary: Merge in WACCM5 development
Purpose of changes:
- Add first WACCM5 chemistry package: waccm_mozart_mam3.
- Add namelist settings and use_cases for WACCM5
- Add features for WACCM5 compatibility.
- Troposphere cloud physics "top" added to more modules.
- Fix CARMA coordinate bug.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
- New configuration file, namelist defaults, and use_cases for WACCM5.
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
A models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/chem_mech.dat
A models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/chem_mech.doc
A models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/chem_mech.in
A models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/chem_mods.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/m_rxt_id.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/m_spc_id.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_adjrxt.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_imp_sol.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_indprd.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_lin_matrix.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_lu_factor.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_lu_solve.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_nln_matrix.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_phtadj.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_prod_loss.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_rxt_rates_conv.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_setrxt.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_mam3/mo_sim_dat.F90
- Add new waccm_mozart_mam3 chemistry package.
A models/atm/cam/test/system/config_files/wm1.9c5h
- Configuration file for WACCM5 tests.
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/build-namelist
A models/atm/cam/bld/config_files/defaults_waccm5.xml
M models/atm/cam/bld/config_files/definition.xml
M models/atm/cam/bld/configure
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1850_cam5.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2000_cam5.xml
- Default settings for WACCM5 added.
- waccm_mozart_mam3 added to configure, namelist defaults added.
M models/atm/cam/src/chemistry/modal_aero/modal_aero_coag.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_newnuc.F90
M models/atm/cam/src/chemistry/mozart/mo_usrrxt.F90
M models/atm/cam/src/physics/cam/aer_rad_props.F90
M models/atm/cam/src/physics/cam/modal_aer_opt.F90
M models/atm/cam/src/physics/cam/ndrop.F90
- Added trop_cloud_top_lev as upper boundary of certain
processes/loops.
M models/atm/cam/src/chemistry/mozart/chemistry.F90
M models/atm/cam/src/chemistry/mozart/mo_chemini.F90
M models/atm/cam/src/physics/cam/cam_diagnostics.F90
M models/atm/cam/src/physics/cam/phys_control.F90
- Added waccm_mozart_mam3 chemistry package to options/conditionals.
M models/atm/cam/src/physics/cam/microp_aero.F90
- Initialize entire lcldn/lcldo arrays to 0._r8, rather than
only initializing below top_lev.
M models/atm/cam/src/physics/carma/cam/carma_intr.F90
- Fixed bug in CARMA interface where dlat was being set
instead of dy.
M models/atm/cam/test/system/input_tests_master
M models/atm/cam/test/system/tests_posttag_bluefire
- Added short WACCM5 tests to bluefire posttag.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS.
edinburgh/lf95: All PASS
edinburgh/pgi or jaguar/pgi: All PASS
The following non-pretag tests, however, are expected to fail.
- WACCM5 posttag baseline test (bl380) will fail, because this is a new
feature and chemistry package. However, sm380 and er380 succeed.
- CARMA baseline tests expected to fail due to bug fix.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- CARMA only, due to bug fix.
- what platforms/compilers: All
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
- Larger than roundoff CARMA, bit-for-bit all other cases, anticipate
same climate.
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_1_37
Originator(s): eaton
Date: Wed Sep 5 08:26:01 MDT 2012
One-line Summary: Remove hardcoded zvir from cam4 physics.
Purpose of changes:
. Call the init routine for the pbl_utils from phys_init, not from
init_eddy_diff and init_hb_diff. This wasn't done previously because a
local value of zvir (0.61) was being used in the hb_diff module. Using a
single call to the init routine which uses the shared value of constants
will change answers (but not climate) for cam3/cam4 physics.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/src/physics/eddy_diff.F90
. remove call to init_pbl
models/atm/cam/src/physics/hb_diff.F90
. remove call to init_pbl
. remove zvir_loc, zvir, rair. they were only used in init_pbl call which has
been moved. Adjust init_hb_diff args accordingly.
models/atm/cam/src/physics/pbl_utils.F90
. change name init_pbl to pbl_utils_init
models/atm/cam/src/physics/physpkg.F90
. add call to pbl_utils_init from phys_init
models/atm/cam/src/physics/vertical_diffusion.F90
. modify actual args in init_hb_diff call
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
004 bl132 TBL.sh e48c4dh ghgrmp+1850_cam4 9s ..............................FAIL! rc= 7 at Fri Aug 31 14:43:43 MDT 2012
006 bl133 TBL.sh e48c4paqdm aqpgro+aquaplanet_cam4 3s .....................FAIL! rc= 7 at Fri Aug 31 14:47:20 MDT 2012
017 bl331 TBL.sh f4c4dh co2rmp+1850_cam4 9s ...............................FAIL! rc= 7 at Fri Aug 31 15:36:07 MDT 2012
023 bl333 TBL.sh f4c4paqdh aqpgro+aquaplanet_cam4 3s ......................FAIL! rc= 7 at Fri Aug 31 15:58:00 MDT 2012
033 bl336 TBL.sh wm4c4h outfrq3s+waccm_1850_cam4 9s .......................FAIL! rc= 7 at Fri Aug 31 16:35:18 MDT 2012
045 bl373 TBL.sh fm1.9c4dh outfrq3s 9s ....................................FAIL! rc= 7 at Tue Sep 4 09:25:23 MDT 2012
049 bl376 TBL.sh fm1.9c4dh+testmech outfrq3s 9s ...........................FAIL! rc= 7 at Tue Sep 4 10:31:57 MDT 2012
052 bl374 TBL.sh fs1.9c4dh outfrq3s+1850-2005_cam4_super_fast_llnl 9s .....FAIL! rc= 7 at Tue Sep 4 11:26:28 MDT 2012
059 bl440 TBL.sh fsoa1.9c4dh outfrq3s+mozart_megan_emis 9s ................FAIL! rc= 7 at Tue Sep 4 13:07:47 MDT 2012
063 bl387 TBL.sh f1.9c4dm outfrq3s+1850-2005_cam4 9s ......................FAIL! rc= 7 at Tue Sep 4 13:51:42 MDT 2012
066 bl388 TBL.sh f1.9c4m outfrq24h+1850_cam4 2d ...........................FAIL! rc= 7 at Tue Sep 4 14:04:16 MDT 2012
069 bl379 TBL.sh wm1.9c4h outfrq24h+waccm_1850_cam4 2d ....................FAIL! rc= 7 at Tue Sep 4 16:14:28 MDT 2012
072 bl731 TBL.sh h16c3aqdh aqua+aquaplanet_cam3 9s ........................FAIL! rc= 7 at Tue Sep 4 16:23:53 MDT 2012
075 bl751 TBL.sh h30c4aqdm outfrq3s+aquaplanet_cam4 9s ....................FAIL! rc= 7 at Tue Sep 4 16:47:10 MDT 2012
081 bl992 TBL_ccsm.sh f19_g16 E 2d ........................................FAIL! rc= 7 at Tue Sep 4 17:39:42 MDT 2012
edinburgh/lf95: All PASS except:
032 bl313 TBL.sh f10c3aqdm outfrq3s+aquaplanet_cam3 3s ....................FAIL! rc= 7 at Mon Sep 3 23:57:16 MDT 2012
035 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7 at Tue Sep 4 01:18:44 MDT 2012
040 bl318 TBL.sh f10c4cdm sat_hist 9s .....................................FAIL! rc= 7 at Tue Sep 4 03:18:48 MDT 2012
edinburgh/pgi: All PASS except:
033 bl313 TBL.sh f10c3aqdm outfrq3s+aquaplanet_cam3 3s ....................FAIL! rc= 7 at Fri Aug 31 15:43:49 MDT 2012
036 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7 at Fri Aug 31 15:55:16 MDT 2012
039 bl315 TBL.sh f10c3dm outfrq3s+cam3 9s .................................FAIL! rc= 7 at Fri Aug 31 16:03:06 MDT 2012
All configurations using cam3 or cam4 physics fail the baseline comparisons.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB except configs using cam3/cam4 physics
===============================================================
===============================================================
Tag name: cam5_1_36
Originator(s): fvitt
Date: 31 Aug 2012
One-line Summary: Changes for MAM7, added chemistry package, and misc bug fixes
Purpose of changes:
- changes for MAM7 (7 mode modal aerosol model):
. use different scaling factor for sea salt emissions
. adjusted the external forcings for trop_mam7 chemistry package
. adjusted the NH4 aerosol optics files
- new chemistry package trop_strat_mam7 (MOZART chemistry with MAM7 aerosols)
- add build-namelist use case for MEGAN emissions that will be used
with MOZSOA compsets
- bug fix in CO+OH reaction rate for super_fast_llnl chemistry
- bug fix in "WMO" method of tropopause calculation
- fix path of chemistry preprocessor generated source files in Filepath
so that CESM build with compile the correct source files
- added regression tests for the chem preprocessor
Bugs fixed (include bugzilla ID):
- correction to CO+OH reaction rate for super_fast_llnl chemistry
- invoke O1D_to_2OH_adj subroutine when chemistry is super_fast_llnl_mam3
- correction to "WMO" method of tropopause calculation
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
D models/atm/cam/src/chemistry/pp_waccm_mozart_sulfur/mo_usrrxt.F90
- removed duplicate code
the code relevant to waccm_mozart_sulfur moved to chemistry/mozart/mo_usrrxt.F90
List all subroutines added and what they do:
A models/atm/cam/test/system/TNE_ccsm.sh
A models/atm/cam/test/system/config_files/testmech
- for chemistry preprocessor regression tests
A models/atm/cam/test/system/config_files/fst7mode1.9c5h
A models/atm/cam/test/system/config_files/f7mode1.9c5dh
A models/atm/cam/test/system/config_files/fst7mode4c5dm
- for trop_mam7 and trop_strat_mam7 chemistry tests
A models/atm/cam/test/system/tests_chem_hybrid
A models/atm/cam/test/system/tests_chem_mpi
- for many chemistry configurations
A models/atm/cam/bld/namelist_files/use_cases/mozart_megan_emis.xml
- use case for MEGAN emission for MOZART chemistry
A models/atm/cam/src/chemistry/pp_trop_strat_mam7
A models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_lu_solve.F90
A models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_setrxt.F90
A models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_adjrxt.F90
A models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_rxt_rates_conv.F90
A models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_imp_sol.F90
A models/atm/cam/src/chemistry/pp_trop_strat_mam7/chem_mods.F90
A models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_prod_loss.F90
A models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_lin_matrix.F90
A models/atm/cam/src/chemistry/pp_trop_strat_mam7/m_rxt_id.F90
A models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_phtadj.F90
A models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_nln_matrix.F90
A models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_lu_factor.F90
A models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_sim_dat.F90
A models/atm/cam/src/chemistry/pp_trop_strat_mam7/mo_indprd.F90
A models/atm/cam/src/chemistry/pp_trop_strat_mam7/m_spc_id.F90
A models/atm/cam/src/chemistry/pp_trop_strat_mam7/chem_mech.doc
A models/atm/cam/src/chemistry/pp_trop_strat_mam7/chem_mech.in
- new chemistry package added
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/TCB.sh
M models/atm/cam/test/system/tests_pretag_bluefire
M models/atm/cam/test/system/TCB_ccsm.sh
M models/atm/cam/test/system/input_tests_master
M models/atm/cam/test/system/TSM.sh
- changes for testing chemistry preprocessor
- added tests for trop_mam7 and trop_strat_mam7 chemistry
M models/atm/cam/bld/configure
M models/atm/cam/bld/config_files/definition.xml
- fixed the path of pre-processor generated source code in Filepath
- new chem option
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- defaults for trop_strat_mam7 chemistry
- change in NH4 optics files for MAM7
- added emission files for MAM7
M models/atm/cam/bld/namelist_files/use_cases/cam5_trop_strat_chem.xml
- adjusted the lght_no_prd_factor namelist parameter
M models/atm/cam/bld/build-namelist
- MAM7 (trop_mam7 and trop_strat_mam7) default namelist
- added the ability to use "$INPUTDATA_ROOT" variable in specifying complex
namelist settings in use cases
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
- update to chemistry preprocessor to handle mechanism with up to 1000 species
M models/atm/cam/src/physics/cam/tropopause.F90
- correction to "WMO" method of tropopause calculation
M models/atm/cam/src/physics/cam/phys_control.F90
M models/atm/cam/src/physics/cam/modal_aer_opt.F90
- additional prognostic modal aerosol model (trop_strat_mam7)
M models/atm/cam/src/chemistry/bulk_aero/progseasalts_intr.F90
- use different scaling factor for sea salt emissions
M models/atm/cam/src/chemistry/pp_trop_mam7/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_mam7/chem_mech.in
- changes in external forcings
M models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
M models/atm/cam/src/chemistry/mozart/chemistry.F90
- fix for super_fast_llnl_mam3
M models/atm/cam/src/chemistry/mozart/mo_usrrxt.F90
- corrected CO+OH rate calculation for super_fast_llnl
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
052 bl374 TBL.sh fs1.9c4dh outfrq3s+1850-2005_cam4_super_fast_llnl 9s .....FAIL! rc= 7 at Fri Aug 31 00:16:50 MDT 2012
- correction to reaction rate
054 bl415 TBL.sh fst1.9c5dh outfrq3s+cam5_trop_strat_chem 9s ..............FAIL! rc= 7 at Fri Aug 31 00:18:08 MDT 2012
- change in lght_no_prd_factor namelist parameter
edinburgh/lf95: All Pass
edinburgh/pgi:
056 bl420 TBL.sh f7mode4c5dm outfrq3s 9s ..................................FAIL! rc= 7 at Thu Aug 30 16:13:18 MDT 2012
changes in trop_mam7
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_1_35
Originator(s): eaton
Date: Wed Aug 29 08:28:47 MDT 2012
One-line Summary: fix vertical_diffusion pbuf fields
Purpose of changes:
. Several of the fields put in the physics buffer by the vertical diffusion
module were using the special pbuf_times dimension incorrectly. This
only results in a bug for the Eulerian dycore (and hence SCAM) since
that's the only dycore for which pbuf_times is greater than 1.
The dimension was removed for the following fields:
- tke is in the buffer to be communicated to microp_aero. There is no
need for multiple time levels when the purpose is just to share a the
most recently computed values of a diagnostic. In fact subtle mistakes
will result when the data consuming module executes before the data
producing module in the tphysbc then tphysac sequence, and both modules
are using the pbuf_old_tim_idx to access the buffer. In the particular
case of tke the bug was that it was being accessed in microp_aero as if
it did not even have a pbuf_times dimension.
- kvh is used by other parameterizations, and as an initial guess in
compute_eddy_diff on subsequent timesteps. In the Eulerian case it
probably doesn't make sense to use values that are 2 timesteps old as
initial guesses rather than 1 timestep old values. Furthermore this
field was being accessed in microp_aero and ndrop as if it did not have
a pbuf_times dimension.
- kvm, tauresx, and tauresy are in the pbuf to be used as an initial
guesses on the next timestep.
- turbtype and smaw are not currently needed in the pbuf.
- kvt is only used by the waccm/gw_drag code. Only the information just
computed in vertical_diffusion needs to be passed (not multiple time
levels).
. Change default vertical remap subcycles for 1-deg FV to nspltvrm=2. This
is to make the 1-deg simulation more stable.
. Fix T_TTEND output diagnostic.
Bugs fixed (include bugzilla ID):
. Several physics buffer fields declared with the pbuf_times dimension were
being incorrectly accessed. This is only a bug for cam5 physics using
Eulerian dycore. Note in particular that this affects SCAM.
. The T_TTEND diagnostic had the wrong sign, and due to incorrect use of
the physics buffer was completely invalid for runs using Eulerian dycore.
Describe any changes made to build system: none
Describe any changes made to the namelist:
. Change default value of nspltvrm to 2 for 1-deg FV.
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes:
. The change of nspltvrm to 2 for 1-deg FV does add an overhead which will
be noticable (5% ?) but hasn't been thoroughly evaluated.
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. set nspltvrm=2 for 1-deg FV
models/atm/cam/src/physics/cam/cam_diagnostics.F90
. Fix access to tke, kvh, kvm (now 2D instead of 3D)
. Fix T_TTEND output diagnostic:
- T_TTEND was added to the pbuf with a pbuf_times index, but was accessing
the field ignoring that index. For the Eulerian core this implies that
on some steps the previous timestep value was accessed, and on the
following steps the value from 2 previous timesteps was accessed. When
computing tendencies for Eulerian the ztodt variable contains twice the
physics timestep, so need to always access the pbuf values from 2
timesteps previous using the pbuf_old_tim_idx function.
- In the diag_conv_tend_ini routine only initialize the T_TTEND pbuf on the
first step of an initial run, and initialize all pbuf_times levels.
- Fix the sign of the T_TTEND tendency calculation.
models/atm/cam/src/physics/cam/convect_shallow.F90
. remove time index from tke access from pbuf
models/atm/cam/src/physics/cam/diffusion_solver.F90
. make sure kvt is present when do_molec_diff=.true.
. remove initializer for kvt. It is intent(out) from compute_molec_diff.
models/atm/cam/src/physics/cam/gw_drag.F90
. remove unused dummy arg kvh
models/atm/cam/src/physics/cam/molec_diff.F90
. change kvt dummy arg to intent(out) from inout
models/atm/cam/src/physics/cam/physpkg.F90
. fix initialization of tke, kvh, kvm to use just 1 time level
. tphysac - remove access of kvt from the pbuf, and remove kvt actual arg
from call to gw_intr
models/atm/cam/src/physics/cam/stratiform.F90
. remove tke which is referenced but not used. It is also initialized
which overwrites the initialization in physpkg. But shouldn't impact
answers because it isn't used by the hack shallow convection.
. remove kvh which is referenced but not used. It is also initialized
which overwrites the initialization in physpkg. Shouldn't impact cam4
answers unless running with IC file that contained KVH.
. remove turbtype and smaw which are referenced but not used
models/atm/cam/src/physics/cam/tracers_suite.F90
. comment out debug output
models/atm/cam/src/physics/cam/vertical_diffusion.F90
. remove pbuf_times dimension from tke. it is set in compute_eddy_diff or
compute_hb_diff and used in moist physics to estimate subgrid vertical
velocity. **N.B.** was being incorrectly accessed without the pbuf_times
dimension in microp_aero.
. remove pbuf_times dimension from kvh. kvh in the pbuf is used by other
physics parameterizations, and as an initial guess in compute_eddy_diff
on the next timestep. **N.B.** was being incorrectly accessed without
the pbuf_times dimension in microp_aero and in ndrop.
. remove pbuf_times dimension from kvm. kvm in the pbuf is only used as an
initial guess in compute_eddy_diff on the next timestep. Thus the kvm
field in the pbuf does not contain the contributions from molecular
diffusion that are added in the call to compute_vdiff.
. remove pbuf_times dimension from turbtype and smaw. these fields in the
pbuf are currently not used, but may be in the future. Also can remove
the pbuf_set_field to zero for the HB case since the fields are init to
zero and not modified.
. tauresx and tauresy are only in the pbuf to be used as initial guesses in
subsequent timesteps. No need for multiple time levels.
. kvt only needs one time level and physpkg scope.
. kvt is set before it is used anywhere -- doesn't need an initialization
in vertical_diffusion_init
. change local array for kvt to pointer into pbuf
. remove obsolete comment about calc_kvt
. associate kvt pointer with pbuf and remove the copy of the local kvt into
the pbuf.
models/atm/cam/src/physics/waccm/gw_drag.F90
. remove dummy arg kvt and access from pbuf instead. add init for
kvt_idx. modify indexing to account for pbuf field using 1 based
indices.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Mon Aug 27 18:54:35 MDT 2012
edinburgh/lf95: All PASS except:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Mon Aug 27 19:24:15 MDT 2012
009 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Mon Aug 27 21:02:18 MDT 2012
011 bl113 TBL.sh e8c5paqdm aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Mon Aug 27 22:30:15 MDT 2012
015 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Mon Aug 27 22:40:37 MDT 2012
edinburgh/pgi: All PASS except:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Mon Aug 27 17:10:52 MDT 2012
010 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Mon Aug 27 17:31:47 MDT 2012
012 bl113 TBL.sh e8c5paqdm aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Mon Aug 27 17:46:04 MDT 2012
016 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Mon Aug 27 17:58:51 MDT 2012
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB except:
. configurations using Eulerian dycore w/ cam5 physics change answers due
to bug fixes
. FV, 1-deg runs change answers (same climate) due to changing the default
number of subcycles for the vertical remapping (nspltvrm=2).
===============================================================
===============================================================
Tag name: cam5_1_34
Originator(s): fischer, eaton, sungsup, fvitt
Date: Fri Aug 24 2012
One-line Summary: obklen, deep convection, ndrop, and WACCM concentration bug fixes
Purpose of changes:
. Mods from Sungsu Park to fix the unreasonable concentration of some
species in WACCM. The fix prevents large fluxes of species with very
small concentrations.
. Fix to wet radius calculation in the modal_aero_wateruptake module.
. Provide a computed value for the Obukhov length in the cam5 physics
package. It was inadvertently left set to zero. This impacts dry
deposition calculations.
. Mods from Guang Zhang in zm_conv to fix some inconsistency in the
initialization of hu, and to define su at cloud base using hu.
. Mods to ndrop from Cathy Chuang, Philip Cameron-smith, and Steve Ghan.
Mods to mo_setsox from Xiaohong Liu to generalize the method for
calculating pH value of cloud water.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
M models/atm/cam/bld/namelist_files/namelist_definition.xml
M models/atm/cam/bld/build-namelist
M models/atm/cam/src/physics/cam/microp_aero.F90
M models/atm/cam/src/physics/cam/vertical_diffusion.F90
. obklen variable fix for cam5
M models/atm/cam/src/physics/cam/zm_conv.F90
. deep convection bug fix
M models/atm/cam/src/physics/cam/uwshcu.F90
. bug fix for unreasonable concentration of tracers with WACCM
M models/atm/cam/src/chemistry/utils/modal_aero_wateruptake.F90
. move calculation of dgncur_awet outside of modal_aero_wateruptake_sub and
into modal_aero_wateruptake_dr. This moves it to a point after the
calculation of wetrad has been completed with the
deliquesence/crystallization hysteresis treatment.
M models/atm/cam/src/physics/cam/ndrop.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_sim_dat.F90
M models/atm/cam/src/chemistry/mozart/mo_setsox.F90
. ndrop bug fix
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
004 bl132 TBL.sh e48c4dh ghgrmp+1850_cam4 9s ..............................FAIL! rc= 7 at Thu Aug 23 09:29:18 MDT 2012
006 bl133 TBL.sh e48c4paqdm aqpgro+aquaplanet_cam4 3s .....................FAIL! rc= 7 at Thu Aug 23 09:29:20 MDT 2012
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Thu Aug 23 09:29:28 MDT 2012
013 bl330 TBL.sh f4c5paqdh aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Thu Aug 23 09:29:30 MDT 2012
017 bl331 TBL.sh f4c4dh co2rmp+1850_cam4 9s ...............................FAIL! rc= 7 at Thu Aug 23 09:29:41 MDT 2012
021 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...............................FAIL! rc= 7 at Thu Aug 23 09:29:52 MDT 2012
023 bl333 TBL.sh f4c4paqdh aqpgro+aquaplanet_cam4 3s ......................FAIL! rc= 7 at Thu Aug 23 09:29:54 MDT 2012
033 bl336 TBL.sh wm4c4h outfrq3s+waccm_1850_cam4 9s .......................FAIL! rc= 7 at Thu Aug 23 09:30:42 MDT 2012
036 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ......................................FAIL! rc= 7 at Thu Aug 23 09:30:49 MDT 2012
040 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .....................................FAIL! rc= 7 at Thu Aug 23 09:31:17 MDT 2012
045 bl373 TBL.sh fm1.9c4dh outfrq3s 9s ....................................FAIL! rc= 7 at Thu Aug 23 09:32:04 MDT 2012
047 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...........................FAIL! rc= 7 at Thu Aug 23 09:32:24 MDT 2012
050 bl374 TBL.sh fs1.9c4dh outfrq3s+1850-2005_cam4_super_fast_llnl 9s .....FAIL! rc= 7 at Thu Aug 23 09:32:51 MDT 2012
052 bl415 TBL.sh fst1.9c5dh outfrq3s 9s ...................................FAIL! rc= 7 at Thu Aug 23 09:34:05 MDT 2012
054 bl430 TBL.sh fm1.9c5dh outfrq3s 9s ....................................FAIL! rc= 7 at Thu Aug 23 09:35:06 MDT 2012
057 bl440 TBL.sh fsoa1.9c4dh outfrq3s 9s ..................................FAIL! rc= 7 at Thu Aug 23 09:35:58 MDT 2012
061 bl387 TBL.sh f1.9c4dm outfrq3s+1850-2005_cam4 9s ......................FAIL! rc= 7 at Thu Aug 23 09:36:11 MDT 2012
064 bl388 TBL.sh f1.9c4m outfrq24h+1850_cam4 2d ...........................FAIL! rc= 7 at Thu Aug 23 09:36:21 MDT 2012
067 bl379 TBL.sh wm1.9c4h outfrq24h+waccm_1850_cam4 2d ....................FAIL! rc= 7 at Thu Aug 23 09:38:37 MDT 2012
070 bl731 TBL.sh h16c3aqdh aqua+aquaplanet_cam3 9s ........................FAIL! rc= 7 at Thu Aug 23 09:38:46 MDT 2012
073 bl751 TBL.sh h30c4aqdm outfrq3s+aquaplanet_cam4 9s ....................FAIL! rc= 7 at Thu Aug 23 09:39:20 MDT 2012
079 bl992 TBL_ccsm.sh f19_g16 E 2d ........................................FAIL! rc= 7 at Thu Aug 23 10:36:02 MDT 2012
082 bl993 TBL_ccsm.sh ne30_g16 FC5 2d .....................................FAIL! rc= 7 at Thu Aug 23 12:23:05 MDT 2012
expected, all bug fixes change answers
edinburgh/lf95:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Wed Aug 22 23:23:39 MDT 2012
009 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Wed Aug 22 23:23:43 MDT 2012
011 bl113 TBL.sh e8c5paqdm aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Wed Aug 22 23:23:44 MDT 2012
015 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Wed Aug 22 23:23:50 MDT 2012
022 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Thu Aug 23 00:12:41 MDT 2012
030 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Thu Aug 23 02:58:48 MDT 2012
032 bl313 TBL.sh f10c3aqdm outfrq3s+aquaplanet_cam3 3s ....................FAIL! rc= 7 at Thu Aug 23 03:39:43 MDT 2012
034 er314 TER.sh wg10c4dm outfrq3s 4+5s ...................................FAIL! rc= 8 at Thu Aug 23 04:32:47 MDT 2012
035 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7 at Thu Aug 23 05:19:42 MDT 2012
037 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 7 at Thu Aug 23 06:49:22 MDT 2012
040 bl318 TBL.sh f10c4cdm sat_hist 9s .....................................FAIL! rc= 7 at Thu Aug 23 07:47:52 MDT 2012
expected, all bug fixes change answers
edinburgh/pgi:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Wed Aug 22 15:10:31 MDT 2012
010 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Wed Aug 22 15:19:55 MDT 2012
012 bl113 TBL.sh e8c5paqdm aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Wed Aug 22 15:24:37 MDT 2012
016 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Wed Aug 22 15:29:44 MDT 2012
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Wed Aug 22 15:47:52 MDT 2012
031 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Wed Aug 22 17:03:07 MDT 2012
033 bl313 TBL.sh f10c3aqdm outfrq3s+aquaplanet_cam3 3s ....................FAIL! rc= 7 at Wed Aug 22 17:11:49 MDT 2012
036 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7 at Wed Aug 22 17:29:57 MDT 2012
039 bl315 TBL.sh f10c3dm outfrq3s+cam3 9s .................................FAIL! rc= 7 at Wed Aug 22 17:43:44 MDT 2012
041 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 7 at Wed Aug 22 17:56:15 MDT 2012
044 bl317 TBL.sh f10c5cdm outfrq3s 9s .....................................FAIL! rc= 7 at Wed Aug 22 18:17:23 MDT 2012
047 bl319 TBL.sh fn10c5dm outfrq3s_bam 9s .................................FAIL! rc= 7 at Wed Aug 22 18:33:39 MDT 2012
050 bl320 TBL.sh fn10c5dm rad_diag 9s .....................................FAIL! rc= 7 at Wed Aug 22 18:38:34 MDT 2012
053 bl321 TBL.sh f10c5cdm atrain 9s .......................................FAIL! rc= 7 at Wed Aug 22 18:46:36 MDT 2012
056 bl420 TBL.sh f7mode4c5dm outfrq3s 9s ..................................FAIL! rc= 7 at Wed Aug 22 19:32:42 MDT 2012
059 bl734 TBL.sh hn16c5aqdm outfrq3s_bam+aquaplanet_cam5 9s ...............FAIL! rc= 7 at Wed Aug 22 20:48:29 MDT 2012
expected, all bug fixes change answers
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
cam5_1_32
obklen, deep convection, water update
cesm1_1_beta15
ndrop bug fix
- platform/compilers:
bluefire, xlf
- configure commandline:
Used cesm build scripts
create_newcase -compset F1850C5 -res f19_f19 -mach bluefire -case /glade/home/fischer/runs/F1850C5_5_1_32_xxxx
- MSS location of output:
/home/fischer/csm/F1850C5_5_1_32_mawu
bug fix for water uptake in the modal aerosol
/home/fischer/csm/F1850C5_5_1_32_obklen
Fix for obklen variable
/home/fischer/csm/F1850C5_5_1_32_zm
Deep convection bug fix
/home/tilmes/csm/f2000_cam5_beta15_modal
ndrop bug fix
MSS location of control simulations used to validate new climate:
/home/fischer/csm/F1850C5_5_1_32_bl
obklen, deep convection, water update
/home/tilmes/csm/f2000_cam5_beta15
ndrop bug
URL for AMWG diagnostics output used to validate new climate:
http://www.cgd.ucar.edu/cms/hannay/internal/cam_dev/cam5.2_dev.html
===============================================================
===============================================================
Tag name: cam5_1_33
Originator(s): fischer
Date: Wed Aug 22, 2012
One-line Summary: external updates and bug fixes
Purpose of changes:
. SVN external updates to beta16 tags
scripts4_120509a -> scripts4_120801a
. Changes to CESM build scripts.
. configure only supports -case and -clean.
cleannamelist, cleanmach and cleanall have been removed.
. RUNDIR moved from env_build.xml to env_run.xml
Default RUNDIR is now set in Machines/config_machines.xml, instead of
being based on $EXEROOT.
drvseq4_1_09 -> drvseq4_1_24
branch_tags/rasm_tags/rasm08_clm4_0_43 -> branch_tags/rasm_tags/rasm11_clm4_0_43
Only update clm tags to cesm1_1_beta15 do to a bug in clm searching for
datm files
cice4_0_20120502 -> cice4_0_20120803
newer cice tag than cesm1_1_beta16 to fix decomp types were missing from
the definition file
docn8_120502 -> docn8_120626
share3_120509 -> share3_120731
timing_120419 -> timing_120731
MCT2_7_0-120413 -> MCT2_8_0_120503
Didn't update MCT tag to github since it broke cam's build scripts
PIO
pio wasn't updated because newer version needs FORTRAN 2003, so all lahey
tests failed.
stubs1_3_02 -> stubs1_3_04
Machines_120515 -> Machines_120802
. Merge in branch tag branch_tags/bldnml_tags/bldnml04_cam5_1_31
.CESM build namelist fixes
. Cloud diagnostics bug fix in cloud_diagnostics.F90.
. Additional optional microphysics output diagnostics.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: fischer
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M SVN_EXTERNAL_DIRECTORIES
. updated externals to match cesm1_1_beta16, except for CLM, MCT,and PIO
PIO has Fortran 2003 code which isn't supported in lahey, there will
be a patch for this. MCT needs to be built using the MCT configure
script, which isn't supported with the cam stand alone build.
CLM's configure needs a definition file from datm.
M models/atm/cam/test/system/TCB.sh
. Update for cice decomp updates
. Make seperate explicit calls to generate_cice_decomp to get decomp_type,
maxblocks, bsize_x, and bsize_y.
M models/atm/cam/test/system/TCB_ccsm.sh
. Renamed env_mach_pes.xml to env_configure.xml
. Add xmlchange for RUNDIR
M models/atm/cam/bld/configure
. new decomp types for cice
M models/atm/cam/bld/cam.buildnml.csh
M models/atm/cam/bld/cam.cpl7.template
. CESM build namelist fixes
M models/atm/cam/bld/build-namelist
. Change debug print messages
M models/atm/cam/src/physics/cam/cloud_diagnostics.F90
. cloud_diagnostics bug fix, gravit*1000 should be just gravit in call to conv_water_4rad
M models/atm/cam/src/physics/cam/micro_mg_cam.F90
M models/atm/cam/src/physics/cam/physpkg.F90
M models/atm/cam/src/physics/cam/microp_driver.F90
. Additional optional microphysics output diagnostics.
M models/atm/cam/src/cpl_esmf/atm_comp_esmf.F90
. Fixed typo
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
079 bl992 TBL_ccsm.sh f19_g16 E 2d ........................................FAIL! rc= 7 at Tue Aug 21 18:07:07 MDT 2012
Change in cice decomp type changes answers for some cases
edinburgh/lf95: all passed
edinburgh/pgi: all passed
===============================================================
===============================================================
Tag name: cam5_1_32
Originator(s): santos, eaton
Date: July 23, 2012
One-line Summary: remove dependencies in phys_control, vert diffusion cleanup
Purpose of changes:
- To break the dependency of phys_control on ref_pres (and thus on the
dynamics and physconst).
- Move namelist options from phys_control to vertical_diffusion.
- Remove #ifdef MODAL_AERO from vertical_diffusion.
- Make trb_mtn_stress module fully portable.
- General cleanup (remove goto statements, unnecessary code).
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
- ref_pres_nl handles options for the reference pressure module.
- trop_cloud_top_press is defined in namelist_definitions and
always added by build_namelist.
- Value for cam5 is 100 Pa. Currently only WACCM configurations
go above this level.
- Value for cam4 is currently 0 Pa. (CAM 4 uses k1mb as the
microphysics top instead.)
- The following namelist variables were moved from phys_control to
vertical_diffusion:
- do_iss
- do_tms
- tms_orocnst
- tms_z0fac
- vdiff_compute_obklen
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Brian Eaton
List all subroutines eliminated:
models/atm/cam/src/physics/cam/phys_control.F90 :: phys_ctl_init
models/atm/cam/src/physics/cam/phys_control.F90 :: trop_cloud_top_init
- These set various top/bottom levels for vertical_diffusion and the
cloud physics, respectively.
- Their functionality has been split between routines in other
modules, in order to remove dependence on ref_pres.
List all subroutines added and what they do:
models/atm/cam/src/physics/cam/diffusion_solver.F90 :: new_fieldlist_vdiff
- Returns a new vdiff_selector ("fieldlist") object, given a
constituent number.
- Split fieldlist creation subroutine off of init_vdiff so that
module does not have to be re-initialized by eddy_diff.
models/atm/cam/src/physics/cam/rad_constituents.F90 :: rad_cnst_get_mam_mmr_idx
models/atm/cam/src/physics/cam/rad_constituents.F90 :: rad_cnst_get_mode_num_idx
- Get constituent indices of mam species mass mixing ratio, and mode
number mixing ratio, respectively.
- Used in cleanup of vertical diffusion.
models/atm/cam/src/physics/cam/ref_pres.F90 :: ref_pres_readnl
- Namelist parser for ref_pres.
- Added to move trop_cloud_top_press option from phys_control.
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/build-namelist
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- Added trop_cloud_top_press to namelist.
models/atm/cam/bld/namelist_files/namelist_definition.xml
- Added trop_cloud_top_press, moved options from phys_control to
vertical_diffusion.
models/atm/cam/src/control/runtime_opts.F90
- Added ref_pres namelist.
models/atm/cam/src/physics/cam/cldwat2m_macro.F90
models/atm/cam/src/physics/cam/cloud_fraction.F90
models/atm/cam/src/physics/cam/micro_mg_cam.F90
models/atm/cam/src/physics/cam/microp_aero.F90
models/atm/cam/src/physics/cam/ndrop_bam.F90
- Updated location of trop_cloud_top_lev from phys_control to ref_pres.
models/atm/cam/src/physics/cam/diffusion_solver.F90
- do_iss is set by input argument rather than gotten from phys_control.
- Fieldlists are created by new_fieldlist_vdiff rather than init_vdiff.
- vdiff_selector uses an allocatable list rather than a pointer, and
the .not. and "any" operators operate without slicing. (Adding "(:)"
actually does produce a semantic difference, though one that's
irrelevant here!)
models/atm/cam/src/physics/cam/eddy_diff.F90
- Get fieldlists from new_fieldlist_vdiff rather than init_vdiff.
models/atm/cam/src/physics/cam/molec_diff.F90
- Use errstring argument to init_molec_diff to remove iulog and endrun
dependencies.
- init_molec_diff now calculates top/bottom of molecular diffusion
models/atm/cam/src/physics/cam/phys_control.F90
- To remove ref_pres dependence, the following were removed
from this module, and correspondingly from phys_getopts:
- trop_cloud_top_init
- trop_cloud_top_lev
- trop_cloud_top_press
- phys_ctl_init
- do_molec_diff
- ntop_eddy/nbot_eddy
- ntop_molec/nbot_molec
- Because they relate specifically to vertical_diffusion, the following
were also removed:
- do_iss
- do_tms
- tms_orocnst
- tms_z0fac
- vdiff_compute_obklen
models/atm/cam/src/physics/cam/physpkg.F90
- phys_ctl_init and trop_cloud_top_init calls removed.
models/atm/cam/src/physics/cam/rad_constituents.F90
- Added rad_cnst_get_mam_mmr_idx and rad_cnst_get_mode_num_idx.
models/atm/cam/src/physics/cam/ref_pres.F90
- Added ref_pres_readnl.
- Add trop_cloud_top_lev calculation.
models/atm/cam/src/physics/cam/trb_mtn_stress.F90
- Used errstring to remove dependency on iulog, took dependence on
shr_const_mod back out.
models/atm/cam/src/physics/cam/tropopause.F90
- Replaced goto statements with cycle/exit.
- Comment typo fixes.
models/atm/cam/src/physics/cam/vertical_diffusion.F90
- Added variables to namelist from phys_control.
- vertical_diffusion now decides if molecular diffusion will be done.
- ntop_molec/nbot_molec are returned from init_molec_diff, not phys_getopts.
- vertical_diffusion now determines ntop_eddy/nbot_eddy values.
- Uses new_fieldlist_vdiff.
- Handles new errstring outputs from init_molec_diff and init_tms.
- New rad_constituent procedures and prog_modal_aero used to remove
#ifdef MODAL_AERO statements.
models/atm/cam/src/physics/rrtmg/rad_solar_var.F90
- Used DIM argument to maxloc to change size 1 vector to scalar.
models/atm/cam/src/physics/waccm/gw_drag.F90
- nbot_molec used from molec_diff rather than queried from phys_getopts
models/atm/cam/src/physics/waccmx/majorsp_diffusion.F90
- Unused Boltzmann constant removed.
- ntop_molec/nbot_molec used from molec_diff rather than phys_control.
- Do loop guaranteed to have exactly one iteration removed (replaced
with simple assignment).
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
- All PASS.
edinburgh/lf95:
- All PASS.
edinburgh/pgi or jaguar/pgi:
- All PASS.
Summarize any changes to answers: None. All bit-for-bit.
===============================================================
===============================================================
Tag name: cam5_1_31
Originator(s):fischer, eaton, jedwards
Date: 5-25-2012
One-line Summary: homme updates, 1/2 cam4 tunings, and bug fixes
Purpose of changes:
. Updated testing scripts to handle CESM jaguarpf to titan name change.
. Added new COSP tests that skip MISR and RADAR simulators because they
cause answer changes when the columns change. This happens when different
pes layouts, or phys_loadbalance changes.
. Added ability to set tunings differently depending on the ocean model
. Merge with cam1/branch_tags/camse_cam5_1_27_tags/01_camse_cam5_1_27/
Updates for homme, memory reduction
Changes to vert_remap_q_alg in namelist_defaults_cam.xml
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
. Added ability to set tunings differently depending on the ocean model
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
. For ne60 with 256 mpi tasks on bluefire memory usage when from
858Mb high water per task to 778Mb
648Mb low water per task to 573Mb
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/test_driver.sh
. added titan
M models/atm/cam/test/system/tests_posttag_bluefire
M models/atm/cam/test/system/input_tests_master
A models/atm/cam/test/system/nl_files/outfrq3s_cosp
. New COSP tests that skips MISR and the RADAR simulators
This is to get around failing tests because of answer changes when
columns are changed.
M models/atm/cam/bld/configure
. remove 'none' as an option for -ocn.
. remove default setting of $ocn_pkg='none' when -ccsm_seq option set
M models/atm/cam/bld/config_files/definition.xml
. replace 'none' by 'pop2' as valid value for ocn.
M models/atm/cam/bld/cam.buildnml.csh
. use $COMP_OCN to set value of -ocn argument to configure (replace
hardcoded value 'none')
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. add tunings for 1/2 deg cam4 for cldfrc_rhminh, cldfrc_rhminl,
cldfrc_sh1, cldwat_icritw
. Merge with cam1/branch_tags/camse_cam5_1_27_tags/01_camse_cam5_1_27/
Changes made to vert_remap_q_alg
M models/atm/cam/bld/build-namelist
. Added test for $cam_build when setting bndtvs and focndomain
M models/atm/cam/src/control/cam_history.F90
. Changed the len of logname from 8 to 16 to fix PGI problem on lynx.
M models/atm/cam/src/physics/cam/microp_aero.F90
. Initialize for ptend was needed.
M models/atm/cam/src/dynamics/homme/interp_mod.F90
. Updated seq_io_mod to use shr_pio_mod
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
. updated homme external to homme1_3_6
M SVN_EXTERNAL_DIRECTORIES
. updated externals tags to cesm1_1_beta14 tags
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
070 bl731 TBL.sh h16c3aqdh aqua+aquaplanet_cam3 9s ........................FAIL! rc= 7 at Thu May 24 11:46:43 MDT 2012
073 bl751 TBL.sh h30c4aqdm outfrq3s+aquaplanet_cam4 9s ....................FAIL! rc= 7 at Thu May 24 11:47:16 MDT 2012
082 bl993 TBL_ccsm.sh ne30_g16 FC5 2d .....................................FAIL! rc= 7 at Thu May 24 16:24:01 MDT 2012
. homme is expected to changes answer
edinburgh/lf95: all passed
edinburgh/pgi:
059 bl734 TBL.sh hn16c5aqdm outfrq3s_bam+aquaplanet_cam5 9s ...............FAIL! rc= 7 at Tue May 22 20:14:51 MDT 2012
062 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ....................FAIL! rc= 7 at Tue May 22 22:07:24 MDT 2012
. homme is expected to changes answer
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_1_30
Originator(s): fvitt bardeenc
Date: 18 May 2012
One-line Summary: Chemisry and CARMA bug fixes
Purpose of changes:
Chemistry bug fixes
- corrected SO2 + OH --> H2SO4 reaction in trop_strat_mam3 chemistry package
- include wet removal of H2SO4 for aerosol chemistry packages other than trop_mam3
- remove use of deprecated commap module from chemistry source code
- include of solar/geomagnetic activity data in history files for WACCM
CARMA bug fixes
- carma_do_clearsky was not being passed to CARMA_Initialize (is false by default)
- the default for carma_do_clearsky was not correct for the cirrus model (should have been false)
CARMA models
- (cirrus) changed namelist defaults to match latest runs on Pleiades
- (pmc) added dust/ice coagulation
- (test_tracers) new model for tracer trajectory studies
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Brian Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
A models/atm/cam/bld/namelist_files/use_cases/cam5_trop_strat_chem.xml
-new build-namelist use case for *STRATMAM compsets
A models/atm/cam/src/physics/carma/models/test_tracers/carma_model_mod.F90
A models/atm/cam/src/physics/carma/models/test_tracers/carma_model_flags_mod.F90
A models/atm/cam/src/physics/carma/models/test_tracers
- (test_tracers) new model for tracer trajectory studies
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/config_files/definition.xml
M models/atm/cam/bld/configure
- added new carma option (test_tracers)
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- moved "diag_cnst_conv_tend" to "cam_diags_opts" namelist group
- added definition for "met_srf_nudge_flux" namelist option
- CARMA namilist options added:
. carma_launch_doy
. carma_emission_rate
M models/atm/cam/bld/perl5lib/Build/ChemNamelist.pm
- remove H2SO4 from wet dep list only for chem package trop_mam3
M models/atm/cam/bld/perl5lib/Build/ChemPreprocess.pm
- resolve $CASEBUILD path so that the chemistry preprocessor execute
properly within CESM scripts
M models/atm/cam/test/system/test_driver.sh
M models/atm/cam/src/physics/carma/cam/carma_intr.F90
- pass do_clearsky option into CARMA model
M models/atm/cam/src/physics/carma/models/pmc/carma_model_mod.F90
- invoke CAMRA_AddCoagulation
M models/atm/cam/src/physics/carma/models/cirrus/carma_cloudfraction.F90
- change "camsrfexch_types" to "camsrfexch" so this will compile
M models/atm/cam/bld/namelist_files/use_cases/sd_waccm_tslt_geos5.xml
M models/atm/cam/bld/namelist_files/use_cases/sd_waccm_geos5.xml
- update to solar irradiance data input
- update to list of GEOS5 meteorology input files
M models/atm/cam/bld/namelist_files/use_cases/cam4_bam_radpsv_geos5.xml
- update to list of GEOS5 meteorology input files
M models/atm/cam/bld/namelist_files/use_cases/cam4_chem_radpsv_geos5.xml
- update to list of GEOS5 meteorology input files
- gas_wetdep_method set to "NEU"
M models/atm/cam/bld/namelist_files/use_cases/cam4_trop_strat_chem.xml
- set do_tms to .false.
- gas_wetdep_method set to "NEU"
M models/atm/cam/bld/build-namelist
- changes for CARMA
M models/atm/cam/src/physics/cam/cam_diagnostics.F90
M models/atm/cam/src/control/runtime_opts.F90
- moved reading of cam_diagnostics namelist options from runtime_opts to cam_diagnostics
M models/atm/cam/src/control/cam_history_support.F90
M models/atm/cam/src/control/cam_history.F90
- restored inclusion of solar/geomagnetic activity data in history files for waccm
M models/atm/cam/src/physics/cam/tropopause.F90
- added TROP_DZ diagnostic history output
M models/atm/cam/src/physics/cam/phys_control.F90
- increased length of cam_chempkg_out optional argument
M models/atm/cam/src/physics/cam/eddy_diff.F90
- corrected pressure units in description of eddy_max_bot_pressure var
M models/atm/cam/src/chemistry/pp_waccm_mozart_v1/mo_sad.F90
M models/atm/cam/src/chemistry/mozart/mo_sad.F90
- removed latndx and lonndx from arg list of sad_strat_calc routine, which are not used
M models/atm/cam/src/chemistry/utils/mo_flbc.F90
- removed use of commap module, which is deprecated
M models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
- removed latndx and lonndx from args passed into routines which don't make use of these variables
M models/atm/cam/src/chemistry/mozart/mo_fstrat.F90
M models/atm/cam/src/chemistry/mozart/mo_setext.F90
M models/atm/cam/src/chemistry/mozart/mo_photoin.F90
- removed latndx and lonndx from arg list, which are not used
M models/atm/cam/src/chemistry/mozart/mo_photo.F90
- removed latndx and lonndx from arg list, which are not used
- set do_jshort to false for fast TUV method
- removed use of deprecated commap module and use phys_grid interface to determine grid point locations
M models/atm/cam/src/chemistry/mozart/mo_seto2.F90
M models/atm/cam/src/chemistry/mozart/mo_chm_diags.F90
- output known aerosol species in units of MMR and gas-phase species as VMR
. the exception to this is trop_mam3 which outputs all species in units of MMR
M models/atm/cam/src/chemistry/mozart/mo_synoz.F90
- removed use of deprecated commap module and use dyn_grid interface to determine latitudes
and latitude weights grid boxes
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/chem_mech.in
- corrected SO2 + OH --> H2SO4 reaction
M models/atm/cam/src/dynamics/sld/grmult.F90
- corrected code to compute virtual temperature (otherwise this would not compile)
M models/atm/cam/src/dynamics/sld/sld_control_mod.F90
- correction to endrun message in dyn_sld_readnl subroutine
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
052 bl415 TBL.sh fst1.9c5dh outfrq3s 9s ...................................FAIL! rc= 7 at Thu May 17 13:16:55 MDT 2012
054 bl430 TBL.sh fm1.9c5dh outfrq3s 9s ....................................FAIL! rc= 7 at Thu May 17 13:17:55 MDT 2012
These failures are expected due to H2SO4 added to list of tracers which are wet removed
edinburgh/lf95: All Pass
edinburgh/pgi:
056 bl420 TBL.sh f7mode4c5dm outfrq3s 9s ..................................FAIL! rc= 7 at Wed May 16 18:35:24 MDT 2012
This failure is expected due to H2SO4 added to list of tracers which are wet removed
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_1_29
Originator(s): Cheryl Craig and Brian Eaton
Date: 5/10/2012
One-line Summary: Introduced driver layer for modal_aero_wateruptake and cleanup and bug fix
Purpose of changes:
Introduced driver layer for modal_aero_wateruptake to allow incorporation of sub-columns
prec_zmc and snow_zmc renamed prec_dp and snow_dp - both stored in pbuf
prec_cmf and snow_cmf renamed prec_sh and snow_sh and - both stored in pbuf
prec_str, snow_str, prec_sed, snow_sed, prec_pcw and snow_pcw now stored in pbuf
Bugs fixed (include bugzilla ID):
bug fix for klambda check in rrtmg/cloud_rad_props.F90, if == 1 then would klambda-1 would be out-of-bound index
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Eaton
List all subroutines eliminated:
D models/atm/cam/src/control/camsrfexch_types.F90
renamed camsrfexch.F90
D models/atm/cam/src/control/srfxfer.F90
moved into camsrfexch.F90 module and renamed cam_export
List all subroutines added and what they do:
A + models/atm/cam/src/control/camsrfexch.F90
renamed from camsrfexch_type.F90 and now contains old srfxfer routine which is renamed cam_export
List all existing files that have been modified, and describe the changes:
M models/atm/cam/src/chemistry/utils/modal_aero_wateruptake.F90
new driver layer (modal_aero_wateruptake_dr) introduced to aid subcolumns
M models/atm/cam/src/physics/cam/cam_diagnostics.F90
M models/atm/cam/src/physics/cam/convect_shallow.F90
M models/atm/cam/src/physics/cam/convect_deep.F90
MM models/atm/cam/src/physics/cam/microp_driver.F90
M models/atm/cam/src/physics/cam/physpkg.F90
M models/atm/cam/src/physics/cam/stratiform.F90
prec_zmc and snow_zmc renamed prec_dp and snow_dp - both stored in pbuf
prec_cmf and snow_cmf renamed prec_sh and snow_sh and - both stored in pbuf
prec_str, snow_str, prec_sed, snow_sed, prec_pcw and snow_pcw now stored in pbuf
M models/atm/cam/src/physics/rrtmg/cloud_rad_props.F90
bug fix for klambda check, if == 1 then klambda-1 would be out-of-bound index
M models/atm/cam/src/chemistry/bulk_aero/aerosol_intr.F90
calls modal_aero_wateruptake_dr now
M models/atm/cam/src/chemistry/bulk_aero/progseasalts_intr.F90
M models/atm/cam/src/chemistry/bulk_aero/dust_intr.F90
M models/atm/cam/src/chemistry/mozart/mo_lightning.F90
M models/atm/cam/src/chemistry/mozart/chemistry.F90
M models/atm/cam/src/chemistry/mozart/mo_drydep.F90
M models/atm/cam/src/chemistry/mozart/mz_aerosols_intr.F90
M models/atm/cam/src/chemistry/pp_none/chemistry.F90
M models/atm/cam/src/chemistry/utils/aerodep_flx.F90
M models/atm/cam/src/chemistry/utils/modal_aero_deposition.F90
M models/atm/cam/src/control/cam_comp.F90
M models/atm/cam/src/control/cam_restart.F90
M models/atm/cam/src/cpl_esmf/atm_comp_esmf.F90
M models/atm/cam/src/cpl_mct/atm_comp_mct.F90
M models/atm/cam/src/dynamics/eul/inital.F90
M models/atm/cam/src/dynamics/eul/stepon.F90
M models/atm/cam/src/dynamics/fv/stepon.F90
M models/atm/cam/src/dynamics/fv/fv_prints.F90
M models/atm/cam/src/dynamics/fv/metdata.F90
M models/atm/cam/src/dynamics/homme/stepon.F90
M models/atm/cam/src/physics/cam/carma_intr.F90
M models/atm/cam/src/physics/cam/cospsimulator_intr.F90
M models/atm/cam/src/physics/cam/flux_avg.F90
M models/atm/cam/src/physics/cam/iop_surf.F90
M models/atm/cam/src/physics/cam/phys_debug.F90
M models/atm/cam/src/physics/cam/radiation.F90
M models/atm/cam/src/physics/cam/radiation_data.F90
M models/atm/cam/src/physics/cam/restart_physics.F90
M models/atm/cam/src/physics/cam/zm_conv_intr.F90
M models/atm/cam/src/physics/carma/cam/carma_cloudfraction.F90
M models/atm/cam/src/physics/carma/cam/carma_intr.F90
M models/atm/cam/src/physics/carma/models/pmc/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/bc_strat/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/sulfate/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/test_swelling/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/test_detrain/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/cirrus/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/test_radiative/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/meteor_smoke/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/test_growth/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/test_passive/carma_model_mod.F90
M models/atm/cam/src/physics/carma/models/sea_salt/carma_model_mod.F90
M models/atm/cam/src/physics/rrtmg/radiation.F90
M models/atm/cam/src/physics/rrtmg/rrtmg_state.F90
M models/atm/cam/src/dynamics/sld/stepon.F90
rename of camsrfexch_types to camsrfexch
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: all BFB
edinburgh/lf95: all BFB
edinburgh/pgi: all BFB
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
Tag name: cam5_1_28
Originator(s): bardeenc santos
Date: May 8, 2012
One-line Summary: CARMA added, sathist multiple columns, cleanup
Purpose of changes:
- Add CARMA (Community Aerosol and Radiation Model for Atmospheres),
along with several CARMA models:
- bc_strat
- cirrus
- meteor_smoke
- pmc
- sea_salt
- sulfate
As well as basic test/tutorial models for CARMA:
- test_detrain
- test_growth
- test_passive
- test_radiative
- test_swelling
- Add new chemistry package: waccm_mozart_sulfur, usable with
the CARMA sulfate model.
- New features used by CARMA, but potentially relevant outside of
CARMA, include:
- The ability to turn off some processes in MG microphysics and
Park macrophysics.
- A flag specifying the stage at which a constituent will undergo
convective transport.
- New spectral flux outputs for radiation.
- The ability to calculate the Obukhov length outside the HB/HBR
eddy scheme.
- Allow sat_hist to use more than one column near a given point, as
well as giving columns from more than one timestep, to allow
interpolation in time and space.
- Update/add to WACCM namelist defaults for certain configurations.
- Tweaked forcing from meteorological data.
- Improve Darwin parallel support.
- Minor cleanup to RRTMG.
- Some options from cldwat have been moved into cloud_fraction.
Bugs fixed (include bugzilla ID):
- Bug #1475 has been partially addressed by this tag, which removed
the faulty option (liu_in == .true. is the only valid behavior for
MG microphysics). The faulty functionality will be restored with
the addition of further ice nucleation options in a future tag.
Describe any changes made to build system:
- Improved support for compilation using Darwin.
- Added configure and namelist options specific to CARMA and
waccm_mozart_sulfur, including also.
- Namelist variables for new satellite history, RRTMG spectral
(per-band) fluxes, changing wv_saturation scheme.
- Moved/combined some variables from cldwat to cloud_fraction
(cldfrc).
Describe any changes made to the namelist:
- New defaults for certain WACCM resolutions:
atm/cam/inic/fv/cami_0000-01-01_1.9x2.5_L64_c070703.nc
atm/cam/inic/fv/cami_0000-01-01_4x5_L64_c090108.nc
atm/cam/inic/fv/cami_0000-01-01_10x15_L64_c081013.nc
atm/waccm/ic/cami_2000-07-01_1.9x2.5_L125_c040928.nc
- WACCM emissions files updated:
atm/cam/chem/1850-2000_emis/IPCC_emissions_houw_NOx_1850-2000_1.9x2.5.c090728.nc
atm/cam/chem/1850-2000_emis/IPCC_emissions_houw_CO_1850-2000_1.9x2.5.c090728.nc
atm/cam/chem/1850-2000_emis/IPCC_emissions_houw_CH2O_1850-2000_1.9x2.5.c090728.nc
atm/cam/chem/1850-2000_emis/IPCC_emissions_houw_SO2_1850-2000_1.9x2.5.c090522.nc
In the waccm_2000_cam4 use case:
atm/waccm/lb/LBC_1765-2005_1.9x2.5_CMIP5_za_c111110.nc
- Defaults for recently added eddy diffusion variables are now set
in namelist_defaults_cam.xml rather than hard-coded in the module:
0.D0
1.D0
1.D0
40.e3
40.e3
100.e3
.false.
- Several changes specific to CARMA and waccm_mozart_sulfur.
- H2SO4 added to drydep and wetdep lists.
- New logical flags for turning off processes in macro/microphysics:
- macro_park_do_cldice
- macro_park_do_cldliq
- macro_park_do_detrain
- micro_mg_do_cldice
- micro_mg_do_cldliq
- Namelist options for cldfrc replacing functionality in cldwat:
- cldfrc_rhminl_adj_land
- cldfrc_premib
- cldfrc_iceopt
- cldfrc_icecrit
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
A models/atm/cam/src/chemistry/pp_waccm_mozart_sulfur/
A models/atm/cam/src/chemistry/pp_waccm_mozart_sulfur/chem_mech.doc
A models/atm/cam/src/chemistry/pp_waccm_mozart_sulfur/chem_mech.in
A models/atm/cam/src/chemistry/pp_waccm_mozart_sulfur/chem_mods.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_sulfur/m_rxt_id.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_sulfur/m_spc_id.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_sulfur/mo_adjrxt.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_sulfur/mo_imp_sol.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_sulfur/mo_indprd.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_sulfur/mo_lin_matrix.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_sulfur/mo_lu_factor.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_sulfur/mo_lu_solve.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_sulfur/mo_nln_matrix.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_sulfur/mo_phtadj.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_sulfur/mo_prod_loss.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_sulfur/mo_rxt_rates_conv.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_sulfur/mo_setrxt.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_sulfur/mo_sim_dat.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_sulfur/mo_usrrxt.F90
- New waccm_mozart chemistry branch.
A models/atm/cam/src/physics/cam/carma_flags_mod.F90
- Module that provides CARMA-specific flags.
A models/atm/cam/src/physics/cam/carma_intr.F90
A models/atm/cam/src/physics/cam/carma_model_flags_mod.F90
- Stubs for CARMA-specific code, used when the corresponding
CARMA modules are not compiled.
A models/atm/cam/src/physics/carma
A models/atm/cam/src/physics/carma/cam/
A models/atm/cam/src/physics/carma/cam/carma_cloudfraction.F90
A models/atm/cam/src/physics/carma/cam/carma_constants_mod.F90
A models/atm/cam/src/physics/carma/cam/carma_getT.F90
A models/atm/cam/src/physics/carma/cam/carma_intr.F90
A models/atm/cam/src/physics/carma/cam/carma_precision_mod.F90
A models/atm/cam/src/physics/carma/models/
A models/atm/cam/src/physics/carma/models/bc_strat/
A models/atm/cam/src/physics/carma/models/bc_strat/carma_model_mod.F90
A models/atm/cam/src/physics/carma/models/cirrus/
A models/atm/cam/src/physics/carma/models/cirrus/carma_cloudfraction.F90
A models/atm/cam/src/physics/carma/models/cirrus/carma_model_flags_mod.F90
A models/atm/cam/src/physics/carma/models/cirrus/carma_model_mod.F90
A models/atm/cam/src/physics/carma/models/cirrus/growevapl.F90
A models/atm/cam/src/physics/carma/models/meteor_smoke/
A models/atm/cam/src/physics/carma/models/meteor_smoke/carma_model_flags_mod.F90
A models/atm/cam/src/physics/carma/models/meteor_smoke/carma_model_mod.F90
A models/atm/cam/src/physics/carma/models/pmc/
A models/atm/cam/src/physics/carma/models/pmc/carma_model_flags_mod.F90
A models/atm/cam/src/physics/carma/models/pmc/carma_model_mod.F90
A models/atm/cam/src/physics/carma/models/sea_salt/
A models/atm/cam/src/physics/carma/models/sea_salt/carma_model_flags_mod.F90
A models/atm/cam/src/physics/carma/models/sea_salt/carma_model_mod.F90
A models/atm/cam/src/physics/carma/models/sulfate/
A models/atm/cam/src/physics/carma/models/sulfate/carma_model_mod.F90
A models/atm/cam/src/physics/carma/models/test_detrain/
A models/atm/cam/src/physics/carma/models/test_detrain/carma_model_mod.F90
A models/atm/cam/src/physics/carma/models/test_growth/
A models/atm/cam/src/physics/carma/models/test_growth/carma_model_mod.F90
A models/atm/cam/src/physics/carma/models/test_passive/
A models/atm/cam/src/physics/carma/models/test_passive/carma_model_mod.F90
A models/atm/cam/src/physics/carma/models/test_radiative/
A models/atm/cam/src/physics/carma/models/test_radiative/carma_model_mod.F90
A models/atm/cam/src/physics/carma/models/test_swelling/
A models/atm/cam/src/physics/carma/models/test_swelling/carma_model_mod.F90
- Added CARMA-CAM interface and several CARMA models.
X models/atm/cam/src/physics/carma/base
- External for CARMA's base code.
A models/atm/cam/test/system/config_files/f1.9c4carmsulm
A models/atm/cam/test/system/config_files/f1.9c5carmbc_m
A models/atm/cam/test/system/config_files/f1.9c5carmcirm
A models/atm/cam/test/system/config_files/f1.9c5carmmetm
A models/atm/cam/test/system/config_files/f1.9c5carmpmcm
A models/atm/cam/test/system/config_files/f1.9c5carmseam
A models/atm/cam/test/system/nl_files/carma24h
A models/atm/cam/test/system/tests_carma
-Added CARMA-specific tests for various models
List all existing files that have been modified, and describe the changes:
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
- Added CARMA base external, updated RRTMG external.
M models/atm/cam/bld/Makefile.in
M models/atm/cam/src/utils/abortutils.F90
M models/atm/cam/tools/interpic/Makefile
- Improved Darwin support.
M models/atm/cam/bld/config_files/definition.xml
M models/atm/cam/bld/configure
- Added CARMA, waccm_mozart_sulfur.
M models/atm/cam/bld/namelist_files/master_drydep_list.xml
M models/atm/cam/bld/namelist_files/master_gas_wetdep_list.xml
- Added H2SO4 to main gas phase lists.
- Removed MAM-specific drydep entry to prevent
double-counting.
M models/atm/cam/bld/build-namelist
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
M models/atm/cam/bld/namelist_files/namelist_definition.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2000_cam4.xml
- Added CARMA, waccm_mozart_sulfur, and sathist options.
- New files for various WACCM resolutions.
- Updated WACCM emis_file defaults (but usually overridden by
use case defaults anyway).
- Eddy diffusion defaults now set in namelist_defaults_cam.
M models/atm/cam/bld/perl5lib/Build/ChemNamelist.pm
- Tweaks to gas_wetdep_list moved from build-namelist to
set_dep_lists.
- Avoid changing answers for trop_mam due to
waccm_mozart_sulfur's addition of H2SO4 to gas wetdep.
M models/atm/cam/src/chemistry/bulk_aero/aerosol_intr.F90
- Notify hub2atm if CARMA is doing drydep.
M models/atm/cam/src/chemistry/bulk_aero/wetdep.F90
- is_strat_cloudborne handled properly as optional argument
rather than using #ifdef MODAL_AERO statements.
M models/atm/cam/src/chemistry/utils/modal_aero_calcsize.F90
- Unnecessary use statement removed.
M models/atm/cam/src/control/physconst.F90
- Some constants from shr_const_mod are now used instead of
recalculated with identical formulae.
M models/atm/cam/src/control/runtime_opts.F90
- Add CARMA readnl, microp/macrop driver readnl, and the new
RRTMG spectralflux option.
M models/atm/cam/src/control/sat_hist.F90
M models/atm/cam/src/dynamics/eul/dyn_grid.F90
M models/atm/cam/src/dynamics/fv/dyn_grid.F90
M models/atm/cam/src/dynamics/homme/dyn_grid.F90
M models/atm/cam/src/dynamics/sld/dyn_grid.F90
- Now can use n closest columns to a given point, rather than
only the nearest column.
- sat_hist.F90 can output multiple timesteps as well.
- Distance is now the great circle distance rather than
lon^2+lat^2 (this will change the closest column found in
some cases, particularly unstructured grids, points near 0
degrees longitude, and points near the poles).
M models/atm/cam/src/dynamics/eul/grmult.F90
M models/atm/cam/src/dynamics/sld/grmult.F90
- Replaced call to virtem with equivalent one-liner.
M models/atm/cam/src/dynamics/eul/inidat.F90
M models/atm/cam/src/dynamics/fv/inidat.F90
M models/atm/cam/src/dynamics/homme/inidat.F90
M models/atm/cam/src/dynamics/sld/inidat.F90
- Add CARMA constituents.
M models/atm/cam/src/dynamics/fv/metdata.F90
- Meteorological data nudges, rather than forces, some CAM
fields by default.
- New namelist variable met_srf_nudge_flux can be set to
.false. to recover the old behavior.
M models/atm/cam/src/physics/cam/cldwat2m_macro.F90
M models/atm/cam/src/physics/cam/cloud_fraction.F90
- Account for CARMA's processes without trying to redo them.
- Some subroutines made publically available for CARMA.
- Some parameters now housed in cloud_fraction and read in
from the namelist rather than set between cldwat2m_macro
and cloud_fraction.
M models/atm/cam/src/physics/cam/macrop_driver.F90
M models/atm/cam/src/physics/cam/micro_mg.F90
M models/atm/cam/src/physics/cam/micro_mg_cam.F90
M models/atm/cam/src/physics/cam/microp_driver.F90
- Account for physical processes executed by CARMA.
- New namelist flags to control this behavior, and new readnl
methods to get those flags.
- The liu_in option has been removed, since setting it to
false has produced garbage for some time now. See bug
#1475 in bugzilla.
- microp_driver now uses switch statement rather than if (this
anticipates MG2 and/or other potential new schemes it will
have to handle).
M models/atm/cam/src/physics/cam/constituents.F90
M models/atm/cam/src/physics/cam/stratiform.F90
M models/atm/cam/src/physics/cam/zm_conv_intr.F90
- Added flag for constituents to undergo convective transport
in first or second stage.
M models/atm/cam/src/physics/cam/eddy_diff.F90
- Instead of trbintd, ustar now calculated from new calc_ustar
subroutine in pbl_utils.
- Surface density now output with ustar.
M models/atm/cam/src/physics/cam/hb_diff.F90
- Made clear that the module declares and uses its own value
for zvir.
- Added HB_ri field for Richardson number.
- Use routines from pbl_utils to calculate virtual
temperature, ustar, kinematic surface fluxes, and Obukhov
length.
- Virtual temperature is now only calculated once and reused,
rather than recalculated several times.
- Removed now-unnecessary arguments from several routines.
- Instead of all constituent fluxes being input, only uses
water vapor.
M models/atm/cam/src/physics/cam/ndrop_bam.F90
- Aerosols with no defined dispersion exempted from CCN loops.
M models/atm/cam/src/physics/cam/phys_control.F90
- Added wvsat_scheme option, since the scheme can now be
changed.
- Added flag vdiff_compute_obklen to force computation of
Obukhov length outside the vertical diffusion.
M models/atm/cam/src/physics/cam/phys_grid.F90
- Added phys_grid_find_cols, identical to phys_grid_find_col
except that it finds multiple nearest columns and uses the
great circle distance to find them.
M models/atm/cam/src/physics/cam/phys_prop.F90
- Added 'trim' to string included in endrun message.
M models/atm/cam/src/physics/cam/physpkg.F90
- Set convective transport stage for water vapor.
- Added calls to CARMA.
M models/atm/cam/src/physics/cam/radconstants.F90
M models/atm/cam/src/physics/cam/radsw.F90
- CAMRT's radsw no longer thinks it's clever to sneak an index
into the hundredth's place of a float. Sure, it seemed like
the thing to do when it was young and headstrong, but times
have changed, and radsw.F90 has grown up. It has adopted the
more conventional approach of accepting a separate integer
array from radconstants.
- Accordingly, wavmin and wavmax are only available through
an improved accessor function.
M models/atm/cam/src/physics/cam/radiation.F90
M models/atm/cam/src/physics/rrtmg/radiation.F90
M models/atm/cam/src/physics/rrtmg/radlw.F90
M models/atm/cam/src/physics/rrtmg/radsw.F90
- Added new spectralflux option. Only used in RRTMG.
M models/atm/cam/src/physics/cam/vertical_diffusion.F90
- No longer sets defaults that are always provided via
namelist.
- Accomodates changes in eddy diffusion subroutine arguments.
- If vdiff_compute_obklen == .true., now calculates obklen
itself regardless of the eddy scheme.
M models/atm/cam/src/physics/cam/wv_saturation.F90
- 'MurphyKoop' scheme added for new wvsat_scheme option.
M models/atm/cam/src/physics/cam/zm_conv.F90
- Removed pcnst dependency (workaround for bug on PGI).
M models/atm/cam/src/physics/rrtmg/ebert_curry.F90
M models/atm/cam/src/physics/rrtmg/oldcloud.F90
M models/atm/cam/src/physics/rrtmg/slingo.F90
- wavmin and wavmax are now set by accessor function.
M models/atm/cam/src/physics/rrtmg/rad_solar_var.F90
M models/atm/cam/src/physics/rrtmg/radconstants.F90
- Tweak to include far-IR in solar variability is now done in
rad_solar_var rather than radconstants. (Lowest bin extends
down to 100 microns, regardless of actual value.)
- Accessor functions (get_sw/lw_spectral_boundaries) have
minor improvements.
M models/atm/cam/test/system/input_tests_master
- Added CARMA tests.
M models/atm/cam/bld/perl5lib/Build/ChemPreprocess.pm
M models/atm/cam/src/physics/cam/phys_debug_util.F90
M models/atm/cam/src/physics/waccm/gw_drag.F90
M models/atm/cam/src/utils/cam_pio_utils.F90
- Cosmetic changes (comments, typos, whitespace).
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: all passed
edinburgh/lf95: all passed except
bl318 TBL.sh f10c4cdm sat_hist 9s
This is an expected answer change, because satellite history
interpolation now uses the correct great circle distance to find
distances to columns.
edinburgh/pgi or jaguar/pgi: all passed
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: Any with satellite history output.
- what platforms/compilers: All
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
- Diagnostic output in satellite history file changes for some
columns; this has no effect on other output, which is all
bit for bit.
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_1_27
Originator(s): jedwards, mtaylor, mlevy
Date: 4-20-2012
One-line Summary: CAM-SE dycore updates, externals update
Purpose of changes: CAM_SE development, compatability with changes in PIO, driver and csm_share externals
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
added vert_remap_q_alg to allow a toggle between three possible algorythms for vertical remapping in camse
renamed pio_inparm to pio_default_inparm for cam standalone for compatibility with cesm build changes
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Edwards, Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
models/atm/cam/bld/cam.buildnml.csh
models/atm/cam/bld/cam.buildexe.csh
Part of new cesm build mechanism
List all existing files that have been modified, and describe the changes:
models/atm/cam/test/system/nl_files/aqua
added FU and FV to history file so that dyn grid variables are tested in history restart
models/atm/cam/test/system/TCB_ccsm.sh
removed obsolete option -skip_rundb
models/atm/cam/test/system/TSM_ccsm.sh
Added -f flag to rm to avoid error message when file not found
models/atm/cam/bld/configure
Changed when variables are resolved in cesm build
models/atm/cam/bld/Makefile.in
Corrected rpath flag
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
Changed default cam-se limiter option
models/atm/cam/bld/namelist_files/namelist_definition.xml
Added cam-se vert_remap_q_alg flag controls vertical remapping in dycore, updated pio namelist variable names
models/atm/cam/bld/build-namelist
Change pio namelist name
models/atm/cam/bld/cam.cpl7.template
Code moved to cam.build*.csh
models/atm/cam/SVN_EXTERNAL_DIRECTORIES
update dynamics/homme/share to latest cam-se dycore code
models/atm/cam/src/dynamics/sld/dyn_grid.F90
models/atm/cam/src/dynamics/eul/dyn_grid.F90
models/atm/cam/src/dynamics/homme/dyn_grid.F90
models/atm/cam/src/dynamics/fv/dyn_grid.F90
models/atm/cam/src/utils/cam_pio_utils.F90
Changed from seq_io_mod to shr_pio_mod, changed ldof arrays to kind=pio_offset
models/atm/cam/src/control/cam_history.F90
Fixed a problem with camse dyn grid variables in history restart file
models/atm/cam/src/physics/cam/tidal_diag.F90
Changed source of constants from shr_const_mod to physconst
models/atm/cam/src/physics/cam/phys_grid.F90
Changed default load balance option from 0 to 2
models/atm/cam/src/dynamics/homme/interp_mod.F90
models/atm/cam/src/dynamics/homme/native_mapping.F90
models/atm/cam/src/dynamics/homme/dyn_comp.F90
models/atm/cam/src/dynamics/homme/restart_dynamics.F90
models/atm/cam/src/dynamics/homme/inital.F90
Updated interface to cam-se share code
models/atm/cam/src/dynamics/homme/inidat.F90
Fixed pertlim to apply perturbations to initial temperature field
SVN_EXTERNAL_DIRECTORIES
Updated externals
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All pass except
070 bl731 TBL.sh h16c3aqdh aqua+aquaplanet_cam3 9s ........................FAIL
073 bl751 TBL.sh h30c4aqdm outfrq3s+aquaplanet_cam4 9s ....................FAIL
082 bl993 TBL_ccsm.sh ne30_g16 FC5 2d .....................................FAIL
edinburgh/lf95: All pass except
040 bl318 TBL.sh f10c4cdm sat_hist 9s .....................................FAIL!
edinburgh/pgi: All pass except
044 bl317 TBL.sh f10c5cdm outfrq3s 9s .....................................FAIL!
059 bl734 TBL.sh hn16c5aqdm outfrq3s_bam+aquaplanet_cam5 9s ...............FAIL!
062 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ....................FAIL!
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
Changes to CAM-SE for which we do not yet have a standard climate.
Changes to COSP tests due to the change in default phys_loadbalance from 0 to 2,
this is a bug to be investigated.
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_1_26
Originator(s): fischer
Date: Mon Apr 9 15:59:20 MDT 2012
One-line Summary: svn external updates, Makefile updates, testing updates
Purpose of changes:
. svn externals updated to cesm1_1_beta11 component tags. csm_share tag had a bug
that was caught by the lahey compiler, so the latest csm_share tag was
needed, along with a newer esmf_wrf_timemgr and clm tags.
. Makefile.in was updated to set CPPDEF flags to specify different compilers
in the latests csm_share tag.
. CESM tests for 2deg FAMIPC5 and T42 FC5 were added to bluefire posttag test.
PES layouts for cesm are being overridden by the cam tests scripts.
lrg_regular queue is being used for bluefire testing.
. fatmgrid was removed from the namelists, now using combined grid/land
fraction data fatmlndfrc
. atm_id was made public in cam_instance.F90 to speed up lookups in shr_pio_mod
. New initial condition files for ne16 and ne60
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
. Makefile.in was updated to set CPPDEF flags for the different compilers
in the latests csm_share tag.
For PGI compiler set -DCPRPGI
For INTEL compiler set -DCPRINTEL
For Pathscale compiler set -DCPRPATHSCALE
For Cray compiler set -DCPRCRAY
Describe any changes made to the namelist:
. new fatmlndfrc files, fatmgrid removed
share/domains/domain.clm/domain.lnd.T85_USGS.111004.nc
share/domains/domain.clm/domain.lnd.T42_USGS.111004.nc
share/domains/domain.lnd.T31_gx3v7.090928.nc
share/domains/domain.lnd.T21_USGS.111004.nc
share/domains/domain.clm/domain.lnd.T5_USGS.111004.nc
share/domains/domain.lnd.fv0.23x0.31_gx1v6.100517.nc
share/domains/domain.lnd.fv0.47x0.63_gx1v6.090407.nc
share/domains/domain.lnd.fv0.9x1.25_gx1v6.090309.nc
share/domains/domain.lnd.fv1.9x2.5_gx1v6.090206.nc
share/domains/domain.lnd.fv2.5x3.33_gx3v7.110223.nc
share/domains/domain.lnd.fv4x5_USGS.110713.nc
share/domains/domain.lnd.fv10x15_USGS.110713.nc
. New ncdata files for CAM5 ne16 and ne60
atm/cam/inic/homme/cami-mam3_0000-01-ne16np4_L30_c090306.nc
atm/cam/inic/homme/cami-mam3_0000-01-ne60np4_L30_c090306.nc
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/tests_posttag_bluefire
M models/atm/cam/test/system/input_tests_master
- add FAMIPC5 and T42 CAM5 tests
M models/atm/cam/test/system/TCB_ccsm.sh
- override cesm scripts pes layouts for cesm tests
M models/atm/cam/test/system/test_driver.sh
- change bluefire queue to lrg_regular until we can replace the
ne30 test with an ne16 tests
- updated hopper library paths for NETCDF
M models/atm/cam/test/system/CAM_runcmnd.sh
- set number of pes pre node to 24 for hopper
M models/atm/cam/bld/Makefile.in
- added CPPDEF compiler flags -DCPRPGI -DCPRINTEL -DCPRPATHSCALE and
-DCPRCRAY, which are needed to build csm_share
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- added ne16 and ne60 CAM5 ncdata files
- removed fatmgrid, switched to new combined grid/land fraction data files
fatmlndfrc
M models/atm/cam/bld/namelist_files/namelist_definition.xml
M models/atm/cam/bld/build-namelist
- removed fatmgrid
M models/atm/cam/bld/cam.cpl7.template
- fixed problem were cam template was being over written
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
- updated rrtmg externals
M models/atm/cam/src/control/cam_instance.F90
- made atm_id public. This was done to make lookups in shr_pio_mod faster.
M SVN_EXTERNAL_DIRECTORIES
- updated to cesm1_1_beta11 component tags
- needed to update to latest clm, esmf_wrf_timemgr, and csm_share tags
because of bugs that caused lahey to fail
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
004 bl132 TBL.sh e48c4dh ghgrmp+1850_cam4 9s ..............................FAIL! rc= 7 at Fri Apr 6 09:21:02 MDT 2012
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Fri Apr 6 09:21:10 MDT 2012
017 bl331 TBL.sh f4c4dh co2rmp+1850_cam4 9s ...............................FAIL! rc= 7 at Fri Apr 6 09:21:20 MDT 2012
021 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...............................FAIL! rc= 7 at Fri Apr 6 09:21:31 MDT 2012
033 bl336 TBL.sh wm4c4h outfrq3s+waccm_1850_cam4 9s .......................FAIL! rc= 7 at Fri Apr 6 09:22:17 MDT 2012
040 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .....................................FAIL! rc= 7 at Fri Apr 6 09:22:43 MDT 2012
045 bl373 TBL.sh fm1.9c4dh outfrq3s 9s ....................................FAIL! rc= 7 at Fri Apr 6 09:23:31 MDT 2012
047 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...........................FAIL! rc= 7 at Fri Apr 6 09:23:50 MDT 2012
050 bl374 TBL.sh fs1.9c4dh outfrq3s+1850-2005_cam4_super_fast_llnl 9s .....FAIL! rc= 7 at Fri Apr 6 09:24:16 MDT 2012
052 bl415 TBL.sh fst1.9c5dh outfrq3s 9s ...................................FAIL! rc= 7 at Fri Apr 6 09:58:04 MDT 2012
054 bl430 TBL.sh fm1.9c5dh outfrq3s 9s ....................................FAIL! rc= 7 at Fri Apr 6 09:59:04 MDT 2012
057 bl440 TBL.sh fsoa1.9c4dh outfrq3s 9s ..................................FAIL! rc= 7 at Fri Apr 6 10:00:27 MDT 2012
061 bl387 TBL.sh f1.9c4dm outfrq3s+1850-2005_cam4 9s ......................FAIL! rc= 7 at Fri Apr 6 10:00:42 MDT 2012
064 bl388 TBL.sh f1.9c4m outfrq24h+1850_cam4 2d ...........................FAIL! rc= 7 at Fri Apr 6 10:15:11 MDT 2012
067 bl379 TBL.sh wm1.9c4h outfrq24h+waccm_1850_cam4 2d ....................FAIL! rc= 7 at Fri Apr 6 11:06:22 MDT 2012
079 bl992 TBL_ccsm.sh f19_g16 E 2d ........................................FAIL! rc= 7 at Fri Apr 6 12:05:46 MDT 2012
082 bl993 TBL_ccsm.sh ne30_g16 FC5 2d .....................................FAIL! rc= 7 at Fri Apr 6 13:34:04 MDT 2012
new clm tag is not bfb and causes baseline failures
edinburgh/lf95:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Mon Apr 9 09:47:37 MDT 2012
010 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Mon Apr 9 09:47:41 MDT 2012
016 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Mon Apr 9 09:47:46 MDT 2012
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Mon Apr 9 09:47:51 MDT 2012
031 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Mon Apr 9 09:48:30 MDT 2012
036 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7 at Mon Apr 9 09:58:14 MDT 2012
039 bl315 TBL.sh f10c3dm outfrq3s+cam3 9s .................................FAIL! rc= 7 at Mon Apr 9 10:13:25 MDT 2012
044 bl317 TBL.sh f10c5cdm outfrq3s 9s .....................................FAIL! rc= 7 at Mon Apr 9 10:48:52 MDT 2012
047 bl319 TBL.sh fn10c5dm outfrq3s_bam 9s .................................FAIL! rc= 7 at Mon Apr 9 11:05:56 MDT 2012
050 bl320 TBL.sh fn10c5dm rad_diag 9s .....................................FAIL! rc= 7 at Mon Apr 9 11:11:25 MDT 2012
053 bl321 TBL.sh f10c5cdm atrain 9s .......................................FAIL! rc= 7 at Mon Apr 9 11:18:50 MDT 2012
056 bl420 TBL.sh f7mode4c5dm outfrq3s 9s ..................................FAIL! rc= 7 at Mon Apr 9 12:04:00 MDT 2012
new clm tag is not bfb and causes baseline failures
edinburgh/pgi:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Fri Apr 6 11:15:26 MDT 2012
010 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Fri Apr 6 11:24:26 MDT 2012
016 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Fri Apr 6 11:35:05 MDT 2012
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Fri Apr 6 11:54:22 MDT 2012
031 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Fri Apr 6 12:50:52 MDT 2012
036 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7 at Fri Apr 6 13:08:14 MDT 2012
039 bl315 TBL.sh f10c3dm outfrq3s+cam3 9s .................................FAIL! rc= 7 at Fri Apr 6 13:16:02 MDT 2012
044 bl317 TBL.sh f10c5cdm outfrq3s 9s .....................................FAIL! rc= 7 at Fri Apr 6 13:34:37 MDT 2012
047 bl319 TBL.sh fn10c5dm outfrq3s_bam 9s .................................FAIL! rc= 7 at Fri Apr 6 13:44:03 MDT 2012
050 bl320 TBL.sh fn10c5dm rad_diag 9s .....................................FAIL! rc= 7 at Fri Apr 6 13:47:54 MDT 2012
053 bl321 TBL.sh f10c5cdm atrain 9s .......................................FAIL! rc= 7 at Fri Apr 6 13:53:37 MDT 2012
056 bl420 TBL.sh f7mode4c5dm outfrq3s 9s ..................................FAIL! rc= 7 at Fri Apr 6 14:25:54 MDT 2012
new clm tag is not bfb and causes baseline failures
Summarize any changes to answers, i.e.: new clm tag is not bfb and causes baseline failures
===============================================================
===============================================================
Tag name: cam5_1_25
Originator(s): fvitt, emmons, mvertens
Date: 23 Mar 2012
One-line Summary: Corrections to tropospheric chemistry mechanisms and emissions inputs.
Purpose of changes:
- Corrections to reactions in tropospheric chemistry packages
to include M in reactions:
MPAN + OH
HCN + OH
- Include HCOOH emissions in default namelist for tropospheric
chemistry packages
- Simplified specifications of default emission files for MAM-chemistry packages
- Change in CAM configure to compile with MPICH2
- Fix multiple instance in cesm scripts (cam.cpl7.template)
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Chris Fischer
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/cam.cpl7.template
- Fix multiple instance in cesm scripts
M models/atm/cam/bld/configure
- check for MPICH2 -- do not set cppdefs NO_MPI2 and MO_MPIMOD for MPICH2
M models/atm/cam/bld/build-namelist
- include emissions files for HCOOH for tropospheric chem pkgs
- corrected default emissions for trop_mozart_soa chem pkg
- pass in ref to hash into get_default_value for emissions files
-- use "ver" attribute for emis files to simplify namelist_defaults_cam.xml
- change default emis files for super_fast_llnl_mam3 and trop_mam7
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- update megan_factors_file
- change location of default trop chem emissions
- specify emissions for mam chemistry with "ver" attribute
to combine many of the deplicate emissions
M models/atm/cam/bld/namelist_files/use_cases/1850-2005_cam4_super_fast_llnl.xml
M models/atm/cam/bld/namelist_files/use_cases/2000_cam4_super_fast_llnl.xml
M models/atm/cam/bld/namelist_files/use_cases/1850_cam4_super_fast_llnl.xml
M models/atm/cam/bld/namelist_files/use_cases/1850_cam5_super_fast_llnl.xml
- use mapped MEGAN emission factors for isoprene
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_rxt_rates_conv.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_phtadj.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/m_spc_id.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_mozart/chem_mech.in
- change in chemistry mechanism results in new preprocessor generated
source files, many of which are just change in format
- molecular masses are more presice with updated preprocessor
- added M to reactions:
MPAN + OH
HCN + OH
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_rxt_rates_conv.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_mozart_soa/chem_mech.in
- added M to reactions:
MPAN + OH
HCN + OH
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_setrxt.F90
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_rxt_rates_conv.F90
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_phtadj.F90
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/m_spc_id.F90
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/chem_mech.in
- change in chemistry mechanism results in new preprocessor generated
source files, many of which are just change in format
- molecular masses are more presice with updated preprocessor
- added M to reactions:
MPAN + OH
HCN + OH (and added .5*CO2 + M products)
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_rxt_rates_conv.F90
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_mozart_mam3/chem_mech.in
- added M to reactions:
MPAN + OH
HCN + OH
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_setrxt.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_rxt_rates_conv.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_strat_mam3/chem_mech.in
- added M to reactions:
MPAN + OH
HCN + OH (and added .5*CO2 + M products)
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
050 bl374 TBL.sh fs1.9c4dh outfrq3s+1850-2005_cam4_super_fast_llnl 9s .....FAIL! rc= 7 at Thu Mar 22 04:17:54 MDT 2012
- Change in MEGAN emissions of isoprene using mapped factors
045 bl373 TBL.sh fm1.9c4dh outfrq3s 9s ....................................FAIL! rc= 7 at Thu Mar 22 04:17:31 MDT 2012
052 bl415 TBL.sh fst1.9c5dh outfrq3s 9s ...................................FAIL! rc= 7 at Thu Mar 22 04:19:06 MDT 2012
054 bl430 TBL.sh fm1.9c5dh outfrq3s 9s ....................................FAIL! rc= 7 at Thu Mar 22 04:20:07 MDT 2012
057 bl440 TBL.sh fsoa1.9c4dh outfrq3s 9s ..................................FAIL! rc= 7 at Thu Mar 22 04:20:58 MDT 2012
- All these failures are due to change in chemitry mechanism and emissions
076 eq991 TEQ_ccsm.sh f19_f19 F f1.9c4m fcase 2d ..........................FAIL! rc= 7 at Thu Mar 22 06:22:27 MDT 2012
- As in previous tag, problems with clm namelist variable being changed from fatmgrid to fatmlndfrc
edinburgh/lf95: All pass
edinburgh/pgi:
056 bl420 TBL.sh f7mode4c5dm outfrq3s 9s ..................................FAIL! rc= 7 at Wed Mar 21 23:47:35 MDT 2012
- Change in emissions inputs caused this baseline failure for trop_mam7 chemistry
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_1_24
Originator(s): fischer, edwards, eaton, neale, mvertens
Date: Mar 16, 2012
One-line Summary: CAM template update, namelist updates for homme, minor bug fixes
Purpose of changes:
. Bring in cam.cpl7.template updates from a branch tag
. Update namelist option for homme.
Updated soil_erod for the default and homme
cldfrc_permit is now being set for homme
Added ne30 and ne120 ncdata files
. Add a CAM5 ne30_g16 CESM homme dycore test.
. Fix memory leak in stratiform.F90. When use_shfrc is false,
then shfrc is allocated, but never deallocated. Allocate was
replaced with a pointer to a local variable.
. Fixes to testing scripts for hopper and titan/jaguar
Wrong number of processors per node.
. Fix regular expression for parsing string values in Namelist.pm
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
M models/atm/cam/bld/configure
. Change how SPMD is set
M models/atm/cam/bld/perl5lib/Build/Namelist.pm
. Fix regular expression for parsing string values in Namelist.pm
A models/atm/cam/bld/user_nl_cam
. Empty user_nl_cam file that is used by new cam.cpl7.template
M models/atm/cam/bld/cam.cpl7.template
. Updated cam.cpl7.template
Describe any changes made to the namelist:
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. Update to newer soil_erod
Default file was updated, this produces answer changes
Add filenames for ne30, ne60, ne120, and ne240, this produces answer changes for homme
. cldfrc_permit being set for homme instead of default being used
. New ncdata added for ne30, ne120. Still need and ne60 ic.
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/tests_pretag_bluefire
M models/atm/cam/test/system/input_tests_master
. Add a CESM ne30 CAM5 test
M models/atm/cam/test/system/TCB_ccsm.sh
. Fix to work with latest cam.cpl7.template
M models/atm/cam/test/system/CAM_runcmnd.sh
. Set number of processors per node to 16 for hopper and titan/jaguarpf
M models/atm/cam/src/physics/cam/cam_diagnostics.F90
. Changed spelling of precipitatable to precipitable
M models/atm/cam/src/physics/cam/stratiform.F90
. Memory leak fix.
M models/atm/cam/src/physics/cam/modal_aer_opt.F90
. Change 4 to 2 in a formatted write
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
076 eq991 TEQ_ccsm.sh f19_f19 F f1.9c4m fcase 2d ..........................FAIL! rc= 7 at Thu Mar 15 22:40:00 MDT 2012
. Problems with clm namelist variable being changed from fatmgrid to fatmlndfrc
082 bl993 TBL_ccsm.sh ne30_g16 FC5 2d .....................................FAIL! rc= 5 at Thu Mar 15 23:57:26 MDT 2012
. Expected, this is a new test
edinburgh/lf95:
030 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Thu Mar 15 23:08:11 MDT 2012
. Expected, updated soil_erod file
edinburgh/pgi:
031 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Thu Mar 15 17:48:37 MDT 2012
. Expected, updated soil_erod file
059 bl734 TBL.sh hn16c5aqdm outfrq3s_bam+aquaplanet_cam5 9s ...............FAIL! rc= 7 at Thu Mar 15 20:11:09 MDT 2012
062 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ....................FAIL! rc= 7 at Thu Mar 15 21:50:02 MDT 2012
. Expected updated soil_erod file and cldfrc_permit being set for homme
Summarize any changes to answers, i.e.,
===============================================================
===============================================================
Tag name: cam5_1_23
Originator(s): Francis Vitt
Date: 12 Mar 2012
One-line Summary: Support for new CLM MEGAN VOC emissions
Purpose of changes:
* New functionality for CLM MEGAN VOC emissions
- flexibility in specifying VOC emissions from MEGAN in CLM
* New functionality for offline dyn driver
- ability added to turn off surface feedbacks to climate
met_srf_feedback (default is false)
* New functionality added for generating chem rate diagnostics
* Revived broken trop_mam7 and super_fast_llnl_mam3 chem packages
* Added new chem packages :
- trop_mozart_mam3
- trop_mozart_soa
- trop_strat_mam3
* Added ability to support chemistry species names up to 16 characters
* Maximum length for history increased to 24 characters
* Fixed bug in physics_buffer
* Fixed path issues in chem perl build modules for CESM scripts
* Chemistry preprocessor changes :
- supports chem species names up to 16 chars
- remove references to het specifications -- all species can have wet removal
- carbon mass of species output
- calculates molecular masses using double precision to get consistent masses wrt invoking
the preprocessor on different platforms and compilers
- new functionality for generating chem rate diagnostics
* Switched on TMS for SC and SD WACCM
* Misc cleanup in chemistry routines
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
Namelist group megan_emis_nl has been added to file drv_flds_in
Namelist variables:
megan_specifier, megan_mapped_emisfctrs, megan_factors_file
megan_specifier is a series of strings where each string contains one
CAM chemistry constituent name (left of = sign) and one or more MEGAN
compounds (seperated by + sign if more than one). The specification of
the MEGAN compounds to the right of the = signs tells the MEGAN VOC
model within CLM how to construct the VOC fluxes using the factors in
megan_factors_file and land surface state.
megan_factors_file read by CLM contains valid MEGAN compound names,
MEGAN class groupings and scalar emission factors
megan_mapped_emisfctrs switch is used to tell the MEGAN model to use
mapped emission factors read in from the CLM surface data input file
rather than the scalar factors from megan_factors_file
Example:
&megan_emis_nl
megan_specifier = 'ISOP = isoprene',
'C10H16 = myrcene + sabinene + limonene + carene_3 + ocimene_t_b + pinene_b + ...',
'CH3OH = methanol',
'C2H5OH = ethanol',
'CH2O = formaldehyde',
'CH3CHO = acetaldehyde',
...
megan_factors_file = '$datapath/megan_emis_factors.nc'
/
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Brian Eaton
List all subroutines eliminated:
D models/atm/cam/src/chemistry/modal_aero/mo_sethet.F90
- this separate version is no longer needed and is removed
D models/atm/cam/src/chemistry/pp_waccm_mozart/m_het_id.F90
D models/atm/cam/src/chemistry/pp_waccm_mozart_v1/m_het_id.F90
D models/atm/cam/src/chemistry/pp_trop_mozart/m_het_id.F90
D models/atm/cam/src/chemistry/pp_super_fast_llnl/m_het_id.F90
D models/atm/cam/src/chemistry/pp_trop_mam3/m_het_id.F90
D models/atm/cam/src/chemistry/pp_trop_mam7/m_het_id.F90
D models/atm/cam/src/chemistry/pp_trop_ghg/m_het_id.F90
- these are no longer used and are removed
List all subroutines added and what they do:
A models/atm/cam/src/chemistry/mozart/rate_diags.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart/mo_rxt_rates_conv.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_v1/mo_rxt_rates_conv.F90
A models/atm/cam/src/chemistry/pp_trop_mozart/mo_rxt_rates_conv.F90
A models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_rxt_rates_conv.F90
A models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_rxt_rates_conv.F90
A models/atm/cam/src/chemistry/pp_none/mo_rxt_rates_conv.F90
A models/atm/cam/src/chemistry/pp_trop_mam3/mo_rxt_rates_conv.F90
A models/atm/cam/src/chemistry/pp_trop_bam/mo_rxt_rates_conv.F90
A models/atm/cam/src/chemistry/pp_trop_mam7/mo_rxt_rates_conv.F90
A models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_rxt_rates_conv.F90
A models/atm/cam/src/chemistry/pp_trop_ghg/mo_rxt_rates_conv.F90
A models/atm/cam/src/chemistry/pp_waccm_ghg/mo_rxt_rates_conv.F90
- these are added to give the user the ability to output reaction rates to history
A models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_lu_solve.F90
A models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_setrxt.F90
A models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_adjrxt.F90
A models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_rxt_rates_conv.F90
A models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_imp_sol.F90
A models/atm/cam/src/chemistry/pp_trop_mozart_mam3/chem_mods.F90
A models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_prod_loss.F90
A models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_lin_matrix.F90
A models/atm/cam/src/chemistry/pp_trop_mozart_mam3/m_rxt_id.F90
A models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_phtadj.F90
A models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_nln_matrix.F90
A models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_lu_factor.F90
A models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_sim_dat.F90
A models/atm/cam/src/chemistry/pp_trop_mozart_mam3/mo_indprd.F90
A models/atm/cam/src/chemistry/pp_trop_mozart_mam3/m_spc_id.F90
A models/atm/cam/src/chemistry/pp_trop_mozart_mam3/chem_mech.doc
A models/atm/cam/src/chemistry/pp_trop_mozart_mam3/chem_mech.in
A models/atm/cam/src/chemistry/pp_trop_mozart_mam3
- new chemistry package added
A models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_lu_solve.F90
A models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_setrxt.F90
A models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_adjrxt.F90
A models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_rxt_rates_conv.F90
A models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_imp_sol.F90
A models/atm/cam/src/chemistry/pp_trop_mozart_soa/chem_mods.F90
A models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_prod_loss.F90
A models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_lin_matrix.F90
A models/atm/cam/src/chemistry/pp_trop_mozart_soa/m_rxt_id.F90
A models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_phtadj.F90
A models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_nln_matrix.F90
A models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_lu_factor.F90
A models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_sim_dat.F90
A models/atm/cam/src/chemistry/pp_trop_mozart_soa/mo_indprd.F90
A models/atm/cam/src/chemistry/pp_trop_mozart_soa/m_spc_id.F90
A models/atm/cam/src/chemistry/pp_trop_mozart_soa/chem_mech.doc
A models/atm/cam/src/chemistry/pp_trop_mozart_soa/chem_mech.in
A models/atm/cam/src/chemistry/pp_trop_mozart_soa
- new chemistry package added
A models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_lu_solve.F90
A models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_setrxt.F90
A models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_adjrxt.F90
A models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_rxt_rates_conv.F90
A models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_imp_sol.F90
A models/atm/cam/src/chemistry/pp_trop_strat_mam3/chem_mods.F90
A models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_prod_loss.F90
A models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_lin_matrix.F90
A models/atm/cam/src/chemistry/pp_trop_strat_mam3/m_rxt_id.F90
A models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_phtadj.F90
A models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_nln_matrix.F90
A models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_lu_factor.F90
A models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_sim_dat.F90
A models/atm/cam/src/chemistry/pp_trop_strat_mam3/mo_indprd.F90
A models/atm/cam/src/chemistry/pp_trop_strat_mam3/m_spc_id.F90
A models/atm/cam/src/chemistry/pp_trop_strat_mam3/chem_mech.doc
A models/atm/cam/src/chemistry/pp_trop_strat_mam3/chem_mech.in
A models/atm/cam/src/chemistry/pp_trop_strat_mam3
- new chemistry package added
A models/atm/cam/test/system/config_files/fsoa1.9c4dh
A models/atm/cam/test/system/config_files/fm1.9c5dh
A models/atm/cam/test/system/config_files/fst1.9c4dh
A models/atm/cam/test/system/config_files/f7mode4c5dm
A models/atm/cam/test/system/config_files/fsm4c5dm
A models/atm/cam/test/system/config_files/fst1.9c5dh
- new tests added
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/tests_pretag_bluefire
M models/atm/cam/test/system/test_driver.sh
M models/atm/cam/test/system/tests_pretag_edinburgh_pgi
M models/atm/cam/test/system/input_tests_master
- new tests added
M models/atm/cam/bld/config_files/definition.xml
M models/atm/cam/bld/configure
- new chem pkgs listed above
- clm_vocsrc congifure option removed (now run-time specification)
- removed DUST and PROGSSLT cppdefs which are no longer used
- corrected setting of chem_proc_src and chem_proc_src paths for CESM scripts
- added NO__MPIMOD and NO_SIZEOF cppdefs for use of new PIO on edinburgh
M models/atm/cam/bld/Makefile.in
- use NO_MPIMOD and NO_MPI2 when _MPISERIAL is used
- fixed LDFLAGS for lahey so that we can build on systems that do have the ../lf6481/lib64 path
the "-Wl,-rpath ..." flags do not seem to be needed
M models/atm/cam/bld/build-namelist
- removed the flds_voc namelist var (now handled in drv_flds_in namelist file)
- removed reference to the clm_vocsrc config setting (no longer applies)
the MEGAN VOC specifications should be done via build-namelist use cases
or user_nl_cam
- the set_dep_list perl routine needs chem_proc_src path to get correct species lists when
the chem preprocessor is invoked.
- added code for 7-mode modal aerosols
- mods for trop_mozart_mam3, trop_mozart_soa, trop_strat_mam3
- corrections to trop_mam7 settings
- include megan_emis_nl (for MEGAN VOC emissions from CLM) in comp_gruops
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- added defaults for new chemistry pkgs
- 7-mode mam phys props and optics files -- some of these defaults are not correct
- defaults added for trop_mam7 emissions files
- set drydep_method to xactive_atm for super_fast_llnl_mam3 and mozart_mam3 chem pckgs
- set dust_emis_fact to 0.35 for all cam5 configurations regardless of chemistry
- default megan factors file
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- moved met namelist opts for offline dyn driver from cam_inparm to metdata_nl
- added new options for offline dyn driver
- made fexcl, fincl and fwrtpr namelist type specifications consistent with array declarations in runtime_opts
- added namelist vars for specification of CLM MEGAN VOC emissions
- removed "flds_voc" namelist var
- for cplflds_custom namlist var seq_cplflds_inparm -> seq_cplflds_userspec
to support latest driver
M models/atm/cam/bld/namelist_files/master_gas_wetdep_list.xml
M models/atm/cam/bld/namelist_files/master_aer_wetdep_list.xml
M models/atm/cam/bld/namelist_files/master_drydep_list.xml
- added species for new trop_mozart_soa chem
M models/atm/cam/bld/namelist_files/use_cases/sd_waccm_tslt_geos5.xml
M models/atm/cam/bld/namelist_files/use_cases/sd_waccm_geos5.xml
- switched on TMS
- new wasolar forcing file which does not include 29 Feb 2100
M models/atm/cam/bld/namelist_files/use_cases/waccmx_2000_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1955-2005_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp26.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp45.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp85.xml
- new wasolar forcing file which does not include 29 Feb 2100
M models/atm/cam/bld/namelist_files/use_cases/1850_cam4_super_fast_llnl.xml
M models/atm/cam/bld/namelist_files/use_cases/1850-2005_cam4_super_fast_llnl.xml
M models/atm/cam/bld/namelist_files/use_cases/2000_cam4_super_fast_llnl.xml
- added megan specifer namelist
M models/atm/cam/bld/perl5lib/Build/ChemNamelist.pm
M models/atm/cam/bld/perl5lib/Build/ChemPreprocess.pm
- fixed a bug set_dep_lists which occurs when the chem preprocessor is used
- fixed chem_src_dir and chem_proc_src paths problem for CESM scripts
- removed setting DUST and PROGSSLT cppdefs
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
- updates to chemistry preprocessor
- supports chem species names upto 16 chars
- remove references to het specifications -- all species can have wet removal
- carbon mass of species output
- calc molecular masses using double precision to get consistent masses wrt invoking
the preprocessor on different platforms and compilers
- new functionality for generating chem rate diagnostics
M models/atm/cam/src/control/runtime_opts.F90
- use history fieldname_len declared in cam_history_support for consistency
- moved met namelist opts for offline dyn driver from cam_inparm to metdata_nl in metdata.F90
M models/atm/cam/src/control/cam_history_support.F90
- increased fieldname_len to support longer constituent names
M models/atm/cam/src/control/cam_comp.F90
M models/atm/cam/src/physics/cam/physpkg.F90
- moved invocation of offline dyn driver get_met subroutines to physpkg level
M models/atm/cam/src/cpl_esmf/atm_comp_esmf.F90
M models/atm/cam/src/cpl_share/cam_cpl_indices.F90
M models/atm/cam/src/cpl_mct/atm_comp_mct.F90
- changes for new MEGAN VOC surface fluxes
M models/atm/cam/src/physics/cam/microp_aero.F90
- changes to be able to run 7 mode version of MAM
- prog_modal_aero needs to be true for super_fast_llnl_mam3 and mozart_mam3
M models/atm/cam/src/physics/cam/phys_control.F90
- need to suppor longer chempkg names
- prog_modal_aero logical added
M models/atm/cam/src/physics/cam/modal_aer_opt.F90
- include hist fields for trop_mam7
M models/atm/cam/src/physics/cam/ndrop.F90
- get prog_modal_aero logical from phys_getopts
M models/atm/cam/src/physics/cam/physics_buffer.F90
M models/atm/cam/src/physics/cam/physics_buffer.F90.in
- changes to support more diminsions physics buffer arrays
M models/atm/cam/src/physics/cam/constituent_burden.F90
- increase burdennam length for longer constituent names
M models/atm/cam/src/dynamics/fv/dp_coupling.F90
M models/atm/cam/src/physics/waccmx/majorsp_diffusion.F90
- waccmx corrections from 1.0.4 release version
M models/atm/cam/src/chemistry/utils/modal_aero_deposition.F90
- changes needed to run trop_mam7
M models/atm/cam/src/chemistry/modal_aero/modal_aero_data.F90
- format change for readability
M models/atm/cam/src/chemistry/bulk_aero/aerosol_intr.F90
- set prog_modal_aero logical from phys_getopts
M models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
- changes needed to have modal aerosol with mozart chemistry
- removal of the obsolete hetcnt reference
M models/atm/cam/src/chemistry/mozart/mo_tracname.F90
- increased species name length (up to 16 chars)
M models/atm/cam/src/chemistry/mozart/mo_exp_sol.F90
- removal of the obsolete hetcnt reference
M models/atm/cam/src/chemistry/mozart/mo_cph.F90
- reaction tag names changed for the waccm chemical potential heating reactions
M models/atm/cam/src/chemistry/mozart/mo_chemini.F90
- pass pbuf pointer into sao_inti
M models/atm/cam/src/chemistry/mozart/mo_chm_diags.F90
- removed assumption made about the indexing of modal aerosols
- removal of the obsolete hetcnt reference
M models/atm/cam/src/chemistry/mozart/chemistry.F90
- prog_modal_aero needs to be true for super_fast_llnl_mam3 and mozart_mam3
- increase chem_name
- chem_reset_fluxes subroutine added to initialize chem srf emissions each time set
including MEGAN emissions from CLM
- soa_register added
M models/atm/cam/src/chemistry/mozart/mo_drydep.F90
- dry dep added for new SOA species
- allocate and initialize fraction_landuse array for modal aerosol dry dep
M models/atm/cam/src/chemistry/mozart/mo_sethet.F90
- changes needed for new chem pkgs (SOA)
- merged in modal version here
- removal of the obsolete hetcnt reference
M models/atm/cam/src/chemistry/mozart/mo_usrrxt.F90
- change tag names of waccm ion reactions
- incorporate modal aerosols
M models/atm/cam/src/chemistry/mozart/mo_setsoa.F90
- changes needed for new SOA chem pkg
M models/atm/cam/src/chemistry/mozart/mo_srf_emissions.F90
- removed the assumption of which species has emissions set elsewhere
-- now simply added what is specified in file to the surf fluxes
- does not set surf fluxes to zero
- cleaned up the MODAL ifdefs
M models/atm/cam/src/physics/cam/radheat.F90
M models/atm/cam/src/dynamics/fv/dyn_comp.F90
M models/atm/cam/src/dynamics/fv/metdata.F90
- changes for offline dyn driver new feature
M SVN_EXTERNAL_DIRECTORIES
- update driver to drvseq4_1_04 for MEGAN
- update clm to megan05_clm4_0_40 for MEGAN
- update pio to pio1_4_2 to support more dimensions in physics buffer
- scripts updated to scripts4_120307 to support new pio
- machines updated to Machines_120307 to support new pio
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/chem_mods.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/chem_mech.doc
M models/atm/cam/src/chemistry/pp_waccm_mozart/chem_mech.in
- change in cph tag names
- cleanup
- added capability to write reaction rates to history
M models/atm/cam/src/chemistry/pp_waccm_mozart_v1/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_v1/chem_mods.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_v1/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_v1/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_v1/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_v1/chem_mech.doc
M models/atm/cam/src/chemistry/pp_waccm_mozart_v1/chem_mech.in
- change in cph tag names
- cleanup
- added capability to write reaction rates to history
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/chem_mech.doc
- cleanup
- added capability to write reaction rates to history
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/chem_mech.doc
- cleanup
- added capability to write reaction rates to history
M models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/chem_mods.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/chem_mech.doc
- cleanup
- added capability to write reaction rates to history
M models/atm/cam/src/chemistry/pp_none/chemistry.F90
A models/atm/cam/src/chemistry/pp_none/mo_rxt_rates_conv.F90
M models/atm/cam/src/chemistry/pp_none/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_none/chem_mods.F90
A models/atm/cam/src/chemistry/pp_none/chem_mech.in
M models/atm/cam/src/chemistry/pp_none/mo_lin_matrix.F90
- cleanup
- added capability to write reaction rates to history
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/chem_mech.doc
- cleanup
- added capability to write reaction rates to history
D models/atm/cam/src/chemistry/pp_trop_bam/m_het_id.F90
A models/atm/cam/src/chemistry/pp_trop_bam/mo_rxt_rates_conv.F90
M models/atm/cam/src/chemistry/pp_trop_bam/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_bam/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_bam/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_bam/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_bam/chem_mech.doc
- cleanup
- added capability to write reaction rates to history
M models/atm/cam/src/chemistry/pp_trop_mam7/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/chem_mech.doc
- cleanup
- added capability to write reaction rates to history
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/chem_mods.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/chem_mech.doc
- cleanup
- added capability to write reaction rates to history
D models/atm/cam/src/chemistry/pp_trop_ghg/m_het_id.F90
A models/atm/cam/src/chemistry/pp_trop_ghg/mo_rxt_rates_conv.F90
M models/atm/cam/src/chemistry/pp_trop_ghg/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_ghg/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_ghg/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_ghg/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_ghg/chem_mech.doc
- cleanup
- added capability to write reaction rates to history
M models/atm/cam/src/chemistry/pp_waccm_ghg/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_waccm_ghg/chem_mods.F90
M models/atm/cam/src/chemistry/pp_waccm_ghg/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_waccm_ghg/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_waccm_ghg/chem_mech.doc
- cleanup
- added capability to write reaction rates to history
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
050 bl374 TBL.sh fs1.9c4dh outfrq3s+1850-2005_cam4_super_fast_llnl 9s .....FAIL! rc= 7 at Mon Mar 12 02:13:11 MDT 2012
srf emis for ISOP has changed (new MEGAN emis) thus this is expected to fail
052 bl415 TBL.sh fst1.9c5dh outfrq3s 9s ...................................FAIL! rc= 5 at Mon Mar 12 02:13:12 MDT 2012
054 bl430 TBL.sh fm1.9c5dh outfrq3s 9s ....................................FAIL! rc= 5 at Mon Mar 12 02:13:12 MDT 2012
057 bl440 TBL.sh fsoa1.9c4dh outfrq3s 9s ..................................FAIL! rc= 5 at Mon Mar 12 02:13:13 MDT 2012
these baseline tests for new chem pckgs are expected to fail
076 eq991 TEQ_ccsm.sh f19_f19 F f1.9c4m fcase 2d ..........................FAIL! rc= 7 at Mon Mar 12 03:13:02 MDT 2012
079 bl992 TBL_ccsm.sh f19_g16 E 2d ........................................FAIL! rc= 7 at Mon Mar 12 03:40:51 MDT 2012
these failures are do to update to CLM which caused namelist inconsistencies
edinburgh/lf95:
035 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7 at Mon Mar 12 06:26:41 MDT 2012
expected to fail due to TMS switch on
edinburgh/pgi:
036 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7 at Sun Mar 11 20:43:34 MDT 2012
expected to fail due to TMS switch on
056 bl420 TBL.sh f7mode4c5dm outfrq3s 9s ..................................FAIL! rc= 5 at Sun Mar 11 22:31:14 MDT 2012
new test expected to fail
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_1_22
Originator(s): fischer
Date: 2-28-2012
One-line Summary: Fixes for cam template
Purpose of changes:
. cam.cpl7.template has two bug fixes in it, this only affects
running with the cesm scripts
- Incorrect number of tasks was being sent to build-namelist
when mutliple instances of cam is running.
- drv_flds_in needed to be copied to the cesm run directory.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/cam.cpl7.template
. diveded ntasks my the number of atm instances
. copy drv_flds_in from Buildconf/camconf to the run directory
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
edinburgh/lf95:
edinburgh/pgi or jaguar/pgi:
Summarize any changes to answers: all bfb
===============================================================
===============================================================
Tag name: cam5_1_21
Originator(s): fischer, eaton, mvertens
Date: Feb 23 07:50:11 MST 2012
One-line Summary: namelist changes, CAM5 RCP and homme fixes, COSP restart
Purpose of changes:
. Namelist changes to support user_nl_comp and preview_namelist scripts
changes in cesm.
. Bug fix for old formatted fields in rad_diag for CAM5 RCP use_cases
. Fixes to get CAM5 homme with chemistry working out of box
. Fix added so COSP branch runs will work when using restart files from
a non COSP run.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
M cam/bld/configure
. Removed -DSPMD for cesm
M cam/bld/build-namelist
. Fixes so aerosol deposition datasets aren't added to the namelist
when running in aqua planet.
. Removed $pathname = strip_rootdir... so cesm user_nl_comp and
preview_namelist work correctly
M cam/bld/cam.cpl7.template
. Changes so cesm user_nl_comp and preview_namelist work correctly.
. -DSMPD is set here
M models/atm/cam/bld/perl5lib/Build/ChemPreprocess.pm
. $CODEROOT dir needed to be resolved for cesm runs
Describe any changes made to the namelist:
M cam/bld/namelist_files/use_cases/2006-2100_cam5_rcp60.xml
M cam/bld/namelist_files/use_cases/2006-2100_cam5_rcp26.xml
M cam/bld/namelist_files/use_cases/2006-2100_cam5_rcp45.xml
M cam/bld/namelist_files/use_cases/2006-2100_cam5_rcp85.xml
. Fields in rad_diag needed to be updated to the new format
M cam/bld/namelist_files/namelist_defaults_cam.xml
. drydep_srf_file fix for homme/cam5
M cam/bld/namelist_files/namelist_definition.xml
. Path for drydep_srf_file was not resolving because input_pathname wasn't being set.
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
A cam/test/system/config_files/s64c5dh
A cam/test/system/config_files/s64adh
A cam/test/system/config_files/s64idh
A cam/test/system/config_files/s64c5paqdh
. config files needed for s64 tests
A cam/test/system/config_files/h16c5aqdm
M cam/test/system/tests_pretag_edinburgh_pgi
M cam/test/system/input_tests_master
. Add new homme test with chemistry
M models/atm/cam/test/system/nl_files/user_nl_cam
M models/atm/cam/test/system/TCB_ccsm.sh
. Changes so cesm user_nl_comp and preview_namelist work in testing.
M models/atm/cam/src/physics/cam/restart_physics.F90
. Added test for cossp_cnt_init, if missing then cosp_cnt_init is set to 0.
This allows a COSP branch run from run that didn't have COSP turned on.
M SVN_EXTERNAL_DIRECTORIES
. update tags for cesm user_nl_comp and preview_namelist.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: all passed
edinburgh/lf95: all passed
edinburgh/pgi or jaguar/pgi:
059 bl735 TBL.sh h16c5aqdm outfrq3s+aquaplanet_cam5 9s ....................FAIL! rc= 5 at Wed Feb 22 15:08:45 MST 2012
expected, new test
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: bfb
===============================================================
===============================================================
Tag name: cam5_1_20
Originator(s): fischer, eaton, craig, mvertens, edwards
Date: Feb 7, 2012
One-line Summary: coupler fields and esmf interface updates, remove csim4
Purpose of changes:
. Update ESMF interfaces for ESMF ver 5.2.0, and update coupler filed names.
Merged with branch_tags/e52r_tags/e52r02_newcplflds03_cam5_1_16
. Remove csim4
. merge pbuf changes from branch_tags/spcam_cam5_1_17_tags/spcam01_cam5_1_17
. Update svn externals
. Add r8 tests for dynamics/homme, skip dynamics/homme/share
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
M models/atm/cam/bld/config_files/definition.xml
. remove csim4
M models/atm/cam/bld/configure
. Refactor configure to have just one method for writing the Filepath file
. csim4 removed
Describe any changes made to the namelist:
M models/atm/cam/bld/build-namelist
. add seq_cplflds_inparm group to the driver namelist file
. set flds_voc = .false.
flds_co2a = .true.
flds_co2b = .false.
flds_co2c = .false.
flds_co2_dmsa = .false.
cplflds_inparm = ""
. set do_rtm to false in the clm namelist file
M models/atm/cam/bld/namelist_files/namelist_definition.xml
. add flds_voc, flds_co2a, flds_co2b, flds_co2c, flds_co2_dmsa,
cplflds_inparm to the driver namelist file (drv_in)
. add do_rtm so river run off can be turned off in the clm namelist file
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
D models/atm/cam/src/utils/cam_csim4
. Removed csim4
D models/atm/cam/src/physics/cam/advnce.F90
. merge in pbuf changes from branch_tags/spcam_cam5_1_17_tags/spcam01_cam5_1_17
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/input_tests_master
. Add s64 tests
. e48c5m test removed
M models/atm/cam/src/physics/cam/physpkg.F90
M models/atm/cam/src/physics/cam/phys_debug.F90
M models/atm/cam/src/physics/cam/eddy_diff.F90
M models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
M models/atm/cam/src/chemistry/mozart/chemistry.F90
. merge in pbuf changes from branch_tags/spcam_cam5_1_17_tags/spcam01_cam5_1_17
M models/atm/cam/test/system/config_files/f10c3dm
. replace csim4 with cice
M models/atm/cam/test/system/tests_posttag_bluefire
. remove s32 tests, add s64 tests
M models/atm/cam/test/system/TR8.sh
. Add R8 test from dynamics/homme, skip dynamics/homme/share
M models/atm/cam/test/system/tests_pretag_edinburgh_lahey
M models/atm/cam/test/system/tests_pretag_edinburgh_pgi
. Move r8 test from lahey test list to pgi test list
. Add f10c3dm test for pgi
M models/atm/cam/src/dynamics/homme/nctopo_util_mod.F90
. fixed end module statement for the R8 test
M models/atm/cam/src/dynamics/homme/dyn_grid.F90
M models/atm/cam/src/dynamics/homme/native_mapping.F90
M models/atm/cam/src/dynamics/homme/inidat.F90
M models/atm/cam/src/dynamics/homme/stepon.F90
. insterted missing _r8s.
M models/atm/cam/bld/cam.cpl7.template
. USE_MPISERIAL test changed to MPILIB == 'mpi-serial'
. extra echo removed
M models/atm/cam/src/cpl_share/cam_cpl_indices.F90
M models/atm/cam/src/chemistry/mozart/mo_srf_emissions.F90
. couple field name changes
M models/atm/cam/src/utils/time_manager.F90
M models/atm/cam/src/utils/cam_dom/ocn_time_manager.F90
M models/atm/cam/src/cpl_esmf/atm_comp_mct.F90
. ESMF interface update to 5.2.0
M models/atm/cam/src/cpl_mct/atm_comp_mct.F90
M models/atm/cam/src/cpl_esmf/atm_comp_esmf.F90
. couple field name changes
. ESMF interface update to 5.2.0
M models/atm/cam/src/utils/cam_dom/cpl_mct/ocn_comp_mct.F90
M models/atm/cam/src/utils/cam_aqua/cpl_mct/ocn_comp_mct.F90
. New fields are being passed to ocn_final_mct from the coupler.
M models/atm/cam/src/physics/cam/convect_shallow.F90
. Bug fix for 6 outfld calls that declared the 1st dimension of the
output array to be size pcols, but then passed an array section of
dimensioned ncol.
M SVN_EXTERNAL_DIRECTORIES
. Update externals
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
063 bl731 TBL.sh h16c3aqdh aqua+aquaplanet_cam3 9s ........................FAIL! rc= 7 at Thu Feb 2 16:12:20 MST 2012
066 bl751 TBL.sh h30c4aqdm outfrq3s+aquaplanet_cam4 9s ....................FAIL! rc= 7 at Thu Feb 2 16:12:53 MST 2012
expected dynamics/homme missing r8s inserted
edinburgh/lf95: all passed
edinburgh/pgi:
039 bl315 TBL.sh f10c3dm outfrq3s+cam3 9s .................................FAIL! rc= 7 at Tue Jan 31 17:40:58 MST 2012
csim4 replaced with cice
056 bl734 TBL.sh hn16c5aqdm outfrq3s_bam+aquaplanet_cam5 9s ...............FAIL! rc= 7 at Tue Jan 31 19:11:09 MST 2012
expected dynamics/homme missing r8s inserted
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers:
HOMME will have answer changes due to missing r8s being insterted in
/dynamics/homme, and not the share directory
CSIM4 replace with CICE in tests caused answer changes.
===============================================================
===============================================================
Tag name: cam5_1_19
Originator(s): Jim Edwards, Jen Kay
Date: 01-20-2012
One-line Summary: Intel compiler port, Make restarts double, COSP bug fix
Purpose of changes: Intel compiler port, Make restarts double, COSP bug fix
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
models/atm/cam/test/system/test_driver.sh
Adjusted modules used on lynx
models/atm/cam/test/system/tests_pretag_edinburgh_pgi
models/atm/cam/test/system/input_tests_master
Added a test 321 for cosp atrain
models/atm/cam/src/control/cam_history.F90
Make sure that restart files are always douple precision
models/atm/cam/src/physics/cam/cldwat2m_macro.F90
models/atm/cam/src/physics/cam/uwshcu.F90
models/atm/cam/src/physics/cam/vertical_diffusion.F90
models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
An assumption was made about the location of cldliq and cldice fields
in the constituents array. This assumption was incorrect and was removed.
models/atm/cam/src/physics/cam/cospsimulator_intr.F90
1) addition of allocate statement circa line 526.
2) match atrain file longitude range to model
models/atm/cam/src/dynamics/eul/inidat.F90
models/atm/cam/src/dynamics/eul/prognostics.F90
models/atm/cam/src/dynamics/eul/stepon.F90
Change subroutine calls so that intel compiler doesn't complain when endlat 0)
- redo ndrop interfaces to pass cam types as per physics interface
- replace local calc of pi with physconst version, and replace hardcoded
value of sqrt(2*pi) by computed value.
. implement rad_constituent interfaces
. add logic so that tendencies are only computed for prognostic MAM
. remove the MODAL_AERO ifdef except around the code that does the tendency
calculations.
. rework local arrays that were declared with a pcnst dimension, but only
needed storage for the modal number densities and species mmr. Use a
local indexing scheme for those constituents.
models/atm/cam/src/physics/cam/phys_prop.F90
. add fields for the modal optics and properties to the physprop_type
structure.
. modify logic to read physprop files for modes as well as for bulk
aerosols.
. move the reading of bulk aerosol properties into separate routine and
only call when appropriate.
. add new method for reading modal physprop files
models/atm/cam/src/physics/cam/physpkg.F90
. remove unused pbuf and phys_state args from rad_cnst_init call
. replace phys_state arg by pbuf in prognostic_aerosol_initialize call
. remove some unnecessary array bounds and just pass entire phys_state array
models/atm/cam/src/physics/cam/rad_constituents.F90
. extensions to support modal aerosols
- a new namelist variable, mode_defs, was added to supply metadata
defining the modes.
- add parsing for new mode_defs variable
- extend parsing of rad_climate and rad_diag_* variables to include
modes.
- create new internal data structures for the mode definitions and for
mode lists which are analogous to the gas and bulk aerosol lists.
- move the initialization of data structures that contain information
specified in the namelist up into rad_cnst_readnl so that this info is
available elsewhere via the rad_cnst_get_info method as early in the
initialization process as possible.
- generic method rad_cnst_get_info has been extended to return info about
the mode definitions
- rad_cnst_get_aer_mmr and rad_cnst_get_aer_props have been extended to
return mixing ratios and props for the mode species
- rad_cnst_get_mode_num and rad_cnst_get_mode_props have been added to
provide the number mixing ratio and properties that are specific to
modes.
- removed some optional args from the get_info method for data that
should remain private in the module.
. rad_cnst_init and init_lists -- remove unused args pbuf & phys_state
models/atm/cam/src/physics/cam/radiation_data.F90
. use rad_constituent interfaces to access mixing ratio data -- old code
was accessing metadata which is meant to be private in the
rad_constituents module by extending rad_cnst_get_info, then using this
metadata to directly access the mixing ratios in the state and pbuf
structures. The rad_cnst_get_gas and rad_cnst_get_aer_mmr interfaces are
supposed to be used for this.
models/atm/cam/src/physics/cam/stratiform.F90
. remove unused cam5 macro/micro physics code
models/atm/cam/src/physics/cam/tphysac.F90
. fix aerosol_emis_intr interface to make ifdef MODAL_AERO unnecessary
models/atm/cam/src/physics/cam/vertical_diffusion.F90
. move the initialization of the physics buffer fields tke and kvh into the
vertical diffusion init method.
models/atm/cam/src/physics/rrtmg/oldcloud.F90
. cleanup -- remove some unused code
models/atm/cam/src/physics/rrtmg/radiation.F90
. move the add_default calls for radiation diagnostics inside the loop over
diagnostic calculations so that when diagnostic calculations are
requested the output from those calculations is added to the history file.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Fri Oct 28 16:05:11 MDT 2011
013 bl330 TBL.sh f4c5paqdh aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Fri Oct 28 16:09:44 MDT 2011
021 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...............................FAIL! rc= 7 at Fri Oct 28 16:28:12 MDT 2011
036 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ......................................FAIL! rc= 7 at Fri Oct 28 16:59:22 MDT 2011
040 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .....................................FAIL! rc= 7 at Fri Oct 28 17:16:34 MDT 2011
047 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...........................FAIL! rc= 7 at Fri Oct 28 18:00:24 MDT 2011
All failures are expected baseline comparison failures.
edinburgh/lf95: All PASS except:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Fri Oct 28 15:36:52 MDT 2011
009 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Fri Oct 28 15:51:40 MDT 2011
011 bl113 TBL.sh e8c5paqdm aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Fri Oct 28 16:03:04 MDT 2011
015 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Fri Oct 28 16:07:47 MDT 2011
022 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Fri Oct 28 16:37:10 MDT 2011
030 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Fri Oct 28 19:09:29 MDT 2011
040 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 7 at Fri Oct 28 19:51:59 MDT 2011
All failures are expected baseline comparison failures.
There is a bug in PGI compiler versions earlier than 11.0 which prevent it
from compiling the new rad_constituents.F90 file. There is no obvious
workaround for this bug, so until edinburgh gets an updated pgi the testing
will be done on one of the cray platforms.
hopper/pgi-11.7: All PASS except:
005 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Fri Oct 28 14:33:20 PDT 2011
010 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Fri Oct 28 14:35:05 PDT 2011
012 bl113 TBL.sh e8c5paqdm aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Fri Oct 28 14:35:06 PDT 2011
016 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Fri Oct 28 14:35:14 PDT 2011
023 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Fri Oct 28 14:35:19 PDT 2011
031 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Fri Oct 28 14:35:39 PDT 2011
041 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 7 at Fri Oct 28 14:35:41 PDT 2011
044 bl317 TBL.sh f10c5cdm outfrq3s 9s .....................................FAIL! rc= 7 at Fri Oct 28 14:35:47 PDT 2011
047 bl319 TBL.sh fn10c5dm outfrq3s_bam 9s .................................FAIL! rc= 5 at Fri Oct 28 14:35:47 PDT 2011
050 bl320 TBL.sh fn10c5dm rad_diag 9s .....................................FAIL! rc= 7 at Fri Oct 28 14:35:50 PDT 2011
053 bl734 TBL.sh hn16c5aqdm outfrq3s_bam+aquaplanet_cam5 9s ...............FAIL! rc= 5 at Fri Oct 28 14:35:50 PDT 2011
These baseline failures are all expected due to the different answers from
cam5 w/ trop_mam3. Two tests, bl319 and bl734 failed due the test
definitions changing and the new test wasn't defined for the baseline. But
when the new test is added to the baseline cam5_1_11, then the failures in
those tests, as in bl320, is due to a roundoff diff in the diagnostic field
CCN3. The simulations are actually identical.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: The answers are bit for bit except for
configurations with cam5 physics and trop_mam3 chemistry. In that case a
single precision size roundoff change was introduced, and has been
validated by the climate runs below.
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced:
Runs done on bluefire using cesm scripts.
control:
$SVNREPO/cam1/branch_tags/r8_cam5_1_03_tags/r8b_cam5_1_03/
create_newcase -compset F1850C5 -res f19_f19 -mach bluefire -case /blhome/fischer/runs/cam5_1_03_1850_r8
experiment:
$SVNREPO/cam1/branch_tags/radcnst_cam5_1_11_tags/radcnst06_cam5_1_11/
create_newcase -compset F1850C5 -res f19_f19 -mach bluefire -case /glade/home/fischer/runs/cam5_1_11_radcnst06
MSS location of control simulations used to validate new climate:
control: /FISCHER/csm/cam5_1_03_1850_r8
experiment: /home/fischer/csm/cam5_1_11_radcnst06
URL for AMWG diagnostics output used to validate new climate:
http://www.cgd.ucar.edu/cms/fischer/cam5/cam5_1_11_radcnst06-cam5_1_03_1850_r8
===============================================================
===============================================================
Tag name: cam5_1_11
Originator(s): fischer
Date: Oct 11 2011
One-line Summary: add missing explicit r8 kind, fix TKE and CUSH init
Purpose of changes:
. Add testing for missing explicit r8 kind
. Add missing explicit r8 kind to F90 code
. Didn't add missing explicit r8 kind or tests for external code
. Fix initialization bug (#1378) for TKE and CUSH.
Bugs fixed (include bugzilla ID): #1378
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: fischer
List all subroutines eliminated:
D models/atm/cam/src/physics/cam/miesubs.F
. not being used
List all subroutines added and what they do:
A models/atm/cam/test/system/TR8.sh
. test for missing explicit r8 kind
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/test_driver.sh
M models/atm/cam/test/system/tests_pretag_edinburgh_pgi
M models/atm/cam/test/system/input_tests_master
. add test for missing explicit r8 kind
M models/atm/cam/src/physics/cam/convect_shallow.F90
. remove initialization of TKE and CUSH (bug #1378)
. add missing explicit r8 kind
M models/atm/cam/src/control/sat_hist.F90
M models/atm/cam/src/control/interpolate_data.F90
M models/atm/cam/src/control/scamMod.F90
M models/atm/cam/src/utils/time_manager.F90
M models/atm/cam/src/utils/cam_csim4/ice_time_manager.F90
M models/atm/cam/src/utils/cam_dom/ocn_time_manager.F90
M models/atm/cam/src/utils/pilgrim/mod_comm.F90
M models/atm/cam/src/utils/pilgrim/unit_testers/ghosttest.F90
M models/atm/cam/src/utils/pilgrim/unit_testers/redistributetest.F90
M models/atm/cam/src/utils/pilgrim/unit_testers/decomptest.F90
M models/atm/cam/src/utils/pilgrim/unit_testers/parutilitiestest.F90
M models/atm/cam/src/utils/pilgrim/unit_testers/parpatterntest.F90
M models/atm/cam/src/utils/repro_sum_mod.F90
M models/atm/cam/src/physics/cam/cam3_aero_data.F90
M models/atm/cam/src/physics/cam/microp_aero.F90
M models/atm/cam/src/physics/cam/tropopause.F90
M models/atm/cam/src/physics/cam/rayleigh_friction.F90
M models/atm/cam/src/physics/cam/flux_avg.F90
M models/atm/cam/src/physics/cam/aer_rad_props.F90
M models/atm/cam/src/physics/cam/radsw.F90
M models/atm/cam/src/physics/cam/phys_prop.F90
M models/atm/cam/src/physics/cam/hirsbt.f90
D models/atm/cam/src/physics/cam/miesubs.F
M models/atm/cam/src/physics/cam/sslt_rebin.F90
M models/atm/cam/src/physics/cam/modal_aer_opt.F90
M models/atm/cam/src/physics/cam/physics_types.F90
M models/atm/cam/src/physics/cam/radlw.F90
M models/atm/cam/src/physics/cam/hb_diff.F90
M models/atm/cam/src/physics/cam/ndrop.F90
M models/atm/cam/src/physics/cam/physpkg.F90
M models/atm/cam/src/physics/cam/phys_debug_util.F90
M models/atm/cam/src/physics/cam/rad_constituents.F90
M models/atm/cam/src/physics/cam/mcshallow.F90
M models/atm/cam/src/physics/cam/cldwat2m_macro.F90
M models/atm/cam/src/physics/cam/convect_deep.F90
M models/atm/cam/src/physics/cam/cospsimulator_intr.F90
M models/atm/cam/src/physics/cam/wv_saturation.F90
M models/atm/cam/src/physics/cam/cldwat2m_micro.F90
M models/atm/cam/src/physics/cam/rad_solar_var.F90
M models/atm/cam/src/physics/cam/phys_grid.F90
M models/atm/cam/src/physics/cam/phys_debug.F90
M models/atm/cam/src/physics/cam/zm_conv_intr.F90
M models/atm/cam/src/physics/cam/zm_conv.F90
M models/atm/cam/src/physics/cam/eddy_diff.F90
M models/atm/cam/src/physics/cam/cldwat.F90
M models/atm/cam/src/physics/cam/uwshcu.F90
M models/atm/cam/src/physics/cam/vertical_diffusion.F90
M models/atm/cam/src/physics/cam/uw_conv.F90
M models/atm/cam/src/physics/waccm/gw_drag.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_v1/mo_photo.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_v1/mo_sad.F90
M models/atm/cam/src/chemistry/utils/mo_constants.F90
M models/atm/cam/src/chemistry/utils/mo_util.F90
M models/atm/cam/src/chemistry/utils/tracer_data.F90
M models/atm/cam/src/chemistry/utils/solar_data.F90
M models/atm/cam/src/chemistry/utils/aircraft_emit.F90
M models/atm/cam/src/chemistry/modal_aero/mo_sethet.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_rename.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_coag.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_data.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_gasaerexch.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_initialize_data.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_wateruptake.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_calcsize.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_newnuc.F90
M models/atm/cam/src/chemistry/bulk_aero/progseasalts_intr.F90
M models/atm/cam/src/chemistry/bulk_aero/dust_intr.F90
M models/atm/cam/src/chemistry/bulk_aero/aerosol_intr.F90
M models/atm/cam/src/chemistry/mozart/mo_ghg_chem.F90
M models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
M models/atm/cam/src/chemistry/mozart/mo_airmas.F90
M models/atm/cam/src/chemistry/mozart/mo_trislv.F90
M models/atm/cam/src/chemistry/mozart/mo_lightning.F90
M models/atm/cam/src/chemistry/mozart/lin_strat_chem.F90
M models/atm/cam/src/chemistry/mozart/mo_schu.F90
M models/atm/cam/src/chemistry/mozart/mo_lymana.F90
M models/atm/cam/src/chemistry/mozart/mo_calcoe.F90
M models/atm/cam/src/chemistry/mozart/mo_aerosols.F90
M models/atm/cam/src/chemistry/mozart/mo_airplane.F90
M models/atm/cam/src/chemistry/mozart/gas_wetdep_opts.F90
M models/atm/cam/src/chemistry/mozart/mo_ps2str.F90
M models/atm/cam/src/chemistry/mozart/mo_fstrat.F90
M models/atm/cam/src/chemistry/mozart/mo_sphers.F90
M models/atm/cam/src/chemistry/mozart/mo_sad.F90
M models/atm/cam/src/chemistry/mozart/mo_setozo.F90
M models/atm/cam/src/chemistry/mozart/mo_drydep.F90
M models/atm/cam/src/chemistry/mozart/mo_photo.F90
M models/atm/cam/src/chemistry/mozart/cfc11star.F90
M models/atm/cam/src/chemistry/mozart/mo_sethet.F90
M models/atm/cam/src/chemistry/mozart/mo_setsoa.F90
M models/atm/cam/src/chemistry/mozart/mz_aerosols_intr.F90
M models/atm/cam/src/chemistry/mozart/mo_neu_wetdep.F90
M models/atm/cam/src/chemistry/mozart/mo_strato_sad.F90
M models/atm/cam/src/chemistry/mozart/mo_usrrxt.F90
M models/atm/cam/src/chemistry/mozart/mo_setcld.F90
M models/atm/cam/src/chemistry/mozart/mo_xsections.F90
M models/atm/cam/src/chemistry/mozart/mo_inter.F90
M models/atm/cam/src/chemistry/mozart/mo_synoz.F90
M models/atm/cam/src/chemistry/mozart/mo_srf_emissions.F90
M models/atm/cam/src/chemistry/mozart/mo_setsox.F90
M models/atm/cam/src/dynamics/sld/dyn_grid.F90
M models/atm/cam/src/dynamics/sld/sld_control_mod.F90
M models/atm/cam/src/dynamics/eul/eul_control_mod.F90
M models/atm/cam/src/dynamics/eul/dyn_grid.F90
M models/atm/cam/src/dynamics/eul/forecast.F90
M models/atm/cam/src/dynamics/eul/inidat.F90
M models/atm/cam/src/dynamics/eul/getinterpnetcdfdata.F90
M models/atm/cam/src/dynamics/eul/iop.F90
M models/atm/cam/src/dynamics/eul/stepon.F90
M models/atm/cam/src/dynamics/fv/dyn_grid.F90
M models/atm/cam/src/dynamics/fv/dryairm.F90
M models/atm/cam/src/dynamics/fv/dyn_comp.F90
M models/atm/cam/src/dynamics/fv/ctem.F90
. add missing explicit r8 kind
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
004 bl132 TBL.sh e48c4dh ghgrmp+1850_cam4 9s ..............................FAIL! rc= 7 at Mon Oct 10 09:50:55 MDT 2011
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Mon Oct 10 09:51:03 MDT 2011
013 bl330 TBL.sh f4c5paqdh aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Mon Oct 10 09:51:05 MDT 2011
017 bl331 TBL.sh f4c4dh co2rmp+1850_cam4 9s ...............................FAIL! rc= 7 at Mon Oct 10 09:51:14 MDT 2011
021 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...............................FAIL! rc= 7 at Mon Oct 10 09:51:26 MDT 2011
033 bl336 TBL.sh wm4c4h outfrq3s+waccm_1850_cam4 9s .......................FAIL! rc= 7 at Mon Oct 10 09:52:15 MDT 2011
036 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ......................................FAIL! rc= 7 at Mon Oct 10 09:52:22 MDT 2011
040 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .....................................FAIL! rc= 7 at Mon Oct 10 09:52:47 MDT 2011
045 bl373 TBL.sh fm1.9c4dh outfrq3s 9s ....................................FAIL! rc= 7 at Mon Oct 10 09:53:35 MDT 2011
047 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...........................FAIL! rc= 7 at Mon Oct 10 09:53:54 MDT 2011
050 bl374 TBL.sh fs1.9c4dh outfrq3s+1850-2005_cam4_super_fast_llnl 9s .....FAIL! rc= 7 at Mon Oct 10 09:54:19 MDT 2011
054 bl387 TBL.sh f1.9c4dm outfrq3s+1850-2005_cam4 9s ......................FAIL! rc= 7 at Mon Oct 10 09:54:32 MDT 2011
057 bl388 TBL.sh f1.9c4m outfrq24h+1850_cam4 2d ...........................FAIL! rc= 7 at Mon Oct 10 09:54:42 MDT 2011
060 bl379 TBL.sh wm1.9c4h outfrq24h+waccm_1850_cam4 2d ....................FAIL! rc= 7 at Mon Oct 10 10:08:24 MDT 2011
072 bl992 TBL_ccsm.sh f19_g16 E 2d ........................................FAIL! rc= 7 at Mon Oct 10 11:27:41 MDT 2011
edinburgh/lf95:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Fri Oct 7 10:05:05 MDT 2011
009 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 5 at Fri Oct 7 10:23:38 MDT 2011
011 bl113 TBL.sh e8c5paqdm aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Fri Oct 7 10:41:40 MDT 2011
015 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 5 at Fri Oct 7 10:47:03 MDT 2011
018 bl115 TBL.sh e8idm idphys 9s ..........................................FAIL! rc= 7 at Fri Oct 7 10:55:59 MDT 2011
022 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Fri Oct 7 11:30:05 MDT 2011
030 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Fri Oct 7 14:53:54 MDT 2011
032 bl313 TBL.sh f10c3aqdm outfrq3s+aquaplanet_cam3 3s ....................FAIL! rc= 7 at Fri Oct 7 15:05:18 MDT 2011
035 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7 at Fri Oct 7 15:38:46 MDT 2011
038 bl315 TBL.sh f10c3dm outfrq3s+cam3 9s .................................FAIL! rc= 7 at Fri Oct 7 15:52:18 MDT 2011
040 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 7 at Fri Oct 7 16:16:54 MDT 2011
043 bl318 TBL.sh f10c4cdm sat_hist 9s .....................................FAIL! rc= 7 at Fri Oct 7 16:40:26 MDT 2011
edinburgh/pgi:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Fri Oct 7 09:50:03 MDT 2011
009 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 5 at Fri Oct 7 09:57:42 MDT 2011
011 bl113 TBL.sh e8c5paqdm aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 5 at Fri Oct 7 10:01:33 MDT 2011
015 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 5 at Fri Oct 7 10:04:17 MDT 2011
018 bl115 TBL.sh e8idm idphys 9s ..........................................FAIL! rc= 7 at Fri Oct 7 10:08:53 MDT 2011
022 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Fri Oct 7 10:19:37 MDT 2011
030 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Fri Oct 7 10:49:54 MDT 2011
032 bl313 TBL.sh f10c3aqdm outfrq3s+aquaplanet_cam3 3s ....................FAIL! rc= 7 at Fri Oct 7 10:55:05 MDT 2011
035 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7 at Fri Oct 7 11:03:37 MDT 2011
038 bl315 TBL.sh f10c3dm outfrq3s+cam3 9s .................................FAIL! rc= 7 at Fri Oct 7 11:09:38 MDT 2011
040 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 7 at Fri Oct 7 11:15:43 MDT 2011
043 bl317 TBL.sh f10c5cdm outfrq3s 9s .....................................FAIL! rc= 7 at Fri Oct 7 11:26:42 MDT 2011
046 bl319 TBL.sh fn10c5dm outfrq3s 9s .....................................FAIL! rc= 7 at Fri Oct 7 11:34:07 MDT 2011
049 bl320 TBL.sh fn10c5dm rad_diag 9s .....................................FAIL! rc= 7 at Fri Oct 7 11:37:14 MDT 2011
052 bl734 TBL.sh hn16c5aqdm outfrq3s+aquaplanet_cam5 9s ...................FAIL! rc= 7 at Fri Oct 7 12:06:39 MDT 2011
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.: Adding in the missing explicit r8 kind caused greater than
roundoff error. But climate is still preserved.
Runs done on bluefire using cesm scripts.
control: $SVNREPO/cam1/trunk_tags/cam5_1_03/
create_newcase -compset F1850C5 -res f19_f19 -mach bluefire -case /blhome/fischer/runs/cam5_1_03_1850_OOB
create_newcase -mach bluefire -res f19_g16 -compset FMOZ -case /blhome/fischer/runs/mozt5_1_03_2000_OOB
create_newcase -mach bluefire -res f19_g16 -compset F1850W -case /blhome/fischer/runs/waccm5_1_03_1850_OOB
create_newcase -res f19_f19 -compset F_SD_WACCM -mach bluefire -case /blhome/fischer/runs/wcm_offline_OOB
experiment: $SVNREPO/cam1/branch_tags/r8_cam5_1_03_tags/r8b_cam5_1_03/
create_newcase -compset F1850C5 -res f19_f19 -mach bluefire -case /blhome/fischer/runs/cam5_1_03_1850_r8
create_newcase -mach bluefire -res f19_g16 -compset FMOZ -case /blhome/fischer/runs/mozt5_1_03_2000_r8
create_newcase -mach bluefire -res f19_g16 -compset F1850W -case /blhome/fischer/runs/waccm5_1_03_1850_r8
create_newcase -res f19_f19 -compset F_SD_WACCM -mach bluefire -case /blhome/fischer/runs/wcm_offline_r8
MSS location of control simulations used to validate new climate:
Control runs:
/FISCHER/csm/cam5_1_03_1850_OOB
/FISCHER/csm/mozt5_1_03_2000_OOB
/FISCHER/csm/waccm5_1_03_1850_OOB
/FISCHER/csm/wcm_offline_OOB
Experiment runs:
/FISCHER/csm/cam5_1_03_1850_r8
/FISCHER/csm/mozt5_1_03_2000_r8
/FISCHER/csm/waccm5_1_03_1850_r8
/FISCHER/csm/wcm_offline_r8
URL for AMWG diagnostics output used to validate new climate:
http://www.cgd.ucar.edu/cms/fischer/cam5/cam5_1_03_1850_r8-cam5_1_03_1850_OOB/
http://www.cgd.ucar.edu/cms/fischer/r8/mozart/mozt5_1_03_2000_r8-mozt5_1_03_2000_OOB/
http://www.cgd.ucar.edu/cms/fischer/r8/waccm/waccm5_1_03_1850_r8-waccm5_1_03_1850_OOB/
http://www.cgd.ucar.edu/cms/fischer/r8/waccm/wcm_offline_r8-wcm_offline_OOB/
===============================================================
===============================================================
Tag name: cam5_1_10
Originator(s): fischer, jedwards, Mark Taylor
Date: Thu Oct 6 2011
One-line Summary: Update externals, memory use reduction, fix column data output
Purpose of changes:
. Updated SVN externals to match cesm1_1_beta03
. Memory reduction in src/chemistry/mozart/mo_drydep.F90 provided my
Mark Taylor
. Fix bug where column data was the same across history files
. Improve column data test
Bugs fixed (include bugzilla ID):
. #1417 CAM history column output: h0 column data same as h1 column
data even when specified to be different
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
. 250MB reduction for CAM5 1/8 degree with changes to
src/chemistry/mozart/mo_drydep.F90
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/nl_files/ghgrmp
. improve column data testing
M models/atm/cam/src/control/cam_history_support.F90
M models/atm/cam/src/control/cam_history.F90
M models/atm/cam/src/utils/cam_pio_utils.F90
. bug #1417 fix
M models/atm/cam/src/chemistry/mozart/mo_drydep.F90
. Memory usage reduction for CAM5
M SVN_EXTERNAL_DIRECTORIES
. Updated to cesm1_1_beta03 svn externals
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: all passed, except for TBL with ghgrmp
edinburgh/lf95: all passed, except for TBL with ghgrmp
edinburgh/pgi: all passed, except for TBL with ghgrmp
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: Test with ghgrmp where changed for better column data
testing. This causes TBLs for these tests to fail.
===============================================================
===============================================================
Tag name: cam5_1_09
Originator(s): Edwards
Date: 09-27-2011
One-line Summary: Add radiation_scheme string
Purpose of changes: Memory usage reduction for cam5
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist: Added variable
radiation_scheme with allowed values 'rrtmg' and 'camrt'
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Memory reduction when using rrtmg since these 4D camrt variables were allocated but not required.
Code reviewed by: Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
bld/namelist_files/namelist_defaults_cam.xml
bld/namelist_files/namelist_definition.xml
bld/build-namelist
physics/cam/phys_control.F90
physics/cam/radae.F90
Add support for radiation_scheme variable, allocate 4D structures only
for camrt radiation
physics/rrtmg/radiation.F90
Remove unnessasary use of radae
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All pass
edinburgh/lf95: All pass
edinburgh/pgi or jaguar/pgi: All pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_1_08
Originator(s): Francis Vitt
Date: 21 Sep 2011
One-line Summary: Misc bug fixes
Purpose of changes:
Bug fixes include :
- correction to O3 chemical loss rate diagnostic
- fixed a restart problem that occurs when empty_htapes=.true. in waccm configurations
- corrections to time-averaging history output
- correction to sathist output of dynamics decomposed variables on the HOMME grid
- replaced "physconst, only: amass => mwdry" with "physconst, only: mwdry" in several
chemistry modules to appease the PGI 9.3 compiler
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Brian Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/namelist_files/use_cases/sd_waccm_geos5.xml
M models/atm/cam/bld/namelist_files/use_cases/cam4_bam_radpsv_geos5.xml
M models/atm/cam/bld/namelist_files/use_cases/sd_waccm_tslt_geos5.xml
M models/atm/cam/bld/namelist_files/use_cases/cam4_chem_radpsv_geos5.xml
- the first meteorology data file contains one month of data so that
one can run more than one day with data which is downloaded from the
subversion inputdata repository
M models/atm/cam/test/system/nl_files/sat_hist
- include more output fields for more thorough tests of sat_hist and
local-time averaging outputs
M models/atm/cam/src/control/cam_history_buffers.F90
- corrections to local-time averaging accumulation
M models/atm/cam/src/control/sat_hist.F90
- correction to column output of dynamics decomposed variables on the HOMME grid
- moved dyn_find_col subroutine code to the dyn_grid module
M models/atm/cam/src/control/cam_history.F90
- fixed a restart problem that occurs when empty_htapes=.true. in waccm configurations
- corrected the buffer allocations for time-averaging output
- corrected the h_normalize subroutine for time-averaging output buffer
M models/atm/cam/src/physics/cam/hb_diff.F90
- correction to the TKE diagnostic so that it will be the same for different
MPI configurations
M models/atm/cam/src/dynamics/sld/dyn_grid.F90
M models/atm/cam/src/dynamics/eul/dyn_grid.F90
M models/atm/cam/src/dynamics/homme/dyn_grid.F90
M models/atm/cam/src/dynamics/fv/dyn_grid.F90
- added untility methods to find column nearest a specified coordinate and
determine coordinates of a specified element (or lat slice), which
are used in sat_hist and local-time averaging history features.
M models/atm/cam/src/physics/cam/vertical_diffusion.F90
- replace WACCM_GHG and WACCM_MOZART cppdefs with WACCM_PHYS
M models/atm/cam/src/chemistry/pp_none/chem_mods.F90
- change nabscol from 0 to 2 so that arrays are not declared to have zero size
M models/atm/cam/src/chemistry/mozart/mo_photo.F90
- declare do_diag variable that is needed when DEBUG is true
M models/atm/cam/src/chemistry/mozart/linoz_data.F90
M models/atm/cam/src/chemistry/mozart/mz_aerosols_intr.F90
M models/atm/cam/src/chemistry/utils/prescribed_aero.F90
M models/atm/cam/src/chemistry/utils/prescribed_ozone.F90
M models/atm/cam/src/chemistry/utils/prescribed_ghg.F90
M models/atm/cam/src/chemistry/utils/aircraft_emit.F90
M models/atm/cam/src/chemistry/utils/prescribed_volcaero.F90
M models/atm/cam/src/chemistry/mozart/tracer_cnst.F90
M models/atm/cam/src/physics/waccm/waccm_forcing.F90
- replaced "physconst, only: amass => mwdry" with "physconst, only: mwdry"
to appease the PGI 9.3 compiler
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
- update to the chemistry preprocessor to correct O3 loss rate diagnostic and
include missing r8 kind specifiers
M models/atm/cam/src/chemistry/pp_waccm_ghg/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_v1/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_none/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_bam/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_ghg/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_imp_sol.F90
- correction to O3 loss rate diagnostic
- include missing "r8" kind specifier
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All pass
edinburgh/lf95:
043 bl318 TBL.sh f10c4cdm sat_hist 9s .....................................FAIL! rc= 7 at Tue Sep 20 15:24:14 MDT 2011
This failure is expected do the corrections in local-time averaging history output.
edinburgh/pgi:
052 bl734 TBL.sh hn16c5aqdm outfrq3s+aquaplanet_cam5 9s ...................FAIL! rc= 5 at Tue Sep 20 11:20:45 MDT 2011
The base-line code (cam5_1_07) did not successfully compile with PGI 9.3
It was noted in tag cam5_1_06 that this did compile after several attempts.
Several attempts were made here without success. The mwdry changes to the
chemistry modes, as noted above, should remedy this problem.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_1_07
Originator(s): jedwards
Date: 9-19-2011
One-line Summary: correct cam.cpl7.template for backward compatibility
Purpose of changes: Correct problems with cam restart file names in
branch runs.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Feddema
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/cam.cpl7.template
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
This file does not affect any cam standalone tests, it is only used in
the cesm build. Pretag testing was done in the CESM test suite
using cesm1_1_alpha02g
===============================================================
===============================================================
Tag name: cam5_1_06
Originator(s): Levy, Mirin, Taylor, Garcia, Edwards, Feddema
Date: 9-1-2011
One-line Summary: CAM-SE dycore update
Purpose of changes: Merge latest SE dycore development into CAM
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Made NETCDF 4.1.3_seq default on bluefire
Made PGI 9.3 pgf95 default on edinburgh
Describe any changes made to the namelist:
updated default settings for a number of dycore specific fields.
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Jedwards, Mtaylor
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
models/atm/cam/tools/interpic_new/README - refined instructions for interpolating to CAM-SE initial file.
models/atm/cam/bld/configure - added code to link libnetcdff.a if it is found in the nc_lib directory
models/atm/cam/bld/cam.cpl7.template - added code to allow branch runs with restart files from older cam
versions which use cam2 in the filename instead of cam
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
models/atm/cam/bld/namelist_files/namelist_definition.xml
models/atm/cam/bld/build-namelist
Added some cam-se specific namelist variables. Changes some default values.
models/atm/cam/test/system/test_driver.sh
Changed default netcdf for bluefire and PGI for edinburgh
models/atm/cam/test/system/TSC.sh
ncdump no longer accepts more than one file on the command line, changed this test
to look at one file at a time.
models/atm/cam/SVN_EXTERNAl_DIRECTORIES
Update cam-se external pointer
models/atm/cam/doc/ChangeLog
models/atm/cam/src/physics/cam/cam_diagnostics.F90
Added 3 new diagnostic fields:
Z1000
PREC_pcw (precip due to prognostic cloud water)
PREC_zmc (Precip duc to ZM convective scheme)
models/atm/cam/src/dynamics/homme/pmgrid.F90
models/atm/cam/src/dynamics/homme/dyn_grid.F90
models/atm/cam/src/dynamics/homme/interp_mod.F90
models/atm/cam/src/dynamics/homme/native_mapping.F90
models/atm/cam/src/dynamics/homme/dyn_comp.F90
models/atm/cam/src/dynamics/homme/inidat.F90
models/atm/cam/src/dynamics/homme/stepon.F90
models/atm/cam/src/dynamics/homme/restart_dynamics.F90
Bug fixes and updates for compatability with the new cam-se externals
Also creates a new HommeMapping.nc file which provides connectivity information
for the history grid.
models/atm/cam/src/dynamics/homme/share
Latest cam-se dycore code from the HOMME standalone trunk. Includes the
ability to run on non-uniform (telescoping) grids.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All pass except:
063 bl731 TBL.sh h16c3aqdh aqua+aquaplanet_cam3 9s ........................FAIL
066 bl751 TBL.sh h30c4aqdm outfrq3s+aquaplanet_cam4 9s ....................FAIL
edinburgh/lf95: All pass
edinburgh/pgi or jaguar/pgi: All pass except:
052 bl734 TBL.sh hn16c5aqdm outfrq3s+aquaplanet_cam5 9s ...................FAIL
(Note that test 050 sm734 TSM.sh hn16c5aqdm had problems building on edinburgh, these problems
seem to be compiler or system related - the build completed and tests passed after repeated attempts,
this test was also run on lynx with no issues)
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
All differences are in the CAM-SE dycore, we do not yet have a baseline climate established for this dycore.
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_1_05
Originator(s): mvertens, fischer, eaton
Date: Wed Aug 31 2011
One-line Summary: Add multi instance support, replace cam2 with cam in
history file names, update externals
Purpose of changes:
. Add multi instance support for cesm.
. Replace cam2 with cam in history file names. Fix tests to handle
the file name changes.
. Modifications to mkDepends to relax the restrictions on
source file name matching module name and only one module per source
file.
. Update to current externals
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
. configure needed to be changed to set the number of instances
for each component. This is needed for cam stand alone, the
CESM scripts were setting these values.
. configure was also updated to work with the newest pio and mct tags
. mkDepends was modified to relax the restrictions on source file
name matching module
Describe any changes made to the namelist:
. vect_map was added to the nameslists and set to cart3d for homme.
The new driver tag was setting this to a default of npfix, which
was causing failures.
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
A models/atm/cam/src/control/cam_instance.F90
. Initializes instance variables inst_name, inst_index, inst_suffix
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/TBR.sh
M models/atm/cam/test/system/TEQ_ccsm.sh
M models/atm/cam/test/system/TER_ccsm.sh
M models/atm/cam/test/system/TSM_ccsm.sh
M models/atm/cam/test/system/nl_files/scm_b4b_o1
Fixed tests to replace cam2 with cam in history files.
M models/atm/cam/bld/configure
. Set NUM_COMP_INST_ATM, NUM_COMP_INST_LND, NUM_COMP_INST_OCN, NUM_COMP_INST_ICE
and NUM_COMP_INST_GLC to 1 for CAM stand alone runs. This is needed for the
new driver tag.
. Updated to work with latest pio and mct tags
M models/atm/cam/bld/namelist_files/namelist_definition.xml
M models/atm/cam/bld/build-namelist
. Add vect_map and set to cart3d for homme. New driver tag was setting the default
for vect_map to npfix
M models/atm/cam/bld/mkDepends
. Modifications to relax the restrictions on source file name matching module name
and only one module per source file. Also added a new "-d objfile" option which
allows an additional object file to be added to every dependence.
M models/atm/cam/bld/cam.cpl7.template
M models/atm/cam/src/control/runtime_opts.F90
M models/atm/cam/src/control/sat_hist.F90
M models/atm/cam/src/control/cam_restart.F90
M models/atm/cam/src/control/cam_comp.F90
M models/atm/cam/src/control/cam_history.F90
M models/atm/cam/src/utils/cam_pio_utils.F90
M models/atm/cam/src/cpl_mct/atm_comp_mct.F90
M models/atm/cam/src/physics/cam/restart_physics.F90
M models/atm/cam/src/cpl_esmf/atm_comp_mct.F90
M models/atm/cam/src/cpl_esmf/atm_comp_esmf.F90
. Changes made for multi-instance support. Replace cam2 with cam
in history filenames
M SVN_EXTERNAL_DIRECTORIES
. Updated externals.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
073 bl992 TBL_ccsm.sh f19_g16 E 2d ........................................FAIL! rc= 7 at Fri Aug 19 11:30:01 MDT 2011
. E compset was updated in the CESM1 scripts
. all other passed, expect for baselines because output files were renamed, but still bfb
edinburgh/lf95: all passed, expect for baselines becuase output files were renamed, but still bfb
edinburgh/pgi: all passed , expect for baselines because output files were renamed, but still bfb
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.: All baseline tests will fail
because history file names where change from having cam2 to cam. However,
this tag is still bfb with the previous.
===============================================================
===============================================================
Tag name: cam5_1_04
Originator(s): fischer, eaton jedwards
Date: 8 Aug 2011
One-line Summary: Fix cloud diagnostics, and remove unused allocate
Purpose of changes:
. Bugfix for the diagnostic output fields CLDLIQSTR and CLDICESTR. The
previous version was based on state values that did not contain the
tendencies from the macrophysics package.
. Allocate statement in mo_drydep.F90 not used and was removed. It could
have caused problems with UNSTRUCTURED grids since the nlon_veg and nlat_veg
variables are not defined in that case.
Bugs fixed (include bugzilla ID): #1386
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/src/physics/cam/macrop_driver.F90
. Bugfix for the diagnostic output fields CLDLIQSTR and CLDICESTR. The
previous version was based on state values that did not contain the
tendencies from the macrophysics package.
M models/atm/cam/src/chemistry/mozart/mo_drydep.F90
. The allocate statement at line 1861 is not used and was removed. It
could have caused problems with UNSTRUCTURED grids since the nlon_veg and
nlat_veg variables are not defined in that case.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: all passed
edinburgh/lf95: all passed
edinburgh/pgi: all passed
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.: all bfb
===============================================================
===============================================================
Tag name: cam5_1_03
Originator(s): Francis Vitt
Date: 1 Aug 2011
One-line Summary: Extension of the prescribed aerosol reading capabilities.
Purpose of changes:
The prescribed aerosol data modules were extended to be able to read in modal
aerosol concentrations and aerosol deposition fluxes. This was done to facilitate
development of prescribed Modal Aerosol Model configurations.
Bugs fixed (include bugzilla ID):
Missing namelist declaration of drydep_srf_file for chem_inparm in chemistry.F90 (1377)
Describe any changes made to build system:
Describe any changes made to the namelist:
New namelist variable added: "prescribed_aero_model"
Switch used to indicate which type of aerosols are prescribed -- bulk or modal.
This is used to set the default prescribed_aero_specifier and aerodep_flx_specifier
namelist variables.
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
D models/atm/cam/src/chemistry/modal_aero/modal_aero_deposition.F90
- this was moved to models/atm/cam/src/chemistry/utils
List all subroutines added and what they do:
A models/atm/cam/src/chemistry/utils/modal_aero_deposition.F90
- moved here so that it can be used when modal aerosols are not prognosed
- modified init method so the the indices can be specified when modal aerosols are not prognosed
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/build-namelist
- modified to set default aerodep_flx_specifier and prescribed_aero_specifier
according to the new "prescribed_aero_model" namelist variable which can be
"bulk" or "modal"
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- new "prescribed_aero_model" namelist variable added
- rewrite of the "prescribed_aero_specifier" namelist description
- corrections to the "type" entries of prescribed_aero_specifier and aerodep_flx_specifier
namelsit variables
M models/atm/cam/src/physics/cam/advnce.F90
- pass pbuf to prescribed data modules
M models/atm/cam/src/physics/waccm/waccm_forcing.F90
M models/atm/cam/src/chemistry/mozart/tracer_cnst.F90
M models/atm/cam/src/chemistry/mozart/tracer_srcs.F90
- remove reference to fields%chnk_offset
M models/atm/cam/src/chemistry/utils/aerodep_flx.F90
- Expanded to accommodate prescribed modal aerosol deposition fluxes.
M models/atm/cam/src/chemistry/utils/prescribed_aero.F90
- This has be generalized to read any prescribed field from datasets
and store them in the physics buffer. The units convertion feature
has been removed
M models/atm/cam/src/chemistry/utils/tracer_data.F90
- now can handle 2D fields that are stored in files that have lev dimension
- no longer store pointer to pbuf locations but store pbuf indices for pbuf fields
M models/atm/cam/src/chemistry/utils/prescribed_ozone.F90
M models/atm/cam/src/chemistry/utils/prescribed_ghg.F90
M models/atm/cam/src/chemistry/utils/aircraft_emit.F90
M models/atm/cam/src/chemistry/utils/prescribed_volcaero.F90
M models/atm/cam/src/chemistry/mozart/linoz_data.F90
- pbuf is now passed into the adv method and operates on the pbuf field
using the stored pbuf indices rather the stored pbuf pointers
M models/atm/cam/src/chemistry/modal_aero/modal_aero_initialize_data.F90
- check for prescribed aero dep fluxes before invoking the modal_aero_deposition_init
routine
M models/atm/cam/src/chemistry/mozart/chemistry.F90
- restored namelist declaration of drydep_srf_file for chem_inparm
- check for prescribed aero dep fluxes before setting the cam_out fluxes to the
prognosed deposition fluxes
M models/atm/cam/src/chemistry/mozart/mz_aerosols_intr.F90
M models/atm/cam/src/chemistry/bulk_aero/dust_intr.F90
- check for prescribed aero dep fluxes before setting the cam_out fluxes to the
prognosed deposition fluxes
M models/atm/cam/src/chemistry/mozart/mo_tgcm_ubc.F90
- removed redundant "save"
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All Pass
edinburgh/lf95: All Pass
edinburgh/pgi: All Pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_1_02
Originator(s): mirin, fischer, eaton
Date: Fri Jun 17 2011
One-line Summary: Implement reference pressure module.
Purpose of changes:
. The physics/chemistry parameterizations only need vertical level
locations expressed as pressures. In many places the hybrid coefficients
were being used to directly compute reference pressures (assuming a
surface reference pressure of 1e5 Pa). Since future dycores may use
different types of vertical coordinates we needed to remove any direct
access to the hybrid coefficients. A new reference pressure module
(ref_pres) has been implemented and all direct references to the hybrid
coefficients in the physics/chemistry code have been refactored to use
this module. Currently the physics is operating on the grid defined by
the dynamics, so each dycore must supply a subroutine (in dyn_grid) which
provides the reference pressures. The ref_pres module calls this
subroutine during initialization to set the reference pressures.
. Do some refactoring of the initialization sequence to move where the
hybrid coefficients are read from dycore independent places to dycore
dependent places.
. Update documentation in namelist_definition.xml (descriptions of
variables and categories).
. Update test lists for bluefire
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
. Update documentation in namelist_definition.xml (descriptions of
variables and categories).
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
models/atm/cam/src/control/hycoef.F90
. file moved to src/utils/
models/atm/cam/src/physics/cam/cldconst.F90
models/atm/cam/src/physics/cam/cldinti.F90
. code consolidated and moved to cloud_fraction.F90
List all subroutines added and what they do:
models/atm/cam/src/physics/cam/ref_pres.F90
. new module to provide access to the reference pressures
models/atm/cam/src/utils/hycoef.F90
. file moved here from src/control/
models/atm/cam/test/system/config_files/f1.9c4cm
models/atm/cam/test/system/config_files/f1.9c5cm
. created new 2deg COSP test with debug turned off
List all existing files that have been modified, and describe the changes:
models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
. change actual arg to set_ub_col from ps to pint(:,1). This is because
the top interface pressure was being calculated using ps and hybrid coefs.
models/atm/cam/src/control/cam_restart.F90
. remove reading of the hybrid coefs from this dycore independent spot.
Moved to the dycore dependent read_restart_dynamics methods.
. add call to initialize the reference pressures before the physics restart
is read.
models/atm/cam/src/control/readinitial.F90
. remove the call that reads the hybrid coefs from the initial file (and
all associated error checks). This is moved to dycore specific dyn_init
methods.
. remove a bunch of unused local variable declarations
models/atm/cam/src/control/startup_initialconds.F90
. remove unused 'include netcdf.inc'
models/atm/cam/src/dynamics/eul/dyn_comp.F90
models/atm/cam/src/dynamics/fv/dyn_comp.F90
models/atm/cam/src/dynamics/homme/dyn_comp.F90
models/atm/cam/src/dynamics/sld/dyn_comp.F90
. add file handle dummy arg to dyn_init method
. call hycoef_init to read and initialize hybrid coefs
models/atm/cam/src/dynamics/eul/dyn_grid.F90
models/atm/cam/src/dynamics/fv/dyn_grid.F90
models/atm/cam/src/dynamics/homme/dyn_grid.F90
models/atm/cam/src/dynamics/sld/dyn_grid.F90
. add dyn_grid_get_pref method. This is to allow the dycore to supply the
reference pressures to the physics.
models/atm/cam/src/dynamics/eul/inital.F90
models/atm/cam/src/dynamics/fv/inital.F90
models/atm/cam/src/dynamics/homme/inital.F90
models/atm/cam/src/dynamics/sld/inital.F90
. add file handle to dyn_init calling args
. call ref_pres_init before call to read initial_conds
models/atm/cam/src/dynamics/eul/initcom.F90
models/atm/cam/src/dynamics/fv/initcom.F90
models/atm/cam/src/dynamics/sld/initcom.F90
. remove call to hycoef_init. moved to dyn_init
models/atm/cam/src/dynamics/eul/restart_dynamics.F90
models/atm/cam/src/dynamics/fv/restart_dynamics.F90
models/atm/cam/src/dynamics/homme/restart_dynamics.F90
models/atm/cam/src/dynamics/sld/restart_dynamics.F90
. call init_restart_hycoef from init_restart_dynamics -- vdimids is now
local and can be removed from init_restart_dynamics args
. call write_restart_hycoef from write_restart_dynamics
. in read_restart_dynamcis add file handle to dyn_init calling args
models/atm/cam/src/physics/cam/cloud_fraction.F90
. put code from cldinti into cldfrc_init
models/atm/cam/src/physics/cam/eddy_diff.F90
. init_eddy_diff -- remove unused pref_mid dummy arg
. replace 'stop' by endrun call
models/atm/cam/src/physics/cam/physics_types.F90
. remove redundant calculation of psdry and pintdry
models/atm/cam/src/physics/cam/physpkg.F90
. change actual args from hyp* to pref_*
. remove call to cldinti -- code moved to cldfrc_init
models/atm/cam/src/physics/cam/uw_conv.F90
. replace a 'stop' by call endrun
models/atm/cam/src/chemistry/mozart/mo_aurora.F90
models/atm/cam/src/chemistry/mozart/mo_photo.F90
models/atm/cam/src/chemistry/mozart/mo_sad.F90
models/atm/cam/src/chemistry/mozart/mo_synoz.F90
models/atm/cam/src/chemistry/mozart/mo_waccm_hrates.F90
models/atm/cam/src/chemistry/utils/mo_msis_ubc.F90
models/atm/cam/src/chemistry/utils/tracer_data.F90
models/atm/cam/src/chemistry/pp_waccm_mozart_v1/mo_photo.F90
models/atm/cam/src/chemistry/pp_waccm_mozart_v1/mo_sad.F90
models/atm/cam/src/physics/cam/boundarydata.F90
models/atm/cam/src/physics/cam/cldwat.F90
models/atm/cam/src/physics/cam/radae.F90
models/atm/cam/src/physics/cam/tphysidl.F90
models/atm/cam/src/physics/cam/tracers_suite.F90
models/atm/cam/src/physics/cam/vertical_diffusion.F90
models/atm/cam/src/physics/rrtmg/radlw.F90
models/atm/cam/src/physics/waccm/gravity_waves_sources.F90
models/atm/cam/src/physics/waccm/qbo.F90
. replace use of hycoef module by use of ref_pres
models/atm/cam/src/physics/cam/convect_deep.F90
models/atm/cam/src/physics/cam/convect_shallow.F90
models/atm/cam/src/physics/cam/gw_drag.F90
models/atm/cam/src/physics/cam/hb_diff.F90
models/atm/cam/src/physics/cam/radheat.F90
models/atm/cam/src/physics/cam/zm_conv_intr.F90
models/atm/cam/src/physics/waccm/gw_drag.F90
models/atm/cam/src/physics/waccm/iondrag.F90
models/atm/cam/src/physics/waccm/nlte_lw.F90
models/atm/cam/src/physics/waccm/radheat.F90
. change dummy args from hyp* to pref_*
models/atm/cam/test/system/input_tests_master
models/atm/cam/test/system/tests_posttag_bluefire
. Turned off debugging for COSP CAM5 2deg test
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: all passed
edinburgh/lf95: all passed
edinburgh/pgi: all passed
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.: 2deg COSP baseline tests will fail
because debugging was turned off in the test. Everything is bfb
===============================================================
===============================================================
Tag name: cam5_1_01
Originator(s): eaton, fvitt, fischer
Date: Mon May 30 18:24:32 MDT 2011
One-line Summary: enable 1/4 deg FV cam5, chemistry bug fixes
Purpose of changes:
. Enable 1/4 deg FV cam5 to run out of the box. This required a code mod
to the interpolation of the landuse map in the dry deposition code used
by the modal aerosols, and providing an appropriate initial file.
. corrected O1D + CCL4 -> 4*CL reaction in the waccm_mozart_v1 mechanism
. removed reference to solar_parms_file from cam4_trop_strat_chem use_case
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets:
. add default IC file for 1/4 deg FV cam5.
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/src/chemistry/mozart/mo_drydep.F90
. subroutine interp_map -- clamp the values of fraction_landuse between 0
and 1. This replaces a check for out of range values which then
triggered a call to endrun.
models/atm/cam/src/chemistry/pp_waccm_mozart/chem_mech.in
. added comment
models/atm/cam/src/chemistry/pp_waccm_mozart_v1/mo_prod_loss.F90
models/atm/cam/src/chemistry/pp_waccm_mozart_v1/mo_lin_matrix.F90
models/atm/cam/src/chemistry/pp_waccm_mozart_v1/chem_mech.doc
models/atm/cam/src/chemistry/pp_waccm_mozart_v1/chem_mech.in
. corrected O1D + CCL4 -> 4*CL reaction in the waccm_mozart_v1 mechanism
. added comment to chem_mech.in
models/atm/cam/bld/namelist_files/use_cases/cam4_trop_strat_chem.xml
. removed reference to solar_parms_file (waccm input file)
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS
edinburgh/lf95: All PASS
edinburgh/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_1_00
Originator(s): fvitt, fischer, jedwards
Date:
One-line Summary: chemistry corrections, new homme ncdata file
Purpose of changes:
- chemistry preprocessor update -- adjust the write format to *.doc output
files to allow negative values in the "troe" reactions
- correction to pp_trop_mozart document file resulting from the above preprocessor change
- numerous corrections to the chem reactions in trop_strat_bam_v1 package
- corrected the reaction: O1D + CCL4 -> 4*CL
- correction to pp_waccm_mozart_v1 document file resulting from the above preprocessor change
- new homme dycor ne240np4 1850 ncdata file
- add output needed for radition driver
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
-atm/cam/inic/homme/cami_0002-01-01_ne240np4_L26_c100522.nc
+atm/cam/inic/homme/cami_1850-01-01_ne240np4_L26_c110314.nc
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M bld/namelist_files/namelist_defaults_cam.xml
- new homme dycor ne240np4 1850 ncdata file
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
- chemistry preprocessor update -- adjust the write format to *.doc output
files to allow negative values in the "troe" reactions
M models/atm/cam/src/chemistry/pp_trop_mozart/chem_mech.doc
- correction to this document file resulting from the above preprocessor change
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_setrxt.F90
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/chem_mech.in
- numerous corrections to the chem reactions in trop_strat_bam_v1 package
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/chem_mech.doc
M models/atm/cam/src/chemistry/pp_waccm_mozart/chem_mech.in
- corrected the reaction: O1D + CCL4 -> 4*CL
M models/atm/cam/src/chemistry/pp_waccm_mozart_v1/chem_mech.doc
- correction to this document file resulting from the above preprocessor change
M models/atm/cam/src/physics/cam/radiation_data.F90
- add output needed for radition driver
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
all passed
edinburgh/lf95:
all passed
edinburgh/pgi:
all passed
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: all bfb
===============================================================
===============================================================
Tag name: cam5_0_57
Originator(s): fischer, eaton, hannay, fvitt, andrew
Date: Tue May 17 13:57:22 MDT 2011
One-line Summary: CAM5.1 tuning, CAM5 default output updates, bug fixes
Purpose of changes:
. CAM5.1 2 degree tuning, new values for uwshcu_rpen and rhminl
. CAM5 default outputs added
. NEU Chemistry bug fixes
. Micro physics bug fixes from Andrew Gettelman
. Modify build-namelist to produce default FV decomposition settings
(npr_yz). This functionality was previously embedded in the CESM scripts
and was not available to user's of CAM standalone scripts.
. Change dataroot for lynx testing
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
. build-namelist has a new argument, -ntasks, which is used to set a
default value for the FV decomposition. The user can always override
this default by explicitly setting npr_yz.
Describe any changes made to the namelist:
. uwshcu_rpen was added to the namelist for CAM5.
Default value is 10.0
At 2deg the value is 5.0, this changes answers
. rhminl was set the .8875 for 2deg CAM5, this changes answers
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
. uwshcu_readnl was added to models/atm/cam/src/physics/cam/uwshcu.F90
uwshcu_readnl reads the namelist variable uwshcu_rpen
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/build-namelist
. add fv_decomp_set method used to set default FV decomp when new -ntasks
arg is set and the user hasn't already explicitly set it (npr_yz).
. add namelist variable uwshcu_rpen
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. add namelist variable uwshcu_rpen for CAM5 and set default to 10.0
uwshcu_rpen is set to 5.0 when resoltion is 2 degree
. set cldfrc_rhminl to .8875 for 2 degree CAM5
models/atm/cam/bld/namelist_files/namelist_definition.xml
. add uwshcu_repn definition
uwshcu_rpen - For penetrative entrainment efficiency
models/atm/cam/src/control/runtime_opts.F90
. add call to uwshcu_readnl to read in uwshcu_rpen
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. set cldfrc_rhminl to .8875 for 2 degree CAM5
. set uwshcu_rpen to 5.0 for 2 degree CAM5 and set default to 10
models/atm/cam/src/physics/cam/uwshcu.F90
. add routine uwshcu_readnl to set uwshcu_rpen
. add error checking to see if uwshcu_rpen is set
models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
. bug fix prevent overwriting of tendencies set by mo_neu_wetdep
models/atm/cam/src/chemistry/mozart/mo_neu_wetdep.F90
. turn off debug print statements
models/atm/cam/src/physics/cam/cldwat2m_micro.F90
models/atm/cam/src/physics/cam/macrop_driver.F90
. micro physics bug fixes from Andrew Gettelman
models/atm/cam/src/physics/cam/microp_driver.F90
. comment fix from Andrew Gettelman
models/atm/cam/src/physics/cam/cam_diagnostics.F90
. T700, T850, ATMEINT added to default output
models/atm/cam/src/physics/rrtmg/radiation.F90
models/atm/cam/src/physics/rrtmg/radiation.F90
. TOT_CLD_VISTAU and TOT_ICLD_VISTAU added to default output
models/atm/cam/src/physics/cam/microp_aero.F90
. CCN3 added to default output
odels/atm/cam/test/system/test_driver.sh
. set dataroot to /glade/proj3/cseg/inputdata for lynx
SVN_EXTERNAL_DIRECTORIES
. update clm tag to clm4_0_31
. update scripts tag to scripts4_110513
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
040 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .....................................FAIL! rc= 7 at Tue May 17 13:23:40 MDT 2011
047 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...........................FAIL! rc= 7 at Tue May 17 14:16:19 MDT 2011
. New CAM5 tunings for uwshcu_rpen and cldfrc_rhminl cause the baseline tests to fail
edinburgh/lf95:
all passed
edinburgh/pgi :
all passed
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: CAM5 at 2 degree will have answer changes due to
new tuning values for cldfrc_rhminl and uwshcu_rpen
===============================================================
===============================================================
Tag name: cam5_0_56
Originator(s): fischer, eaton
Date: Thu May 12 12:24:30 MDT 2011
One-line Summary: update component svn externals
Purpose of changes:
. update component svn externals
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
-lnd/clm2/pftdata/pft-physiology.c101006.nc
+lnd/clm2/pftdata/pft-physiology.c110425.nc
. Update
+lnd/clm2/initdata/clmi.BCN_0051-01-01_48x96_gx3v7_simyr2000_c110509.nc
+lnd/clm2/initdata/clmi.BCN.1850-01-01_48x96_gx3v7_simyr1850_c110421.nc
+lnd/clm2/surfdata/surfdata_512x1024_simyr1850_c100315.nc
+lnd/clm2/surfdata/surfdata.pftdyn_48x96_hist_simyr1850-2005_c110114.nc
+lnd/clm2/surfdata/surfdata.pftdyn_0.9x1.25_rcp6.0_simyr1850-2100_c100812.nc
+lnd/clm2/surfdata/surfdata.pftdyn_1.9x2.5_rcp6.0_simyr1850-2100_c100813.nc
+lnd/clm2/surfdata/surfdata.pftdyn_10x15_rcp6.0_simyr1850-2100_c100901.nc
. Added
-lnd/clm2/surfdata/surfdata_48x96_simyr2000_c090928.nc
+lnd/clm2/surfdata/surfdata_48x96_simyr2000_c100505.nc
-lnd/clm2/surfdata/surfdata_48x96_simyr1850_c090928.nc
+lnd/clm2/surfdata/surfdata_48x96_simyr1850_c110114.nc
-lnd/clm2/surfdata/surfdata_0.47x0.63_simyr1850_c100305.nc
+lnd/clm2/surfdata/surfdata_0.47x0.63_simyr1850_c100826.nc
-lnd/clm2/surfdata/surfdata.pftdyn_0.9x1.25_rcp8.5_simyr1850-2100_c100205.nc
+lnd/clm2/surfdata/surfdata.pftdyn_0.9x1.25_rcp8.5_simyr1850-2100_c100319.nc
. Changed
-lnd/clm2/surfdata/surfdata.pftdyn_0.9x1.25_rcp8.5_simyr1850-2100_c100205.nc
-lnd/clm2/surfdata/surfdata.pftdyn_0.9x1.25_rcp4.5_simyr1850-2100_c100406.nc
-lnd/clm2/surfdata/surfdata.pftdyn_0.9x1.25_rcp2.6_simyr1850-2100_c100323.nc
. Removed
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. Updated fsurdat, fptfcon, and finidat to match clm4_0_29
. Added rcp6.0 fpftdyn
. Removed extra > at line 249
. Removed RCP 2000-2100 fpftdyn
SVN_EXTERNAL_DIRECTORIES
. Updated all externals to match cesm1_0_beta20
SVN_EXTERNAL_DIRECTORIES
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
004 bl132 TBL.sh e48c4dh ghgrmp+1850_cam4 9s ..............................FAIL! rc= 7 at Wed May 11 16:36:52 MDT 2011
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Wed May 11 16:59:47 MDT 2011
. bl132 and bl137 failed because spun-up data is being used now
072 bl992 TBL_ccsm.sh f19_g16 E 2d ........................................FAIL! rc= 7 at Sat May 7 12:17:46 MDT 2011
. bl992 failed because of updated drv tag
edinburgh/lf95:
all passed
edinburgh/pgi:
all passed
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.: bfb except for: ccsm tests, and T31.
===============================================================
===============================================================
Tag name: cam5_0_55
Originator(s): jenkay, eaton, fischer
Date: 5 May 2011
One-line Summary: Fixes for COSP, reduce CAM5 default output.
Purpose of changes:
. Mods in modis simulator to workaround an xlf compiler problem.
. Modify both CAM standalone and CESM builds to allow modis simulator to
use the abortutils module instead of calling stop.
. Move some 'add_default' calls for CAM5 diagnostic fields inside conditionals.
. Update to 110505 COSP external
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
. Add abortutils dependency so cosp code can make use of subroutine endrun.
$(COSP_LIBDIR)/libcosp.a: abortutils.o
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
models/atm/cam/SVN_EXTERNAL_DIRECTORIES
. update to the 110505 COSP external
models/atm/cam/bld/Makefile.in
models/atm/cam/bld/configure
. mods to allow cosp to use associate CAM's abortutils module
models/atm/cam/src/chemistry/bulk_aero/dust_intr.F90
models/atm/cam/src/chemistry/bulk_aero/progseasalts_intr.F90
models/atm/cam/src/chemistry/modal_aero/modal_aero_calcsize.F90
models/atm/cam/src/chemistry/modal_aero/modal_aero_coag.F90
models/atm/cam/src/chemistry/modal_aero/modal_aero_gasaerexch.F90
models/atm/cam/src/chemistry/modal_aero/modal_aero_initialize_data.F90
models/atm/cam/src/chemistry/modal_aero/modal_aero_newnuc.F90
models/atm/cam/src/chemistry/mozart/mo_chm_diags.F90
models/atm/cam/src/chemistry/mozart/mo_extfrc.F90
models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
models/atm/cam/src/chemistry/mozart/mz_aerosols_intr.F90
models/atm/cam/src/physics/cam/aer_rad_props.F90
models/atm/cam/src/physics/cam/ndrop.F90
models/atm/cam/src/physics/cam/vertical_diffusion.F90
. remove some 'add_default' calls for CAM5 diagnostic fields. Many
add_default calls for modal aerosols were moved inside conditionals that
are turned on using the history_aerosol namelist variable.
models/atm/cam/src/physics/cam/cospsimulator_intr.F90
. fix long_name in some addfld calls
. fix bug in output fields clhmodis, clmmodis, cllmodis
. changes in cloud fraction weighting code
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
015 br381 TBR.sh f1.9c4cdm outfrq3s+1850-2005_cam4 6+3s ...................FAIL! rc= 12 at Thu May 5 00:05:32 MDT 2011
016 bl381 TBL.sh f1.9c4cdm outfrq3s+1850-2005_cam4 9s .....................FAIL! rc= 5 at Thu May 5 00:11:40 MDT 2011
019 br382 TBR.sh f1.9c5cdm outfrq3s+1850-2005_cam5 6+3s ...................FAIL! rc= 12 at Thu May 5 01:37:03 MDT 2011
020 bl382 TBL.sh f1.9c5cdm outfrq3s+1850-2005_cam5 9s .....................FAIL! rc= 5 at Thu May 5 01:45:22 MDT 2011
Branch tests failed due to changing task geometry, when keeping the task geometry consistent the branch tests passed
Baselines for COSP are expected to fail
edinburgh/lf95:
all passed
edinburgh/pgi:
043 bl317 TBL.sh f10c5cdm outfrq3s 9s .....................................FAIL! rc= 7 at Thu May 5 13:45:00 MDT 2011
Baselines for COSP are expected to fail
CAM tag used for the baseline comparison tests if different than previous
tag:cam5_0_54
Summarize any changes to answers: BFB except for COSP diagnostic output
===============================================================
===============================================================
Tag name: cam5_0_54
Originator(s): Francis Vitt
Date: 27 Apr 2011
One-line Summary: New chemistry features and minor corrections to trop_mozart reactions.
Purpose of changes:
- provide an additional optional wet deposition method for gas phase chemical species
- include short wavelength photolysis in tropospheric chemistry
- new treatment of WACCM stratospheric aerosols
- new chemistry package added : trop_strat_bam_v1
- adjustments to the trop_mozart chemistry mechanism
- provide dynamic creation of default deposition species lists
namelist settings using master lists
- correction to the wet deposition lists of species for
super_fast_llnl, and waccm_mozart
- correction to the configure options for the offline driver regression test
Science changes effect (baseline test will fail) these chemistry packages:
trop_mozart
- inclusion of the short wavelength photolysis
- adjustments to the chemistry mechanism (reaction rates)
super_fast_llnl
- correction to the wet deposition list (include CH3OOH)
waccm_mozart (development version of waccm)
- correction to the wet deposition list (N2O5 removed from list)
- new treatment of stratospheric aerosols
Bugs fixed (include bugzilla ID): None
Describe any changes made to build system:
Describe any changes made to the namelist:
namelist variables added
- gas_wetdep_list
list of gas phase species that under go wet depostion
- gas_wetdep_method
'MOZ' to select the mozart scheme (default)
'NEU' to select the
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Brian Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
A models/atm/cam/bld/namelist_files/use_cases/cam4_trop_strat_chem.xml
- new use case for F_TROP_STRAT_CHEM compset
A models/atm/cam/bld/namelist_files/master_aer_wetdep_list.xml
A models/atm/cam/bld/namelist_files/master_gas_wetdep_list.xml
A models/atm/cam/bld/namelist_files/master_mam_dep_list.xml
A models/atm/cam/bld/namelist_files/master_drydep_list.xml
A models/atm/cam/bld/perl5lib/Build/ChemNamelist.pm
- sets default depostion species lists namelist variables for the desired
chemistry mechanism using the above master lists
A models/atm/cam/src/chemistry/mozart/gas_wetdep_opts.F90
- reads wet depostion options for gas phase species
A models/atm/cam/src/chemistry/mozart/mo_neu_wetdep.F90
- new wet depostion method provided by J Neu
A models/atm/cam/src/chemistry/pp_trop_strat_bam_v1
A models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_lu_solve.F90
A models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_setrxt.F90
A models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_adjrxt.F90
A models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_imp_sol.F90
A models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/chem_mods.F90
A models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_prod_loss.F90
A models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_lin_matrix.F90
A models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/m_rxt_id.F90
A models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_phtadj.F90
A models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_nln_matrix.F90
A models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/m_het_id.F90
A models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_lu_factor.F90
A models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_sim_dat.F90
A models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/mo_indprd.F90
A models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/m_spc_id.F90
A models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/chem_mech.doc
A models/atm/cam/src/chemistry/pp_trop_strat_bam_v1/chem_mech.in
- new chem pkg added (133 species, interactive stratosphere and troposphere chem)
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/config_files/fmo1.9c4dh
- correction to the configure options for the offline driver test
M models/atm/cam/bld/configure
- new chemistry package added 'trop_strat_bam_v1'
- set $chem_src_dir to $cam_root/.... which is an absolute path so that
build-namelist deposition lists generation will function within CESM scripts
M models/atm/cam/bld/config_files/definition.xml
- new chemistry package added 'trop_strat_bam_v1' (-chem option)
M models/atm/cam/bld/build-namelist
- changes for the new trop_strat_bam_v1 chemisty package
- invokes the ChemNamelist module to generate depostion lists
- removed the if( $waccm_chem ) clause around the strat_aero_feedback/rad_climate setting
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- default emissions files added for trop_strat_bam_v1
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- namelist variables added
- gas_wetdep_list
list of gas phase species that under go wet depostion
- gas_wetdep_method
'MOZ' to select the mozart scheme (default)
'NEU' to select the
M models/atm/cam/bld/namelist_files/use_cases/sd_waccm_geos5.xml
M models/atm/cam/bld/namelist_files/use_cases/sd_waccm_tslt_geos5.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_refb1.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1850-2005_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2000_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp26.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp45.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1955-2005_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp85.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1974-2005_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1850_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_sc_2000_cam4.xml
- history fields names change related to SAD code changes
HNO3_CD1 replaced by HNO3_STS and HNO3_CD3 replaced by HNO3_NAT
M models/atm/cam/bld/perl5lib/Build/ChemPreprocess.pm
- added get_species_list routine and is made public so it can be used by ChemNamelist.pm
- set the chem_src_dir for customized mechanism
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
- updated chemistry preprocessor
M models/atm/cam/src/control/sat_hist.F90
- added a check to avoid an error which occurs when the last observation
time is exactly the same as the model time
M models/atm/cam/src/chemistry/pp_waccm_mozart_v1/mo_photo.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_v1/mo_sad.F90
- inteface changes to be constitent with changes in mo_gas_phase_chemdr
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_setrxt.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/m_het_id.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_mozart/chem_mech.in
- changes to chemistry reactions
- chemistry solver includes wash-out loss rates term for all species
which are set to non-zero for only the gas_wetdep_list namelist
M models/atm/cam/src/chemistry/pp_waccm_mozart/chem_mods.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/m_het_id.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/chem_mech.doc
M models/atm/cam/src/chemistry/pp_waccm_mozart/chem_mech.in
- chemistry solver includes wash-out loss rates term for all species
which are set to non-zero for only the gas_wetdep_list namelist
- this is a bit-for-bit change
M models/atm/cam/src/chemistry/pp_super_fast_llnl/chem_mods.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/m_het_id.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/chem_mech.doc
M models/atm/cam/src/chemistry/pp_super_fast_llnl/chem_mech.in
- chemistry solver includes wash-out loss rates term for all species
which are set to non-zero for only the gas_wetdep_list namelist
- the master lists give a different list of wet dep species so
this is not a bit-for-bit change, otherwise it is bit-for-bit
M models/atm/cam/src/chemistry/pp_waccm_mozart_v1/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_v1/chem_mods.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_v1/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_v1/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_v1/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_v1/m_het_id.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_v1/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart_v1/chem_mech.doc
M models/atm/cam/src/chemistry/pp_waccm_mozart_v1/chem_mech.in
- chemistry solver includes wash-out loss rates term for all species
which are set to non-zero for only the gas_wetdep_list namelist
- this is a bit-for-bit change
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_phtadj.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/m_het_id.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/m_spc_id.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_mam3/chem_mech.in
- chemistry solver includes wash-out loss rates term for all species
which are set to non-zero for only the gas_wetdep_list namelist
- this is a bit-for-bit change
M models/atm/cam/src/chemistry/pp_trop_bam/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_trop_bam/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_trop_bam/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_bam/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_trop_bam/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_bam/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_trop_bam/mo_phtadj.F90
M models/atm/cam/src/chemistry/pp_trop_bam/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_bam/m_het_id.F90
M models/atm/cam/src/chemistry/pp_trop_bam/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_trop_bam/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_bam/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_trop_bam/m_spc_id.F90
M models/atm/cam/src/chemistry/pp_trop_bam/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_bam/chem_mech.in
- chemistry solver includes wash-out loss rates term for all species
which are set to non-zero for only the gas_wetdep_list namelist
- this is a bit-for-bit change
M models/atm/cam/src/chemistry/pp_trop_mam7/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/m_het_id.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_mam7/chem_mech.in
- chemistry solver includes wash-out loss rates term for all species
which are set to non-zero for only the gas_wetdep_list namelist
- this is a bit-for-bit change
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/chem_mods.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/m_het_id.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/chem_mech.doc
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/chem_mech.in
- chemistry solver includes wash-out loss rates term for all species
which are set to non-zero for only the gas_wetdep_list namelist
- this is a bit-for-bit change
M models/atm/cam/src/chemistry/pp_trop_ghg/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_ghg/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_trop_ghg/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_ghg/m_het_id.F90
M models/atm/cam/src/chemistry/pp_trop_ghg/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_ghg/chem_mech.doc
M models/atm/cam/src/chemistry/pp_trop_ghg/chem_mech.in
- chemistry solver includes wash-out loss rates term for all species
which are set to non-zero for only the gas_wetdep_list namelist
- this is a bit-for-bit change
M models/atm/cam/src/chemistry/pp_waccm_ghg/chem_mods.F90
M models/atm/cam/src/chemistry/pp_waccm_ghg/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_waccm_ghg/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_waccm_ghg/m_het_id.F90
M models/atm/cam/src/chemistry/pp_waccm_ghg/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_waccm_ghg/chem_mech.doc
M models/atm/cam/src/chemistry/pp_waccm_ghg/chem_mech.in
- chemistry solver includes wash-out loss rates term for all species
which are set to non-zero for only the gas_wetdep_list namelist
- this is a bit-for-bit change
M models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
- HNO3_CD1 renamed to HNO3_STS
- HNO3_CD2 renamed to HNO3_NAT
- removed HNO3_CD2, SAD_SNAT, RAD_SNAT
- added call to neu_wetdep (new wet deposition scheme)
M models/atm/cam/src/chemistry/mozart/mo_strato_rates.F90
- chenges to names of variables
M models/atm/cam/src/chemistry/mozart/mo_chemini.F90
- call to neu_wetdep_init added
M models/atm/cam/src/chemistry/mozart/mo_jshort.F90
M models/atm/cam/src/chemistry/mozart/mo_photo.F90
- changes to use jshort for low top model if the needed fields are available
(short wavelength photolsys for CAM-Chem)
M models/atm/cam/src/chemistry/mozart/mo_sad.F90
- extensive changes provided by Doug Kinnison
this is a new treatment of the the WACCM stratospheric aerosols
M models/atm/cam/src/chemistry/mozart/chemistry.F90
- call gas_wetdep_readnl added to read wet dep namelist options
- call to neu_wetdep_tend added
M models/atm/cam/src/chemistry/mozart/mo_chem_utls.F90
M models/atm/cam/src/chemistry/mozart/mo_sethet.F90
(N2O5 removed from list) - changed need to specify which species have wet dep with namelist
setting
- chemistry solver includes wash-out loss rates term for all species
which are set to non-zero for only the gas_wetdep_list namelist
M models/atm/cam/src/chemistry/mozart/mz_aerosols_intr.F90
- now uses the gas_wetdep_list to check that none of the gas phase
species are included in the aerosol wet dep list
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
033 bl336 TBL.sh wm4c4h outfrq3s+waccm_1850_cam4 9s .......................FAIL! rc= 7 at Tue Apr 26 20:18:43 MDT 2011
the development version of waccm_mozart (not waccm_mozart_v1) is expected
to fail due to
- correction to the wet deposition list (N2O5 removed from list)
- new treatment of stratospheric aerosols
045 bl373 TBL.sh fm1.9c4dh outfrq3s 9s ....................................FAIL! rc= 7 at Tue Apr 26 21:24:59 MDT 2011
trop_mozart is expected to fail due to
- inclusion of the short wavelength photolysis
- adjustments to the chemistry mechanism (reaction rates)
050 bl374 TBL.sh fs1.9c4dh outfrq3s+1850-2005_cam4_super_fast_llnl 9s .....FAIL! rc= 7 at Tue Apr 26 22:07:42 MDT 2011
super_fast_llnl is expected to fail due to correction to the wet deposition list (include CH3OOH)
edinburgh/lf95: All Pass
edinburgh/pgi : All Pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_0_53
Originator(s): eaton
Date: Mon Apr 25 14:11:19 MDT 2011
One-line Summary: fix cosp external
Purpose of changes:
. The new cosp external didn't get 'propset' before the last commit.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes: none
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: not done
edinburgh/lf95: not done
edinburgh/pgi or jaguar/pgi: not done
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_0_52
Originator(s): fischer, eaton, fvitt, jenkay, tcraig
Date: 19 April 2011
One-line Summary: Update COSP, Fix for offline driver, performance fix for modal aerosol, ESMF updates
Purpose of changes:
. Fix performance problem introduced in cam5_0_27 in the modal aerosol
optics calculation.
. COSP updates, Includes the mods from LLNL to incorporate the effects of snow and other
bug fixes.
. Fix offline driver, add sd_waccm_ghg use case, fix SDBAM.
. ESMF updates from Tony Craig
Bugs fixed (include bugzilla ID):
. Fix for offline driver
. Fix the use case(cam4_bam_radpsv_geos5) used in SDBAM compset
Describe any changes made to build system:
Describe any changes made to the namelist:
. Added cosp_histfile_aux to the COSP namelist.
When set to true, additional output is generated.
. Use case cam4_bam_radpsv_geos5.xml changes
set strat aerosol data cycle year, file, and type
set ext_frc_type, srf_emis_type, and tracer_cnst_type to SERIAL
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
CAM5 1deg is ~ 6.5% faster than the previous tag.
previous tag .902 hrs for a 1deg 10 day run
cam5_0_52 .842 hrs for a 1deg 10 day run
Code reviewed by: Brian Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
A bld/namelist_files/use_cases/sd_waccm_geos5.xml
A bld/namelist_files/use_cases/waccm_sc_2000_cam4.xml
. New WACCM use cases from fvitt
A test/system/config_files/f1.9c4cdm
A test/system/config_files/f1.9c5cdm
. created 2deg COSP tests
List all existing files that have been modified, and describe the changes:
M test/system/input_tests_master
M test/system/tests_posttag_bluefire
. added 2deg COSP tests to bluefire post-tag tests
M test/system/nl_files/off1.9x2.5
. changed start_ymd, met_data_file
. set bnd_topo, and ncdata
M src/utils/time_manager.F90
M src/cpl_esmf/atm_comp_mct.F90
M src/cpl_esmf/atm_comp_esmf.F90
. ESMF updates from Tony Craig
M models/atm/cam/bld/namelist_files/namelist_definition.xml
. added cosp_histfile_aux and cosp_histfile_aux_num to the namelist
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. BAM and MAM aerosol emissions file changes
M models/atm/cam/bld/build-namelist
. added tracer_cnst_filelist
M models/atm/cam/bld/namelist_files/use_cases/cam4_bam_radpsv_geos5.xml
. set strat aerosol data cycle year, file, and type
. set ext_frc_type, srf_emis_type, and tracer_cnst_type to SERIAL
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
. Updated COSP external
M models/atm/cam/src/physics/cam/cospsimulator_intr.F90
M models/atm/cam/src/physics/rrtmg/radiation.F90
. COSP updates
M models/atm/cam/src/dynamics/fv/pfixer.F90
. set sum1 private for omp
M models/atm/cam/src/dynamics/fv/inidat.F90
M models/atm/cam/src/dynamics/fv/uv3s_update.F90
. added if defined for OFFLINE_DYN, WACCM_GHG, and WACCM_MOZART
M models/atm/cam/src/dynamics/fv/dyn_comp.F90
. made do loops omp parallel
M models/atm/cam/src/dynamics/fv/metdata.F90
. removed if debug
. added tests for TS in met dataset
models/atm/cam/src/physics/cam/modal_aer_opt.F90
. lookup the indicies of the modal aerosol species in the aerosol list at
initialization and save them (this is the main performance fix)
. assign pointers to modal aerosol specie mixing ratios outside the
calculation loops (not sure whether this is having a significant
performance impact)
models/atm/cam/src/physics/cam/rad_constituents.F90
. make rad_cnst_get_aer_idx public
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
all passed
edinburgh/lf95:
043 bl318 TBL.sh f10c4cdm sat_hist 9s .....................................FAIL! rc= 7 at Tue Apr 19 18:19:26 MDT 2011
Changes is COSP caused bl318 to fail
edinburgh/pgi or jaguar/pgi:
043 bl317 TBL.sh f10c5cdm outfrq3s 9s .....................................FAIL! rc= 7 at Tue Apr 19 12:05:14 MDT 2011
Changes is COSP caused bl317 to fail
CAM tag used for the baseline comparison tests if different than previous
tag: cam5_0_51
Summarize any changes to answers: BFB except for COSP diagnostic output
===============================================================
===============================================================
Tag name: cam5_0_51
Originator(s): Francis Vitt
Date: 4 April 2011
One-line Summary:
Added various chemistry features, bug fixes and code clean up.
Purpose of changes:
- provide more intuitive data-timing namelist variables
- update trop_mozart with latest MOZART4 mechanism
- restore lighting NOx production in super_fast_llnl
- give flexibility to the units of the emissions datasets
- use more up-to-date dataset for waccm_ghg forcing
- include CO2 reactions for WACCM to improve concentrations in upper regions
- updates to dry deposition module
- bug fixes listed below
Bugs fixed (include bugzilla ID):
- corrected vertically integrated wet deposition rates diagnostics
- corrected MASS and AREA output fields
- corrected chemical prod/loss rates diagnostics
Describe any changes made to build system:
Describe any changes made to the namelist:
- data-timing namelist changes:
* srf_emis_ymd --> srf_emis_cycle_yr
srf_emis_fixed_ymd
* srf_emis_tod --> srf_emis_fixed_tod
* ext_frc_ymd --> ext_frc_cycle_yr
ext_frc_fixed_ymd
* ext_frc_tod --> ext_frc_fixed_tod
* flbc_date --> flbc_cycle_yr
flbc_fixed_ymd
flbc_fixed_tod
* flbc_yr_offset removed
* prescribed_aero_ymd --> prescribed_aero_cycle_yr
prescribed_aero_fixed_ymd
* prescribed_aero_tod --> prescribed_aero_fixed_tod
* aerodep_flx_ymd --> aerodep_flx_cycle_yr
aerodep_flx_fixed_ymd
* aerodep_flx_tod --> aerodep_flx_fixed_tod
* prescribed_ghg_ymd --> prescribed_ghg_cycle_yr
prescribed_ghg_fixed_ymd
* prescribed_ghg_tod --> prescribed_ghg_fixed_tod
* prescribed_ozone_ymd --> prescribed_ozone_cycle_yr
prescribed_ozone_fixed_ymd
* prescribed_ozone_tod --> prescribed_ozone_fixed_tod
* prescribed_volcaero_ymd --> prescribed_volcaero_cycle_yr
prescribed_volcaero_fixed_ymd
* prescribed_volcaero_tod --> prescribed_volcaero_fixed_tod
* sad_date --> sad_cycle_yr
sad_fixed_ymd
sad_fixed_tod
* tgcm_ubc_ymd --> tgcm_ubc_cycle_yr
tgcm_ubc_fixed_ymd
* tgcm_ubc_tod --> tgcm_ubc_fixed_tod
* chlorine_loading_ymd --> chlorine_loading_cycle_yr
chlorine_loading_fixed_ymd
* chlorine_loading_tod --> chlorine_loading_fixed_tod
* linoz_data_ymd --> linoz_data_cycle_yr
linoz_data_fixed_ymd
* linoz_data_tod --> linoz_data_fixed_tod
* tracer_cnst_ymd --> tracer_cnst_cycle_yr
tracer_cnst_fixed_ymd
* tracer_cnst_tod --> tracer_cnst_fixed_tod
* tracer_srcs_ymd --> tracer_srcs_cycle_yr
tracer_srcs_fixed_ymd
* tracer_srcs_tod --> tracer_srcs_fixed_tod
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Brian Eaton
List all subroutines eliminated:
D models/atm/cam/bld/config_files/defaults_waccm_ghg.xml
D models/atm/cam/bld/config_files/defaults_trop_mozart.xml
D models/atm/cam/bld/config_files/defaults_trop_mam3.xml
D models/atm/cam/bld/config_files/defaults_trop_bam.xml
D models/atm/cam/bld/config_files/defaults_trop_mam7.xml
D models/atm/cam/bld/config_files/defaults_waccm_mozart.xml
D models/atm/cam/bld/config_files/defaults_super_fast_llnl_mam3.xml
D models/atm/cam/bld/config_files/defaults_trop_ghg.xml
D models/atm/cam/bld/config_files/defaults_super_fast_llnl.xml
- these were not needed and removed
D models/atm/cam/src/physics/cam/chemistry.F90
- moved to models/atm/cam/src/chemistry/pp_none
D models/atm/cam/src/chemistry/mozart/tgcm_forcing.F90
D models/atm/cam/src/chemistry/mozart/iondrag.F90
- iondrag.F90 moved to physics/waccm/
- tgcm_forcing.F90 replaced by physics/waccm/waccm_forcing.F90
List all subroutines added and what they do:
A models/atm/cam/bld/config_files/defaults_waccm.xml
- configure defaults in this file are used for all veriation of waccm
A models/atm/cam/bld/namelist_files/use_cases/cam4_bam_radpsv_geos5.xml
A models/atm/cam/bld/namelist_files/use_cases/sd_waccm_tslt_geos5.xml
A models/atm/cam/bld/namelist_files/use_cases/2000_cam4_trop_chem.xml
A models/atm/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp26.xml
A models/atm/cam/bld/namelist_files/use_cases/cam4_chem_radpsv_geos5.xml
A models/atm/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp85.xml
- added build-namelist use cases
A models/atm/cam/src/physics/waccm/waccm_forcing.F90
- for new waccm_ghg forcing data
A models/atm/cam/src/physics/waccm/iondrag.F90
- moved from models/atm/cam/src/chemistry/mozart
A models/atm/cam/src/chemistry/modal_aero/mo_sethet.F90
- old mo_sethet.F90 moved here so that trop_mam3 is bit-for-bit unchanged
A models/atm/cam/src/chemistry/pp_trop_bam/chem_mech.doc
A models/atm/cam/src/chemistry/pp_trop_bam/chem_mech.in
A models/atm/cam/src/chemistry/pp_trop_mozart/chem_mech.doc
A models/atm/cam/src/chemistry/pp_trop_mozart/chem_mech.in
A models/atm/cam/src/chemistry/pp_trop_mam3/chem_mech.doc
A models/atm/cam/src/chemistry/pp_trop_mam3/chem_mech.in
A models/atm/cam/src/chemistry/pp_trop_mam7/chem_mech.doc
A models/atm/cam/src/chemistry/pp_trop_mam7/chem_mech.in
A models/atm/cam/src/chemistry/pp_waccm_mozart/chem_mech.doc
A models/atm/cam/src/chemistry/pp_waccm_mozart/chem_mech.in
A models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/chem_mech.doc
A models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/chem_mech.in
A models/atm/cam/src/chemistry/pp_trop_ghg/chem_mech.doc
A models/atm/cam/src/chemistry/pp_trop_ghg/chem_mech.in
A models/atm/cam/src/chemistry/pp_super_fast_llnl/chem_mech.doc
A models/atm/cam/src/chemistry/pp_super_fast_llnl/chem_mech.in
A models/atm/cam/src/chemistry/pp_waccm_mozart_v1/chem_mech.doc
A models/atm/cam/src/chemistry/pp_waccm_mozart_v1/chem_mech.in
A models/atm/cam/src/chemistry/pp_waccm_ghg/chem_mech.doc
A models/atm/cam/src/chemistry/pp_waccm_ghg/chem_mech.in
- added chem package documents
A models/atm/cam/src/chemistry/pp_waccm_mozart_v1
A models/atm/cam/src/chemistry/pp_waccm_mozart_v1/mo_lu_solve.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_v1/mo_setrxt.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_v1/mo_photo.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_v1/mo_adjrxt.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_v1/mo_imp_sol.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_v1/chem_mods.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_v1/mo_prod_loss.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_v1/mo_lin_matrix.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_v1/mo_sad.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_v1/mo_jeuv.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_v1/mo_strato_rates.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_v1/m_rxt_id.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_v1/mo_phtadj.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_v1/mo_nln_matrix.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_v1/m_het_id.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_v1/mo_lu_factor.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_v1/mo_indprd.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_v1/mo_sim_dat.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart_v1/m_spc_id.F90
- this was added for backwards compatibility (b4b same as previous waccm_mozart)
A models/atm/cam/src/chemistry/pp_none
A models/atm/cam/src/chemistry/pp_none/mo_lu_solve.F90
A models/atm/cam/src/chemistry/pp_none/mo_setrxt.F90
A models/atm/cam/src/chemistry/pp_none/mo_phtadj.F90
A models/atm/cam/src/chemistry/pp_none/mo_nln_matrix.F90
A models/atm/cam/src/chemistry/pp_none/chemistry.F90
A models/atm/cam/src/chemistry/pp_none/mo_adjrxt.F90
A models/atm/cam/src/chemistry/pp_none/mo_lu_factor.F90
A models/atm/cam/src/chemistry/pp_none/mo_indprd.F90
A models/atm/cam/src/chemistry/pp_none/m_spc_id.F90
A models/atm/cam/src/chemistry/pp_none/mo_imp_sol.F90
A models/atm/cam/src/chemistry/pp_none/chem_mods.F90
A models/atm/cam/src/chemistry/pp_none/mo_prod_loss.F90
A models/atm/cam/src/chemistry/pp_none/mo_lin_matrix.F90
- this chem pkg was added to inable removal of cppdefs
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/config_files/wm1.9c4h
- use waccm_mozart_v1 chemistry package for this test
M models/atm/cam/bld/configure
- copy chemistry mechanism docs to $cam_bld
- added waccm_mozart_v1 chemistry package (same as the CMIP version of waccm_mozart)
- reference only defaults_waccm.xml config defaults files
- references to TROPCHEM and SUPERFASTLLNL cppdefs removed
M models/atm/cam/bld/config_files/definition.xml
- waccm_mozart_v1 chem option added
- added chem_proc_src chemistry preprocessor generated source code directory path
- new default emissions files for trop_mozart
- defaults added for waccm_mozart_v1
- set cam_chempkg to 'waccm_mozart' for the waccm_mozart_v1 mechanism
M models/atm/cam/bld/build-namelist
- chem_rad_passive --> atm_dep_flux=.false.
- airpl_emis_file replaced with ext frc files
- offline_dyn --> do_tms=.false.
- cftgcm --> waccm_forcing_file
- CB2 and OC2 default emis files added in addition to HCN,CH3CN,C2H2,CH3COOH for trop_mozart
- changes in dep lists for new trop_mozart mechanism
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- replaced *_ymd namelist variables with *_cycle_yr
- replaced cftgcm defaults with waccm_forcing defaults for waccm_ghg
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- met_data_path namelist added
- waccm_forcing_* namelist options added for waccm ghg data
-- itgcmcyc removed
-- cftgcm removed
- data-timing namelist changes (see above).
M models/atm/cam/bld/namelist_files/use_cases/1850_cam4_trop_bam.xml
M models/atm/cam/bld/namelist_files/use_cases/2005_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/aquaplanet_cam3.xml
M models/atm/cam/bld/namelist_files/use_cases/aquaplanet_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/aquaplanet_cam5.xml
M models/atm/cam/bld/namelist_files/use_cases/2005-2100_cam4_rcp85v2.xml
M models/atm/cam/bld/namelist_files/use_cases/2005-2100_cam4_rcp60.xml
M models/atm/cam/bld/namelist_files/use_cases/1850-2005_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/1850-2005_cam5.xml
M models/atm/cam/bld/namelist_files/use_cases/2005-2100_cam4_rcp26.xml
M models/atm/cam/bld/namelist_files/use_cases/2005-2100_cam4_rcp45.xml
M models/atm/cam/bld/namelist_files/use_cases/2005-2100_cam4_rcp85.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1850-2005_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2000_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/1850-2005_cam4_bgc.xml
M models/atm/cam/bld/namelist_files/use_cases/2000_cam4_super_fast_llnl.xml
M models/atm/cam/bld/namelist_files/use_cases/1850_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/1850-2005_cam4_super_fast_llnl.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp45.xml
M models/atm/cam/bld/namelist_files/use_cases/1850_cam5.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1974-2005_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1850_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/1850_cam4_super_fast_llnl.xml
M models/atm/cam/bld/namelist_files/use_cases/1850_cam4_2xco2.xml
M models/atm/cam/bld/namelist_files/use_cases/1850_cam5_super_fast_llnl.xml
- mods for the data-timing namelist changes
M models/atm/cam/bld/perl5lib/Build/ChemPreprocess.pm
- copy over preprocessor input and *.doc files to $cam_bld directory
- set $chem_proc_src dir when preprocessor is invoked
- clean up
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
- use updated preprocessor which has corrections to the chem prod/loss rate diagnostics
M models/atm/cam/src/control/runtime_opts.F90
- removed itgcmcyc and cftgcm
and added call to waccm_forcing_readnl
- broadcast nlte_use_mo
- met_data_path namelist variable added
M models/atm/cam/src/physics/cam/tropopause.F90
- TROPCHEM cppdefs removed
- history field names changed:
'chem_trop' --> 'hstobie_trop'
'chem_trop_linoz' --> 'hstobie_linoz'
'chem_trop_tropop' --> 'hstobie_tropop'
M models/atm/cam/src/physics/cam/advnce.F90
M models/atm/cam/src/physics/cam/radheat.F90
- radheat_timestep_init interface changed for new waccm_ghg
M models/atm/cam/src/physics/waccm/nlte_lw.F90
M models/atm/cam/src/physics/waccm/radheat.F90
- changes for waccm_ghg
M models/atm/cam/src/physics/cam/tphysac.F90
M models/atm/cam/src/physics/cam/tphysbc.F90
M models/atm/cam/src/physics/cam/physpkg.F90
M models/atm/cam/src/chemistry/bulk_aero/aerosol_intr.F90
- WACCM_MOZART and TROPCHEM cppdefs removed
M models/atm/cam/src/physics/cam/phys_control.F90
- cam_chempkg_out added to getopts
M models/atm/cam/src/physics/cam/cam_diagnostics.F90
- tidal diagnostics added
M models/atm/cam/src/physics/cam/tidal_diag.F90
- get_tidal_coeffs subroutine added
M models/atm/cam/src/physics/cam/chem_surfvals.F90
- data-timing namelist changes
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_setrxt.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/chem_mods.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/m_het_id.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/m_spc_id.F90
- changed trop_mozart chem mechanism
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_bam/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_trop_ghg/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_imp_sol.F90
M models/atm/cam/src/chemistry/pp_waccm_ghg/mo_imp_sol.F90
- fixed bug in chem prod/loss diagnostics
M models/atm/cam/src/chemistry/utils/aerodep_flx.F90
M models/atm/cam/src/chemistry/utils/prescribed_aero.F90
M models/atm/cam/src/chemistry/utils/prescribed_ozone.F90
M models/atm/cam/src/chemistry/utils/mo_flbc.F90
M models/atm/cam/src/chemistry/utils/prescribed_ghg.F90
M models/atm/cam/src/chemistry/utils/tracer_data.F90
M models/atm/cam/src/chemistry/utils/aircraft_emit.F90
M models/atm/cam/src/chemistry/utils/prescribed_volcaero.F90
M models/atm/cam/src/chemistry/utils/m_types.F90
M models/atm/cam/src/chemistry/mozart/chlorine_loading_data.F90
M models/atm/cam/src/chemistry/mozart/mo_tgcm_ubc.F90
M models/atm/cam/src/chemistry/mozart/upper_bc.F90
M models/atm/cam/src/chemistry/mozart/mo_chemini.F90
M models/atm/cam/src/chemistry/mozart/tracer_cnst.F90
M models/atm/cam/src/chemistry/mozart/tracer_srcs.F90
M models/atm/cam/src/chemistry/mozart/mo_extfrc.F90
M models/atm/cam/src/chemistry/mozart/linoz_data.F90
M models/atm/cam/src/chemistry/mozart/mo_strato_sad.F90
- data-timing namelist changes
M models/atm/cam/src/chemistry/bulk_aero/dust_sediment_mod.F90
- replaced "stop" statement with call to endrun
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_setrxt.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/chem_mods.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_indprd.F90
- waccm mech changes (added CO2 reactions near upper boundary)
M models/atm/cam/src/chemistry/pp_super_fast_llnl/chem_mods.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_indprd.F90
- added ext frcing of NO for lightning production
M models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
- set sulfate passed to usrrxt from vmr(so4_ndx) if SO4 is active
- corrected units on rxt (tag_names) history feilds
- call set_fstrat_h2o only if h2o is not chem active
M models/atm/cam/src/chemistry/mozart/mo_airplane.F90
M models/atm/cam/src/chemistry/mozart/mo_setext.F90
- moved the NO lightning code outside the if (has_airpl_src) block
M models/atm/cam/src/chemistry/mozart/mo_lightning.F90
- added has_no_lightning_prod var so that cppdefs can be removed from physpkg.F90
M models/atm/cam/src/chemistry/mozart/spedata.F90
- removed the "spe_restart" option
- no longer save the spe data file pathname in restart file
M models/atm/cam/src/chemistry/mozart/mo_strato_rates.F90
- if mixing ratios are small (< 1e-16) then rates are set to zero
M models/atm/cam/src/chemistry/mozart/mo_chm_diags.F90
- NJEUV increased to 26 (neuv) for the new waccm upper boundary jeuv CO2 photolysis
- corrected AREA, MASS, and WD_* (vert integrated wet dep rate) diagnostics
M models/atm/cam/src/chemistry/mozart/mo_setinv.F90
- added *_dens and *_vmr history fields
M models/atm/cam/src/chemistry/mozart/chemistry.F90
- data-timing namelist changes
- "chem_rad_passive" switch added -> sets ptend%lq(1) = .false. so that the
chemistry does not change water vapor
- removed the spedata restart calls
- removed cppdefs that set chem_name
- chem_is now invokes phys_control->cam_chempkg_is function
M models/atm/cam/src/chemistry/mozart/mo_drydep.F90
- added code for HCN, CH3CN, HCOOH
- if offline-dyn use meteorology TS to determine icefrac which is
applied only when drydep_method = 'xactive_lnd'
- HCN and CH3CN zero over land (only ocnice_dvel applied)
- code corrected for Carbon aerosols
- CO tags added
M models/atm/cam/src/chemistry/mozart/mo_photo.F90
- max zenith for waccm 94.0 --> 97.01 degress
M models/atm/cam/src/chemistry/mozart/mo_sethet.F90
- Louisa's wet dep changes
- check for unset het rates added
M models/atm/cam/src/chemistry/mozart/mo_usrrxt.F90
- CO tags added
- added check for usr_N2O5_aer if mech has het1 to avoid double counting
in tropospheric aerosol rates
M models/atm/cam/src/chemistry/mozart/mo_jeuv.F90
- CO2 photolysis added (near upper boundary)
M models/atm/cam/src/chemistry/mozart/mo_srf_emissions.F90
- data-timing namelist changes
- flexable units of input data
M models/atm/cam/src/dynamics/fv/ctem.F90
- "defined on ilev" added to description of 2D outputs
M models/atm/cam/src/dynamics/fv/cd_core.F90
M models/atm/cam/src/dynamics/fv/metdata.F90
- offline dyn changes
- can read bottom portion of met data
- added code to determine icefrac from surface temperature
M SVN_EXTERNAL_DIRECTORIES
- use driver tag drvseq3_1_51 and CLM tag clm4_0_26
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
033 bl336 TBL.sh wm4c4h outfrq3s+waccm_1850_cam4 9s .......................FAIL! rc= 7
This is expected to fail due to the changes in waccm_mozart chemistry mechanism
045 bl373 TBL.sh fm1.9c4dh outfrq3s 9s ....................................FAIL! rc= 7
This is expected to fail due to the changes in trop_mozart chemistry mechanism,
dry deposition and wet deposition
050 bl374 TBL.sh fs1.9c4dh outfrq3s+1850-2005_cam4_super_fast_llnl 9s .....FAIL! rc= 7
This is expected to failure due to the inclusion of lightning produced NOx
060 bl379 TBL.sh wm1.9c4h outfrq24h+waccm_1850_cam4 2d ....................FAIL! rc= 7
This failure is cause by changes in dry deposition. If the drydep_method were set to
'xactive_atm' then this baseline test would pass since this uses waccm_mozart_v1 which
has exactly the same mechanism as the previous waccm_mozart
edinburgh/lf95:
035 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7
This is expected to fail due to changes in waccm_ghg forcing data
edinburgh/pgi:
035 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7
This is expected to fail due to changes in waccm_ghg forcing data
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_0_50
Originator(s): Jim Edwards
Date: March 30, 2011
One-line Summary: Bug fixes for cosp and inithist
Purpose of changes: bug fixes, code cleanup
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes:
Code reviewed by: Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
models/atm/cam/src/control/cam_history_support.F90 : field_copy
provides an overload on the = operator to properly copy the
field_info derived type
List all existing files that have been modified, and describe the changes:
models/atm/cam/test/system/TER.sh
added code to prevent inithist file from initial runs with
inithist='endofrun' from being compared
models/atm/cam/test/system/nl_files/ghgrmp
models/atm/cam/test/system/nl_files/outfrq3s
added inithist='endofrun' to namelist to capture writting inithist
after a restart
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
updated some of the homme init files
models/atm/cam/src/control/cam_history_buffers.F90
models/atm/cam/src/control/cam_history_support.F90
models/atm/cam/src/control/cam_history.F90
Added support for individual fields to define and use their own fill values, replaced some cosp
bounds variables improperly removed in cam5_0_47, fixed a bug that caused inithist files to fail
when trying to write after a restart. Removed global declarations of plat,plon,plev and plevp.
Changed the variable time_bnds written to inithist files to properly reflect time of files, this
will cause some baseline tests to fail.
models/atm/cam/src/physics/cam/cospsimulator_intr.F90
Added support for cosp fillvalues to be used in cam_history instead of overwritting one fillvalue
with another.
models/atm/cam/src/physics/cam/tropopause.F90
models/atm/cam/src/physics/cam/radiation.F90
models/atm/cam/src/physics/cam/aer_rad_props.F90
models/atm/cam/src/physics/cam/radiation_data.F90
models/atm/cam/src/physics/cam/modal_aer_opt.F90
models/atm/cam/src/physics/cam/microp_driver.F90
models/atm/cam/src/physics/cam/cldwat2m_micro.F90
models/atm/cam/src/physics/rrtmg/oldcloud.F90
models/atm/cam/src/physics/rrtmg/radiation.F90
models/atm/cam/src/dynamics/fv/inidat.F90
Moved references to fillvalue from cam_history.F90 to cam_history_support.F90
models/atm/cam/src/dynamics/homme/native_mapping.F90
Modified output mapping file to better conform to scrip/esmf standards
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All pass except:
004 bl132 TBL.sh e48c4dh ghgrmp+1850_cam4 9s ..............................FAIL! rc= 7 at Thu Mar 31 07:44:32 MDT 2011
006 bl133 TBL.sh e48c4paqdm aqpgro+aquaplanet_cam4 3s .....................FAIL! rc= 7 at Thu Mar 31 07:44:34 MDT 2011
013 bl330 TBL.sh f4c5paqdh aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Thu Mar 31 07:44:36 MDT 2011
017 bl331 TBL.sh f4c4dh co2rmp+1850_cam4 9s ...............................FAIL! rc= 7 at Thu Mar 31 07:44:44 MDT 2011
021 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...............................FAIL! rc= 7 at Thu Mar 31 07:44:52 MDT 2011
023 bl333 TBL.sh f4c4paqdh aqpgro+aquaplanet_cam4 3s ......................FAIL! rc= 7 at Thu Mar 31 07:44:53 MDT 2011
033 bl336 TBL.sh wm4c4h outfrq3s+waccm_1850_cam4 9s .......................FAIL! rc= 7 at Thu Mar 31 07:45:19 MDT 2011
045 bl373 TBL.sh fm1.9c4dh outfrq3s 9s ....................................FAIL! rc= 7 at Thu Mar 31 07:45:39 MDT 2011
050 bl374 TBL.sh fs1.9c4dh outfrq3s+1850-2005_cam4_super_fast_llnl 9s .....FAIL! rc= 7 at Thu Mar 31 07:45:54 MDT 2011
054 bl387 TBL.sh f1.9c4dm outfrq3s+1850-2005_cam4 9s ......................FAIL! rc= 7 at Thu Mar 31 07:46:01 MDT 2011
066 bl751 TBL.sh h30c4aqdm outfrq3s+aquaplanet_cam4 9s ....................FAIL! rc= 7 at Thu Mar 31 08:04:48 MDT 2011
edinburgh/lf95: All pass except:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Thu Mar 31 08:17:23 MDT 2011
009 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Thu Mar 31 08:17:25 MDT 2011
011 bl113 TBL.sh e8c5paqdm aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Thu Mar 31 08:17:25 MDT 2011
015 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Thu Mar 31 08:17:28 MDT 2011
022 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Thu Mar 31 08:17:30 MDT 2011
030 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Thu Mar 31 09:08:13 MDT 2011
032 bl313 TBL.sh f10c3aqdm outfrq3s+aquaplanet_cam3 3s ....................FAIL! rc= 7 at Thu Mar 31 09:08:14 MDT 2011
035 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7 at Thu Mar 31 09:08:17 MDT 2011
038 bl315 TBL.sh f10c3dm outfrq3s+cam3 9s .................................FAIL! rc= 7 at Thu Mar 31 09:08:18 MDT 2011
040 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 7 at Thu Mar 31 09:08:19 MDT 2011
edinburgh/pgi : ALL pass except:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Thu Mar 31 08:37:53 MDT 2011
009 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Thu Mar 31 08:37:54 MDT 2011
011 bl113 TBL.sh e8c5paqdm aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Thu Mar 31 08:37:55 MDT 2011
015 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Thu Mar 31 08:37:57 MDT 2011
022 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Thu Mar 31 08:37:59 MDT 2011
030 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Thu Mar 31 08:55:29 MDT 2011
032 bl313 TBL.sh f10c3aqdm outfrq3s+aquaplanet_cam3 3s ....................FAIL! rc= 7 at Thu Mar 31 08:55:30 MDT 2011
035 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7 at Thu Mar 31 08:55:33 MDT 2011
038 bl315 TBL.sh f10c3dm outfrq3s+cam3 9s .................................FAIL! rc= 7 at Thu Mar 31 08:55:34 MDT 2011
040 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 7 at Thu Mar 31 08:55:35 MDT 2011
043 bl317 TBL.sh f10c5cdm outfrq3s 9s .....................................FAIL! rc= 7 at Thu Mar 31 08:55:41 MDT 2011
046 bl319 TBL.sh fn10c5dm outfrq3s 9s .....................................FAIL! rc= 7 at Thu Mar 31 08:55:44 MDT 2011
052 bl734 TBL.sh hn16c5aqdm outfrq3s+aquaplanet_cam5 9s ...................FAIL! rc= 7 at Thu Mar 31 08:56:04 MDT 2011
All results are bfb, tests failed due to changes in the inithist file - for ghgrmp and outfrq3s an additional file is
present which was not in the baseline. The others failed because the variable time_bnds was corrected to
properly reflect the (instantaneous) bounds of the inithist file and because the nlon and wnummax variables
were removed.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_0_49
Originator(s): Edwards, Taylor, Mirin
Date: 03/21/2011
One-line Summary: homme dycore updates, New column output method
Purpose of changes: Developement of homme dycore and related
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Additional homme hgrid support
Describe any changes made to the namelist:
updated some defaults for high resolution
Exposed additional homme namelist variables to CAM's build-namelist:
homme_ftype: selects type of forcing (adjustment, tendencies or
hybrid
phys_tscale: HOMME's forcing allows for the the physics dt
(phys_tscale) to be different than the
physics update frequency (dtime)
default is phys_tscale=dtime
nu_q hyper-viscosity for tracers (default=nu)
qsplit controlls tracer/dynamics time step splitting
List any changes to the defaults for the boundary datasets:
updated default topography data sets to use new
hyperviscosity-smoothed versions.
added a 128x256 eul initial dataset with 101 startdate
Describe any substantial timing or memory changes:
Code reviewed by: Edwards, Eaton
List all subroutines eliminated:
models/atm/cam/src/dynamics/homme/io_dist.F90
models/atm/cam/src/dynamics/fv/io_dist.F90
Obsolete functions to move distributed arrays to single processor IO.
List all subroutines added and what they do:
two new modules added - they are data processing utilities which
rely on both CAM and HOMME infrastructure so included here but
not used for normal simulations.
models/atm/cam/src/dynamics/homme/native_mapping.F90
Allows the creation of scrip format mapping files based on the
homme dycore basis functions. These are smoother than bilinear
interpolation but may not be sign preserving.
models/atm/cam/src/dynamics/homme/nctopo_util_mod.F90
Applys hyper-viscosity smoothing to PHIS, SGH, SGH30 fields,
writes smoothed fields to history files as PHIS_SM, SGH_SM and SGH30_SM
List all existing files that have been modified, and describe the changes:
models/atm/cam/test/system/nl_files/ghgrmp
Added a third history file output with new column data format but
same data as second history file
models/atm/cam/bld/config_files/horiz_grid.xml
models/atm/cam/bld/build-namelist
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
models/atm/cam/bld/namelist_files/namelist_definition.xml
Added support for ne16np8 and ne30np8 hgrids
Updated default topo files for ne30np4, ne60np4, ne120np4 and ne240np4 hgrids
Added support for new column output method, method is triggered with namelist
variable collect_column_output = .true. this is an array of size 6, one for
each history file.
models/atm/cam/src/control/cam_history.F90
models/atm/cam/src/control/cam_history_support.F90
Added an interface to interp_mod so that information about the interpolation
method is written to the history files.
Added an option to collect column output into a single field and
output using the ncol format which is defined for unstructured grids.
This is a much faster method when you are outputing several columns
models/atm/cam/src/utils/spmd_utils.F90
Refactored in an effort to move toward removeing mpishorthand.F and
making this the sole location in cam where mpif.h is included.
models/atm/cam/src/dynamics/homme/pmgrid.F90
models/atm/cam/src/dynamics/homme/dyn_grid.F90
models/atm/cam/src/dynamics/homme/dp_coupling.F90
models/atm/cam/src/dynamics/homme/dyn_comp.F90
models/atm/cam/src/dynamics/homme/inidat.F90
models/atm/cam/src/dynamics/homme/stepon.F90
models/atm/cam/src/dynamics/homme/restart_dynamics.F90
models/atm/cam/src/dynamics/homme/spmd_dyn.F90
models/atm/cam/src/dynamics/homme/inital.F90
Added support for physics to use more mpi tasks than dynamics,
removed SPMD cpp ifdefs , remove use of mpishorthand module
models/atm/cam/src/dynamics/homme/interp_mod.F90
Added code to add attributes to the history file indicating
the options used in interpolation, fixed a bug in vector
field interpolation.
models/atm/cam/src/dynamics/fv/dyn_grid.F90
models/atm/cam/src/dynamics/sld/dyn_grid.F90
models/atm/cam/src/dynamics/eul/dyn_grid.F90
Added stubs for get_block_ldof_d - plan to move code
models/atm/cam/src/dynamics/fv/interp_mod.F90
models/atm/cam/src/dynamics/sld/interp_mod.F90
models/atm/cam/src/dynamics/eul/interp_mod.F90
update stub api to support unstructured grids
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All pass except
004 bl132 TBL.sh e48c4dh ghgrmp+1850_cam4 9s ..............................FAIL! rc= 7 at Tue Mar 22 11:21:47 MDT 2011
021 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...............................FAIL! rc= 7 at Tue Mar 22 12:01:36 MDT 2011
063 bl731 TBL.sh h16c3aqdh aqua+aquaplanet_cam3 9s ....................... FAIL
066 bl751 TBL.sh h30c4aqdm outfrq3s+aquaplanet_cam4 9s ....................FAIL
edinburgh/lf95: All pass except:
009 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Tue Mar 22 11:32:18 MDT 2011
030 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Tue Mar 22 15:59:09 MDT 2011
edinburgh/pgi : All pass except:
009 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Tue Mar 22 11:18:46 MDT 2011
030 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Tue Mar 22 12:00:41 MDT 2011
052 bl734 TBL.sh hn16c5aqdm outfrq3s+aquaplanet_cam5 9s ...................FAIL
ghgrmp baseline tests fail due to the addition of a history file with collected column output
all existing history files were comfirmed to be bfb with baseline
HOMME default output was changed from interpolated to native grid
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_0_48
Originator(s): mvr,bundy,jedwards,dennis,hannay,eaton
Date: Wed Mar 16 2011
One-line Summary: added functionality for fv 2.5x3.33 resolution; small fixes
and cleanup to test scripts; added support for 0.25deg cam-homme; tweak of
cam5 tuning parameters; bug fixes to flushing of output buffers; updated
externals definitions for component tags
=======
Purpose of changes:
-machine jaguar was decommissioned 3/7/11 so support was stripped from test scripts
-adjusted tuning parameters dust_emis_fact and cldfrc_rhminl for cam5_1...tuning
was done at 1deg but applies to all resolutions for now - tuning specific to
other resolutions may follow
-externals updated from cesm1_0_beta13 to cesm1_0_beta16
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets:
- added default files for new 2.5x3.33 resolution
- added default files for 0.25deg cam-homme
Describe any substantial timing or memory changes: not checked
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do:
A models/atm/cam/test/system/config_files/f2.5c5dm
- new configuration options test file for 2.5x3.33
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/TCB.sh
M models/atm/cam/test/system/TDD.sh
M models/atm/cam/test/system/TCB_ccsm.sh
M models/atm/cam/test/system/TBL.sh
M models/atm/cam/test/system/TBR.sh
M models/atm/cam/test/system/TPF.sh
M models/atm/cam/test/system/TER.sh
M models/atm/cam/test/system/TEQ_ccsm.sh
M models/atm/cam/test/system/TER_ccsm.sh
M models/atm/cam/test/system/TSM_ccsm.sh
M models/atm/cam/test/system/TMC.sh
M models/atm/cam/test/system/TBL_ccsm.sh
M models/atm/cam/test/system/TSC.sh
M models/atm/cam/test/system/TEQ.sh
M models/atm/cam/test/system/TSM.sh
- fix behavior of test scripts in how they clean up after themselves
M models/atm/cam/test/system/CAM_decomp.sh
- added code for new 2.5x3.33 resolution to fv dynamics decomposition utility
M models/atm/cam/test/system/test_driver.sh
- removed test support for decommissioned machine, jaguar
- made all module loading/unloading in test scripts to refer to generic
compilers, netcdf, etc and removed the 'module purge' from lynx
M models/atm/cam/test/system/nl_files/ghgrmp
- enhanced namelist options test file with additional vars on aux history
tapes (jedwards)
M models/atm/cam/test/system/input_tests_master
- modified test definition to now run at new 2.5x3.33 resolution
M models/atm/cam/test/system/CAM_runcmnd.sh
- removed test support for decommissioned machine, jaguar
M models/atm/cam/bld/config_files/horiz_grid.xml
- added support for new 2.5x3.33 resolution
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- added default files for new 2.5x3.33 resolution
- added default files for 0.25deg cam-homme
- adjusted tuning parameters dust_emis_fact and cldfrc_rhminl for cam5
M models/atm/cam/src/control/ncdio_atm.F90
M models/atm/cam/src/utils/abortutils.F90
M models/atm/cam/src/dynamics/eul/iop.F90
- now flushing correct output buffer rather than unit 6
M models/atm/cam/src/control/cam_history.F90
- bug fix to string length problem (jedwards)
M models/atm/cam/src/utils/time_manager.F90
- added method from release branch for timemgr_set_date_time
M models/atm/cam/src/utils/cam_aqua/shr_flux_mod.F90
- change to interface to coincide with changes to share and drv code
M models/atm/cam/src/cpl_mct/atm_comp_mct.F90
M models/atm/cam/src/cpl_esmf/atm_comp_esmf.F90
- added call to flush buffers after initialization stage
M SVN_EXTERNAL_DIRECTORIES
M .
- new external component tags for drv, clm, cice, docn, share, timing,
scripts, slnd, sice, socn, sglc - consistent with update from cesm1_0_beta13
to cesm1_0_beta16
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Tue Mar 15 16:35:21 MDT 2011
013 bl330 TBL.sh f4c5paqdh aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Tue Mar 15 16:36:56 MDT 2011
021 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...............................FAIL! rc= 7 at Tue Mar 15 16:49:59 MDT 2011
036 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ......................................FAIL! rc= 7 at Tue Mar 15 17:14:53 MDT 2011
040 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .....................................FAIL! rc= 7 at Tue Mar 15 17:28:40 MDT 2011
047 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...........................FAIL! rc= 7 at Tue Mar 15 17:56:35 MDT 2011
072 bl992 TBL_ccsm.sh f19_g16 E 2d ........................................FAIL! rc= 7 at Tue Mar 15 19:20:21 MDT 2011
- cam5 baselines were expected to fail due to changes in tuning parameters
- bl992 was expected to fail due to changes to component tags between cesm1_0_beta13 and cesm1_0_beta16, namely:
"landfrac is no longer sent at run phase" (beta15) and "driver mapping mods" (beta16)
edinburgh/lf95:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Tue Mar 15 14:47:47 MDT 2011
009 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Tue Mar 15 15:00:24 MDT 2011
011 bl113 TBL.sh e8c5paqdm aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Tue Mar 15 15:07:27 MDT 2011
015 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Tue Mar 15 15:13:55 MDT 2011
022 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Tue Mar 15 15:41:20 MDT 2011
030 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Tue Mar 15 18:13:40 MDT 2011
040 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 7 at Tue Mar 15 18:47:52 MDT 2011
- cam5 baselines were expected to fail due to changes in tuning parameters
- bl312 was expected to fail due to it changing to 2.5x3.33 resolution which did not exist in baseline code
edinburgh/pgi:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Wed Mar 16 10:30:16 MDT 2011
009 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Wed Mar 16 10:35:59 MDT 2011
011 bl113 TBL.sh e8c5paqdm aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Wed Mar 16 10:39:20 MDT 2011
015 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Wed Mar 16 10:42:43 MDT 2011
022 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Wed Mar 16 10:52:47 MDT 2011
030 bl312 TBL.sh f2.5c5dm ghgrmp 9s .......................................FAIL! rc= 7 at Wed Mar 16 11:17:07 MDT 2011
040 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 7 at Wed Mar 16 11:28:02 MDT 2011
043 bl317 TBL.sh f10c5cdm outfrq3s 9s .....................................FAIL! rc= 7 at Wed Mar 16 11:28:08 MDT 2011
046 bl319 TBL.sh fn10c5dm outfrq3s 9s .....................................FAIL! rc= 7 at Wed Mar 16 11:28:09 MDT 2011
049 bl320 TBL.sh fn10c5dm rad_diag 9s .....................................FAIL! rc= 7 at Wed Mar 16 11:28:11 MDT 2011
052 bl734 TBL.sh hn16c5aqdm outfrq3s+aquaplanet_cam5 9s ...................FAIL! rc= 7 at Wed Mar 16 11:29:31 MDT 2011
- cam5 baselines were expected to fail due to changes in tuning parameters
- bl312 was expected to fail due to it changing to 2.5x3.33 resolution which did not exist in baseline code
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: b4b except for cam5, where all configurations had answers change due to
the tweaking of tuning parameters...these produce a new climate and the diagnostics can be found here:
http://wwwint.cgd.ucar.edu/project/diagnostics/cesm-1.0/amwg/hannay/b40_1850_1d_b08c5cn_138j/b40_1850_1d_b08c5cn_138j-obs_yr200-209/
======================================================================
===============================================================
>>>>>>> .merge-right.r27350
Tag name: cam5_0_47
Originator(s): jedwards, eaton
Date: Mar, 08 2011
One-line Summary: Refactoring cam history
Purpose of changes:
. Refactoring in cam history to allow for extra dimensions beyond the
current spatial/temporal ones. Previously the fields in a history file
time sample only had (lon,lat,lev,time) or (ncol,lev,time) dimensions.
The "addfld" calls allowed lev to be set to 1, plev or plevp. To work
around this constraint in the past optional flags were added to the
addfld subroutine that allowed more values of lev, but this workaround
did not provide extra dimensions. In cases where multiple non-horizontal
dimensions were needed, e.g., level and optical depth, or level and
subcolumn, the technique employed was to combine multiple dimensions into
a single mixed dimension. We have eliminated the need to do that.
Optional arguments have been added to the addfld subroutine which allow
defining multiple dimensions rather than just a single "lev" dimension.
. The support for 4-byte history buffers has been removed. This feature
added substantial complexity to cam_history, but no demonstrated benefit.
It also was not part of the regression testing, so we don't know whether
it was working correctly.
. The old isccp cloudsimulator code was removed. The latest ISCCP
simulator is available as part of the COSP package.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Edwards, Eaton
List all subroutines eliminated:
models/atm/cam/src/physics/cam/cloudsimulator.F90
models/atm/cam/src/physics/cam/cloudsimulator_38.F90
models/atm/cam/src/physics/cam/icarus_scops.F90
models/atm/cam/src/physics/cam/icarus_scops_38.F90
. The old ISCCP simulator code has been removed.
List all subroutines added and what they do:
models/atm/cam/src/control/cam_history_buffers.F90
models/atm/cam/src/control/cam_history_support.F90
- These files seperate out some of the distinct functionality that was
in cam_history.
List all existing files that have been modified, and describe the changes:
models/atm/cam/test/system/CAM_compare.sh
Added -d option so that cprnc can compare sat_hist files
models/atm/cam/test/system/test_driver.sh
Point to new cprnc location on bluefire and edinburgh, this change should
be reverted when the new cprnc code is moved to the standard location
models/atm/cam/test/system/nl_files/outfrq1m
Change ndens from 1 to 2 so that 4 byte history files are included in the test
suite.
models/atm/cam/test/system/input_tests_master
Modify test 318 to incude sat history file and local time buf tests
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
models/atm/cam/bld/namelist_files/namelist_definition.xml
models/atm/cam/bld/build-namelist
Remove isccp code, add drydep_srf_file for homme, remove flags for 4-byte history buffers
models/atm/cam/bld/cam.cpl7.template
Remove phys_loadbalance setting - this is handled by cams configure script and varies with dycore
models/atm/cam/SVN_EXTERNAL_DIRECTORIES
Update cprnc tag
models/atm/cam/src/utils/time_manager.F90
models/atm/cam/src/control/mpishorthand.F
Remove global save statement
models/atm/cam/src/control/runtime_opts.F90
Remove isccp code, remove 4-byte history buffer code
models/atm/cam/src/control/cam_comp.F90
models/atm/cam/src/control/camsrfexch_types.F90
Added code to deallocate derived data types used in srf exchanges
models/atm/cam/src/physics/cam/radiation.F90
models/atm/cam/src/physics/rrtmg/radiation.F90
models/atm/cam/src/control/filenames.F90
Remove isccp code
models/atm/cam/src/control/cam_restart.F90
Removed unused reference to shr_sys_getenv, cam_history mtapes is now ptapes
models/atm/cam/src/control/ioFileMod.F90
Added an optional argument to getfil, lexist, that will be set to true if the
file is found and false otherwise.
models/atm/cam/src/control/cam_history.F90
Moved buffer code to cam_history_buffers.F90, moved some shared header code to
cam_history_support.F90, removed 4-byte buffer code. Added support for more than
1 non-decomposed dimension in fields. Refactored cosp and
sat_hist code, removed isccp specific and other dead code.
models/atm/cam/src/utils/cam_pio_utils.F90
Moved code that had been moved here from cam_history to cam_history_support.F90
Added support for more than 1 non-decomposed dimension in fields.
models/atm/cam/src/utils/cam_dom/ocn_comp.F90
Removed reference to unused shr functions
models/atm/cam/src/physics/cam/cospsimulator_intr.F90
Seperated dimensions that were required to be combined by restrictions in cam_history
Added only clause to use statements.
models/atm/cam/src/chemistry/mozart/mo_drydep.F90
Improved error checking for reading drydep_srf_file used by unstructured grids
models/atm/cam/src/dynamics/homme/interp_mod.F90
Bug fix in interpolation of vector fields
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: all pass
edinburgh/lf95: all pass except:
043 bl318 TBL.sh f10c4cdm sat_hist 9s Test was modified to include sat_hist, I back ported the modified test and
ran manually with bfb results.
Tests 317 and 318 require the updated cprnc (cprnc_110307c)
edinburgh/pgi or jaguar/pgi: all pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_0_46
Originator(s): jet, P. Worley, eaton
Date:
One-line Summary: Fix spmd_comm namelist parameters for all dycores.
Purpose of changes: SPMD controlling namelist parameters were broken for all dycores but FV
by my cam5_0_25 tag. Additionally the inline documentation for the parameters in
spmd_dyn were also screwed up. Both of these are now fixed. This tag also continues
the namelist refactor that defines and sets the default namelist values as module
data instead of using the setopts routines.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
Added a new spmd_utils namelist to atm_in called spmd_utils_nl.
This contains the spmd controlling namelist parameters
swap_comm_protocol
swap_comm_maxreq
fc_gather_flow_cntl
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:mvr
List all subroutines eliminated:
fc_gather_setopts
swap_comm_setopts
List all subroutines added and what they do:
spmd_utils_readnl - allows for namelist input for
the swap_comm/gather_flow spmd
parameters
List all existing files that have been modified, and describe the changes:
M models/atm/cam/doc/ChangeLog
M models/atm/cam/bld/namelist_files/namelist_definition.xml
Added spmd_utils_nl as the group for the swap_comm/gather_flow parameters
M models/atm/cam/src/control/runtime_opts.F90
Added call for reading new namelist
M models/atm/cam/src/utils/spmd_utils.F90
Added new subroutine to read spmd_utils_nl namelist
M models/atm/cam/src/dynamics/sld/spmd_dyn.F90
Got rid of spmd setopts calls
M models/atm/cam/src/dynamics/eul/spmd_dyn.F90
Got rid of spmd setopts calls
M models/atm/cam/src/dynamics/fv/spmd_dyn.F90
Got rid of spmd setopts calls
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All Pass
edinburgh/lf95: All Pass
edinburgh/pgi or jaguar/pgi: All Pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,BFB
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
Tag name: cam5_0_45
Originator(s): mvr,jenkay
Date: Fri Feb 18 2011
One-line Summary: fixing a problem with the weighted averages for several COSP
output variables
Purpose of changes:
- problem affects COSP answers and fix was needed before CFMIP integrations could begin.
output variables affected: MEANCLDALB_ISCCP, MEANPTOP_ISCCP, and MEANTAU_ISCCP
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: not checked
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
M models/atm/cam/src/physics/cam/cospsimulator_intr.F90
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: ALL PASS
edinburgh/lf95:
043 bl318 TBL.sh f10c4cdm outfrq3s 9s .....................................FAIL! rc= 7 at Fri Feb 18 13:06:55 MST 2011
- baselines for configurations using cosp were expected to fail - differences in the secondary history tapes only
edinburgh/pgi:
043 bl317 TBL.sh f10c5cdm outfrq3s 9s .....................................FAIL! rc= 7 at Fri Feb 18 13:07:06 MST 2011
- baselines for configurations using cosp were expected to fail - differences in the secondary history tapes only
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: b4b, except for cosp variables MEANCLDALB_ISCCP, MEANPTOP_ISCCP, and MEANTAU_ISCCP
======================================================================
===============================================================
Tag name: cam5_0_44
Originator(s): Francis Vitt
Date: 18 Feb 2011
One-line Summary: Added new history capabilities
Purpose of changes:
The following capabilities were added to CAM history to help with
comparisons to observations:
- local time averaging
- sampling along a specified track
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
To activate local time averaging use 'L' history averaging flag and set lcltod_start, and
lcltod_stop namelist variable to specify the local start and stop times (in seconds) of the
averaging region. For example:
hist_avgflag_pertape = 'L'
lcltod_start = 0
lcltod_stop = 3600
To activate sampling along a specified track use the following namelists:
sathist_track_infile - netcdf file that specifies the track(s) the sample along
sathist_fincl - list of history fields to be sampled
sathist_mfilt - number of columns output to each individual file
sathist_hfilename_spec - specifies the output file template "%c.cam2.sat.%y-%m-%d-%s.nc"
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Brian Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
A models/atm/cam/src/control/sat_hist.F90
this manages the output of columns along a specified track
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/namelist_files/namelist_definition.xml
M models/atm/cam/src/control/runtime_opts.F90
M models/atm/cam/src/control/cam_comp.F90
M models/atm/cam/src/control/cam_history.F90
changes for sampling and local time averaging
M models/atm/cam/src/dynamics/sld/dyn_grid.F90
M models/atm/cam/src/dynamics/eul/dyn_grid.F90
M models/atm/cam/src/dynamics/fv/dyn_grid.F90
made get_gcol_block_d interface the same as the dynamics/homme version
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: ALL PASS
edinburgh/lf95: ALL PASS
edinburgh/pgi: ALL PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_0_43
Originator(s): mvr,jenkay,eaton
Date: Wed Feb 16 2011
One-line Summary: cosp bug fixes/code improvement; cleanup of cam microphysics;
fix for sld configurations
Purpose of changes:
- cosp mods include:
1) reff_cosp mods per Yuying's e-mail, described in cospsimulator_intr.F90
2) I checked that all fields to run COSP off-line are available, and changed
documentation at the top of cospsimulator_intr.F90.
3) Jim Boyle found a problem when using cosp_nradsteps = 3:
"The data were saved as instantaneous - the values were saved as missing.
If there is a bug I could imagine the missing values cycling through all the
longitudes for a month and producing seemingly complete fields."
This was indeed a problem - radiation is called a number of times in a
single timestep (loop over chunks in physpkg.F90). Therefore, my cosp_cnt
counter, which was meant to advance only as timestep advanced, was advancing
within a timestep. I redid the logic in radiation.F90. cosp_cnt now has an
lchnk dimension... each chunk has its own counter.
Note: for 3), the monthly outputs from a long COSP run looked ok - I think
they are just sub-sampling the columns in a different way. So 3) is really
an answer issue more for instantaneous forecast values, and a consistency
issue for monthly averages.
4) An aside code mod not related to COSP: I added ATMEINT addfld/outfld calls
in cam_diagnostics.F90, a useful diagnostic for meridional heat transport
calculations.
- microphysics changes involve:
1. Removing unnecessary passed variables and use statements for the CAM5 cloud
microphysics (cldwat2m_micro.F90) [Hugh: Oct 31st Mods]
2. Adding functionality for a logical flag called 'sub_column' to be passed to
the CAM5 microphysics that changes how it works, but with the flag set (false)
to reproduce the standard code (cldwat2m_micro.F90)
- anything using sld dycore was broken with mods in cam5_0_41, where a
namespace issue was addressed for coupling with cism
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: not checked
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
M models/atm/cam/src/utils/sgexx.F90
M models/atm/cam/src/dynamics/sld/nmmatrix.F90
- fix for sld - was broken with mods to address namespace problems in cam5_0_41
M models/atm/cam/src/physics/cam/initindx.F90
M models/atm/cam/src/physics/cam/microp_aero.F90
M models/atm/cam/src/physics/cam/radiation.F90
M models/atm/cam/src/physics/cam/tphysbc.F90
M models/atm/cam/src/physics/cam/cam_diagnostics.F90
M models/atm/cam/src/physics/cam/stratiform.F90
M models/atm/cam/src/physics/cam/convect_shallow.F90
M models/atm/cam/src/physics/cam/macrop_driver.F90
M models/atm/cam/src/physics/cam/cospsimulator_intr.F90
M models/atm/cam/src/physics/cam/tphysac.F90
M models/atm/cam/src/physics/cam/cldwat.F90
M models/atm/cam/src/physics/rrtmg/radlw.F90
M models/atm/cam/src/physics/rrtmg/radiation.F90
M models/atm/cam/src/dynamics/fv/dyn_comp.F90
- mods for cosp work (see above)
M models/atm/cam/src/physics/cam/cldwat2m_micro.F90
MM models/atm/cam/src/physics/cam/microp_driver.F90
- mods included those for cosp work and microphysics cleanup (see above)
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: ALL PASS
edinburgh/lf95:
043 bl318 TBL.sh f10c4cdm outfrq3s 9s .....................................FAIL! rc= 7 at Tue Feb 15 13:38:13 MST 2011
- baselines for configurations turning on cosp were expected to fail - differences in the secondary history tapes only
edinburgh/pgi:
043 bl317 TBL.sh f10c5cdm outfrq3s 9s .....................................FAIL! rc= 7 at Tue Feb 15 13:21:21 MST 2011
- baselines for configurations turning on cosp were expected to fail - differences in the secondary history tapes only
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: b4b, except for cosp variables
======================================================================
===============================================================
Tag name: cam5_0_42
Originator(s): Edwards, Eaton
Date: Mon Feb 7 09:57:52 MST 2011
One-line Summary: Refactoring in rad_constituents and modal aerosol code.
Purpose of changes:
. This is another step on the path to providing prescribed modal aerosol
functionality.
. The rad_constituent interfaces have been generalized to treat the climate
and diagnostic calculations more uniformly. The use of an optional
argument to identify a diagnostic calc is replaced by a non-optional
argument which is set to zero to indicate the climate calculation, and to
1,...,N_DIAG for the diagnostic calculation. This allows us to
consolidate the separate code for climate and diagnostic calcs in the
radiation driver (rrtmg only).
. Removed qqcw array from pbuf structure and replaced with individual cloud
bourne constituant species. Code clean up including replacing goto flow
structures and redefining static variables as parameters.
. Add test for diagnostic rad calcs with BAM and rrtmg
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes:
. Removal of qqcw array reduces the number of 3D arrays in pbuf by 10.
Code reviewed by: Eaton
List all subroutines eliminated: none
List all subroutines added and what they do:
models/atm/cam/test/system/nl_files/rad_diag
List all existing files that have been modified, and describe the changes:
models/atm/cam/src/chemistry/modal_aero/modal_aero_calcsize.F90
. update qqcw access
. cleanup code structure, remove "go to"s
models/atm/cam/src/chemistry/modal_aero/modal_aero_data.F90
. move parameter settings here from the initialize module
. add methods to access qqcw fields
models/atm/cam/src/chemistry/modal_aero/modal_aero_initialize_data.F90
. add register method for adding fields to constituent arrays and pbuf.
. fix output unit
models/atm/cam/src/chemistry/modal_aero/modal_aero_coag.F90
models/atm/cam/src/chemistry/modal_aero/modal_aero_gasaerexch.F90
models/atm/cam/src/chemistry/modal_aero/modal_aero_rename.F90
models/atm/cam/src/chemistry/modal_aero/modal_aero_wateruptake.F90
. cleanup
models/atm/cam/src/chemistry/mozart/chemistry.F90
. call new modal_aero_register method
models/atm/cam/src/chemistry/mozart/mo_mass_xforms.F90
. add methods qqcw2vmr and vmr2qqcw to handle new qqcw implementation
models/atm/cam/src/physics/cam/cldwat2m_micro.F90
. cleanup unused rad_constituents use association
models/atm/cam/src/physics/cam/microp_aero.F90
models/atm/cam/src/physics/cam/microp_driver.F90
. update rad_cnst interfaces
. update qqcw access
models/atm/cam/src/chemistry/bulk_aero/aerosol_intr.F90
models/atm/cam/src/chemistry/mozart/mo_chm_diags.F90
models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
models/atm/cam/src/chemistry/mozart/mo_setsox.F90
models/atm/cam/src/chemistry/mozart/mz_aerosols_intr.F90
models/atm/cam/src/physics/cam/ndrop.F90
. update qqcw access
. cleanup
models/atm/cam/src/physics/cam/stratiform.F90
models/atm/cam/src/physics/cam/tphysbc.F90
. remove unused ifdef MODAL_AERO conditionals
models/atm/cam/src/physics/cam/aer_rad_props.F90
models/atm/cam/src/physics/cam/cospsimulator_intr.F90
models/atm/cam/src/physics/cam/modal_aer_opt.F90
models/atm/cam/src/physics/cam/rad_constituents.F90
models/atm/cam/src/physics/cam/radiation.F90
models/atm/cam/src/physics/cam/radiation_data.F90
models/atm/cam/src/physics/rrtmg/radiation.F90
models/atm/cam/src/physics/waccm/nlte_lw.F90
. update rad_cnst interfaces
models/atm/cam/test/system/input_tests_master
models/atm/cam/test/system/tests_pretag_edinburgh_pgi
. add test {sm,er,bl}320 fn10c5dm rad_diag 9s
This tests rad diagnostic calcs for bulk aerosols in rrtmg
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: all pass
edinburgh/lf95: all pass
edinburgh/pgi: all pass except
049 bl320 TBL.sh fn10c5dm rad_diag 9s .....................................FAIL! rc= 5 at Sat Feb 5 17:51:45 MST 2011
This failure is expected since it's a new test and there is no comparable
test in cam5_0_41 for the baseline comparison.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_0_41
Originator(s): mvr,jedwards,eaton
Date: Mon Jan 31 2011
One-line Summary: enabled use of intel and pathscale compilers; updated clm tag
allowed cleanup of misc.h and preproc.h; filepath changes for clm; new default
climatological datasets for prescribed aerosol, ozone; cleanup of namelist
definitions; mods to allow coupling with cism
Purpose of changes:
- new aerosol and ozone datasets were for 2000 climatology
- coupling with land ice model (cism) was hampered by namespace conflicts
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets:
- new default climatological datasets for prescribed aerosol, ozone
Describe any substantial timing or memory changes: not checked
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/test_driver.sh
- enabled testing with pathscale and intel compilers on lynx
M models/atm/cam/bld/configure
- new clm tag allowed cleanup of misc.h and preproc.h; filepath changes for clm
M models/atm/cam/bld/Makefile.in
- changes needed to enable pathscale and intel compilers
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- new default climatological datasets (2000 climatology) for prescribed aerosol
and prescribed ozone
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- cleanup of old clm entries
M models/atm/cam/bld/namelist_files/use_cases/2005_cam4.xml
- changed the prescribed aero, deposition, and ozone datasets for present day
use_case to the versions used by rcp4.5
M models/atm/cam/src/utils/time_manager.F90
M models/atm/cam/src/physics/cam/stratiform.F90
M models/atm/cam/src/physics/cam/convect_shallow.F90
- minor fixes needed to appease intel or pathscale compiler
M models/atm/cam/src/utils/sgexx.F90
- converted to a module to prevent namespace conflicts when coupled with cism
M models/atm/cam/src/dynamics/eul/settau.F90
- modified to now use new module sgexx
M SVN_EXTERNAL_DIRECTORIES
M .
- updated land tag to clm4_0_22
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Mon Jan 31 09:50:55 MST 2011
036 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ......................................FAIL! rc= 7 at Mon Jan 31 09:51:06 MST 2011
040 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .....................................FAIL! rc= 7 at Mon Jan 31 09:51:29 MST 2011
072 bl992 TBL_ccsm.sh f19_g16 E 2d ........................................FAIL! rc= 4 at Mon Jan 31 10:40:13 MST 2011
- any baseline where prescribed aerosols or prescribed ozone was used was expected to fail
edinburgh/lf95:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Thu Jan 27 12:13:18 MST 2011
009 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Thu Jan 27 12:22:16 MST 2011
015 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Thu Jan 27 12:34:43 MST 2011
022 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Thu Jan 27 12:57:50 MST 2011
030 bl312 TBL.sh f10c5dm ghgrmp 9s ........................................FAIL! rc= 7 at Thu Jan 27 13:20:08 MST 2011
035 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7 at Thu Jan 27 13:41:30 MST 2011
043 bl318 TBL.sh f10c4cdm outfrq3s 9s .....................................FAIL! rc= 7 at Thu Jan 27 13:57:43 MST 2011
- any baseline where prescribed aerosols or prescribed ozone was used was expected to fail
edinburgh/pgi:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Thu Jan 27 12:09:01 MST 2011
009 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Thu Jan 27 12:14:38 MST 2011
015 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Thu Jan 27 12:21:44 MST 2011
022 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Thu Jan 27 12:32:49 MST 2011
030 bl312 TBL.sh f10c5dm ghgrmp 9s ........................................FAIL! rc= 7 at Thu Jan 27 12:41:59 MST 2011
035 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7 at Thu Jan 27 12:49:30 MST 2011
043 bl317 TBL.sh f10c5cdm outfrq3s 9s .....................................FAIL! rc= 7 at Thu Jan 27 13:02:30 MST 2011
046 bl319 TBL.sh fn10c5dm outfrq3s 9s .....................................FAIL! rc= 7 at Thu Jan 27 13:06:58 MST 2011
049 bl734 TBL.sh hn16c5aqdm outfrq3s+aquaplanet_cam5 9s ...................FAIL! rc= 7 at Thu Jan 27 13:33:53 MST 2011
- any baseline where prescribed aerosols or prescribed ozone was used was expected to fail
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: b4b, except for configurations using default datasets for prescribed aerosols, ozone
======================================================================
===============================================================
Tag name: cam5_0_40
Originator(s): mvr,hannay
Date: Thu Jan 27 2011
One-line Summary: bug fixes and retuning of cam5 physics
Purpose of changes:
1. A fix to immersion freezing (small changes)
2. Fixing the diagnostic output FICE to trap for roundoff errors (diagnostic only)
3. Bug fix for subgrid cloud water treatment in contact nucleation (small changes)
4. Bug fix for size of snow particles used in radiation (they get smaller by a factor of 3 and more reflective: some impact).
5. Fix to guarantee the in-stratus LWC to be within the specified range even when the cloud fraction is very small (< 1.e-5 )
6 Fix to prevent model crash by dividing by near zero cloud fraction in the droplet activation routine
We also include retuning parameters as in: b40_1850_1d_b08c5cn_138j
release(2deg) Bugfix+CN (1deg)
Dcs 325 400
rhminl 0.89 0.90
c0_lnd 0.0020 0.0059
c0_ocn 0.0150 0.0450
dp1 0.04 0.10
dp2 675. 500.
div24del2flag 2 4
Notice that the 2 degree version has not been retuned and we use the
1 degree parameter for all resolution.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes: not checked
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- default settings for rhminl, c0_lnd, c0_ocn, dp1, dp2, div24del2flag
retuned for cam5 physics
M models/atm/cam/src/physics/cam/ndrop.F90
- Fix to prevent model crash by dividing by near zero cloud fraction in
the droplet activation routine
M models/atm/cam/src/physics/cam/cldwat2m_macro.F90
- bug fixes, see above
M models/atm/cam/src/physics/cam/cldwat2m_micro.F90
- rhosn, Dcs retuned; bug fixes, see above
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Wed Jan 26 11:49:08 MST 2011
013 bl330 TBL.sh f4c5paqdh aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Wed Jan 26 11:49:10 MST 2011
021 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...............................FAIL! rc= 7 at Wed Jan 26 11:49:24 MST 2011
036 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ......................................FAIL! rc= 7 at Wed Jan 26 11:50:28 MST 2011
040 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .....................................FAIL! rc= 7 at Wed Jan 26 11:51:12 MST 2011
047 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...........................FAIL! rc= 7 at Wed Jan 26 11:51:49 MST 2011
- cam5 baseline tests failed as expected
edinburgh/lf95:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Wed Jan 26 15:36:19 MST 2011
009 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Wed Jan 26 15:42:36 MST 2011
011 bl113 TBL.sh e8c5paqdm aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Wed Jan 26 15:48:01 MST 2011
015 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Wed Jan 26 15:49:49 MST 2011
022 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Wed Jan 26 16:06:20 MST 2011
030 bl312 TBL.sh f10c5dm ghgrmp 9s ........................................FAIL! rc= 7 at Wed Jan 26 16:36:45 MST 2011
040 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 7 at Wed Jan 26 17:27:13 MST 2011
- cam5 baseline tests failed as expected
edinburgh/pgi:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Wed Jan 26 15:34:11 MST 2011
009 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Wed Jan 26 15:38:35 MST 2011
011 bl113 TBL.sh e8c5paqdm aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Wed Jan 26 15:41:42 MST 2011
015 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Wed Jan 26 15:42:39 MST 2011
022 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Wed Jan 26 15:49:55 MST 2011
030 bl312 TBL.sh f10c5dm ghgrmp 9s ........................................FAIL! rc= 7 at Wed Jan 26 15:54:17 MST 2011
040 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 7 at Wed Jan 26 16:06:28 MST 2011
043 bl317 TBL.sh f10c5cdm outfrq3s 9s .....................................FAIL! rc= 7 at Wed Jan 26 16:17:26 MST 2011
046 bl319 TBL.sh fn10c5dm outfrq3s 9s .....................................FAIL! rc= 7 at Wed Jan 26 16:24:58 MST 2011
049 bl734 TBL.sh hn16c5aqdm outfrq3s+aquaplanet_cam5 9s ...................FAIL! rc= 7 at Wed Jan 26 17:01:15 MST 2011
- cam5 baseline tests failed as expected
lynx/pgi:
004 bl157 TBL.sh e64c5m ghgrmp 9s .........................................FAIL! rc= 7 at Tue Jan 25 12:58:50 MST 2011
006 bl158 TBL.sh e64c5paqm aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Tue Jan 25 12:58:52 MST 2011
012 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Tue Jan 25 12:59:05 MST 2011
021 bl372 TBL.sh f1.9c5m fvvp_lb2 9s ......................................FAIL! rc= 7 at Tue Jan 25 13:00:44 MST 2011
026 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...........................FAIL! rc= 7 at Tue Jan 25 13:01:16 MST 2011
- cam5 baseline tests failed as expected
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: new climate for cam5 physics
The validation run is:
http://wwwint.cgd.ucar.edu/project/diagnostics/cesm-1.0/amwg/hannay/cam5_0_39_1d_u265/cam5_0_39_1d_u265-obs/
======================================================================
===============================================================
Tag name: cam5_0_39
Originator(s): mvr,mvertens
Date: Fri Jan 21 2011
One-line Summary: mods to compute coupler indices locally
Purpose of changes:
The coupler is being changed to not depend on hard-wired indices. Each component
will now compute its own indices. The calculation of the indices will reside in
a new module called cam_cpl_indices.F90 - that is in a new directory cpl_share
that is shared between cpl_esmf and cpl_mct. Use of seq_flds_indices is being
removed.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes: not checked
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
A models/atm/cam/src/cpl_share
A models/atm/cam/src/cpl_share/cam_cpl_indices.F90
- new directory and file for computing coupler indices locally
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/configure
- addition of new directory cpl_share to filepath
M models/atm/cam/src/cpl_mct/atm_comp_mct.F90
M models/atm/cam/src/chemistry/mozart/mo_srf_emissions.F90
M models/atm/cam/src/cpl_esmf/atm_comp_esmf.F90
- replaced use of seq_flds_indices with cam_cpl_indices
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: ALL PASS
edinburgh/lf95: not tested
edinburgh/pgi: not tested
lynx/pgi: ALL PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers:
======================================================================
===============================================================
Tag name: cam5_0_38
Originator(s): mvr
Date: Fri Jan 07 2011
One-line Summary: updated for esmf metadata capability
Purpose of changes:
The "register" interface in esmf is changing. This change only
impacts the esmf interfaces, nothing else. This should all be bfb
and really only impacts the esmf interface version of the components.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes: not checked
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
M models/atm/cam/src/cpl_esmf/atm_comp_mct.F90
M models/atm/cam/src/cpl_esmf/atm_comp_esmf.F90
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: ALL PASS
edinburgh/lf95: not tested
edinburgh/pgi: not tested
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers:
======================================================================
===============================================================
Tag name: cam5_0_37
Originator(s): Jim Edwards, Mark Taylor
Date: 01-03-2011
One-line Summary: Merge homme dycore development, inline history interpolation
Purpose of changes: Improve support for unstructured grids
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist: Added analysis namelist for homme dycore
with the following variables:
interpolate_analysis : logical array which indicates whether each history file should be interpolated to a lat/lon grid
interp_nlat, interp_nlon : integer variables indicating the size of the lat lon grid to interpolate to
interp_gridtype: integer indicating the type of output grid to use.
! gridtype = 1 equally spaced, including poles (FV scalars output grid)
! gridtype = 2 Gauss grid (CAM Eulerian)
! gridtype = 3 equally spaced, no poles (FV staggered velocity)
interp_type : The type of interpolation to perform, currently supported values are
! interp_type = 0 Native grid high order interpolation
! interp_type = 1 Bilinear interpolation
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Inline Grid interpolation appears to require some 6-10x more time per history file than writting on the native grid,
this feature is added as a convenience for the scientist. If high frequency output is required, or it is required that
the interpolation meet extraordinary quality standards, then it is recommended to
output on the native grid and interpolate using the post processing tool provided in homme.
Code reviewed by: Eaton, Taylor, Edwards
List all subroutines eliminated:
List all subroutines added and what they do: interp_mod.F90 provides an interface between cam_history.F90
and the interpolation functions which may be unique to each dycore.
List all existing files that have been modified, and describe the changes:
models/atm/cam/SVN_EXTERNAL_DIRECTORIES
models/atm/cam/src/dynamics/homme/external
The homme/external directory was formerly a copy of the homme source tree,
we are now pointing directly at the homme source code repository.
models/atm/cam/bld/Makefile.in
Replaced -btextpsize:64k with -btextpsize:32K for AIX builds, this is a workaround for a bug in AIX which prevents
debuggers from properly reporting source code lines.
models/atm/cam/bld/build-namelist
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
models/atm/cam/bld/namelist_files/namelist_definition.xml
Added support for the new interpolation fields outlined above
models/atm/cam/src/dynamics/sld/interp_mod.F90
models/atm/cam/src/dynamics/eul/interp_mod.F90
models/atm/cam/src/dynamics/fv/interp_mod.F90
Dummy code added to support the interpolation api. This code could also be
used to interpolate from one lat/lon grid to another.
models/atm/cam/src/control/cam_history.F90
models/atm/cam/src/utils/cam_pio_utils.F90
Added code to support history file interpolation, including support for interpolating vector pairs
which may require simultanious treatment.
models/atm/cam/src/dynamics/homme/dyn_grid.F90
models/atm/cam/src/dynamics/homme/interp_mod.F90
models/atm/cam/src/dynamics/homme/dp_coupling.F90
models/atm/cam/src/dynamics/homme/dyn_comp.F90
models/atm/cam/src/dynamics/homme/stepon.F90
Added code for inline interpolation of history files to a lat-lon grid, merged in latest homme dycore
development.
models/atm/cam/src/control/print_memusage.F90
Bug fix for intrepid (BGQ)
models/atm/cam/chemistry/bulk_aero/dust_intr.F90
Replaced netcdf calls with pio calls and added interpolation.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL
021 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...............................FAIL
036 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ......................................FAIL
040 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .....................................FAIL
045 bl373 TBL.sh fm1.9c4dh outfrq3s 9s ....................................FAIL
047 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...........................FAIL
063 bl731 TBL.sh h16c3aqdh aqua+aquaplanet_cam3 9s ........................FAIL
066 bl751 TBL.sh h30c4aqdm outfrq3s+aquaplanet_cam4 9s ....................FAIL
edinburgh/lf95:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL
009 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL
015 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL
022 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL
030 bl312 TBL.sh f10c5dm ghgrmp 9s ........................................FAIL
edinburgh/pgi or jaguar/pgi:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL
009 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL
015 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL
022 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL
030 bl312 TBL.sh f10c5dm ghgrmp 9s ........................................FAIL
043 bl317 TBL.sh f10c5cdm outfrq3s 9s .....................................FAIL
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
Failing baseline cases are due to the change in dust_intr.F90, testing shows that this was a round-off
change for the fv cases. The EUL cases however were not on the same grid as the dataset, though they
were at the same resolution, which causes the dataset to now be interpolated and thus greater
than roundoff error.
HOMME dycore baselines are failing because the default history file is now interpolated to a latlon
grid. These were tested to be bfb with interpolate_analysis set to .false.
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_0_36
Originator(s): mvr
Date: Tue Jan 05 2011
One-line Summary: updated external definitions of component tags to those
of cesm1_0_beta13
Purpose of changes:
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets:
- new default datasets for plant function types and river routing
Describe any substantial timing or memory changes: not checked
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- new default datasets for plant function types and river routing
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- removed entry for obsolete namelist var, hist_crtinic
M models/atm/cam/bld/namelist_files/use_cases/aquaplanet_cam5.xml
- bug fix to remove hard-coded file path to solar input datafile
M SVN_EXTERNAL_DIRECTORIES
M .
- new component tags for clm, cice (non-bfb), csm_share, and scripts
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
004 bl132 TBL.sh e48c4dh ghgrmp+1850_cam4 9s ..............................FAIL! rc= 7 at Tue Jan 4 13:34:59 MST 2011
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Tue Jan 4 13:35:07 MST 2011
017 bl331 TBL.sh f4c4dh co2rmp+1850_cam4 9s ...............................FAIL! rc= 7 at Tue Jan 4 13:36:36 MST 2011
021 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...............................FAIL! rc= 7 at Tue Jan 4 13:36:44 MST 2011
033 bl336 TBL.sh wm4c4h outfrq3s+waccm_1850_cam4 9s .......................FAIL! rc= 7 at Tue Jan 4 13:37:14 MST 2011
036 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ......................................FAIL! rc= 7 at Tue Jan 4 13:37:21 MST 2011
040 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .....................................FAIL! rc= 7 at Tue Jan 4 13:37:42 MST 2011
045 bl373 TBL.sh fm1.9c4dh outfrq3s 9s ....................................FAIL! rc= 7 at Tue Jan 4 13:38:08 MST 2011
047 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...........................FAIL! rc= 7 at Tue Jan 4 13:38:23 MST 2011
050 bl374 TBL.sh fs1.9c4dh outfrq3s+1850-2005_cam4_super_fast_llnl 9s .....FAIL! rc= 7 at Tue Jan 4 13:38:39 MST 2011
054 bl387 TBL.sh f1.9c4dm outfrq3s+1850-2005_cam4 9s ......................FAIL! rc= 7 at Tue Jan 4 13:38:47 MST 2011
057 bl388 TBL.sh f1.9c4m outfrq24h+1850_cam4 2d ...........................FAIL! rc= 7 at Tue Jan 4 13:38:53 MST 2011
060 bl379 TBL.sh wm1.9c4h outfrq24h+waccm_1850_cam4 2d ....................FAIL! rc= 7 at Tue Jan 4 13:40:34 MST 2011
072 bl992 TBL_ccsm.sh f19_g16 E 2d ........................................FAIL! rc= 7 at Tue Jan 4 13:40:40 MST 2011
- baseline failures were expected for any configuration using cice, where changes were not bfb
edinburgh/lf95:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Tue Jan 4 13:31:35 MST 2011
009 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Tue Jan 4 13:31:39 MST 2011
015 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Tue Jan 4 13:32:41 MST 2011
022 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Tue Jan 4 13:32:52 MST 2011
030 bl312 TBL.sh f10c5dm ghgrmp 9s ........................................FAIL! rc= 7 at Tue Jan 4 13:32:58 MST 2011
035 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7 at Tue Jan 4 13:33:03 MST 2011
043 bl318 TBL.sh f10c4cdm outfrq3s 9s .....................................FAIL! rc= 7 at Tue Jan 4 13:34:52 MST 2011
- baseline failures were expected for any configuration using cice, where changes were not bfb
edinburgh/pgi:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Wed Jan 5 09:09:58 MST 2011
009 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Wed Jan 5 09:10:00 MST 2011
015 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Wed Jan 5 09:10:03 MST 2011
022 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Wed Jan 5 09:10:06 MST 2011
030 bl312 TBL.sh f10c5dm ghgrmp 9s ........................................FAIL! rc= 7 at Wed Jan 5 09:10:09 MST 2011
035 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7 at Wed Jan 5 09:10:12 MST 2011
043 bl317 TBL.sh f10c5cdm outfrq3s 9s .....................................FAIL! rc= 7 at Wed Jan 5 09:10:19 MST 2011
046 bl319 TBL.sh fn10c5dm outfrq3s 9s .....................................FAIL! rc= 7 at Wed Jan 5 09:10:20 MST 2011
- baseline failures were expected for any configuration using cice, where changes were not bfb
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers:
Answer changes were due soley to the new cice tag. These changes were described as greater than round-off, but
same climate - please see the cice documentation for more information.
======================================================================
===============================================================
Tag name: cam5_0_35
Originator(s): Jerry Olson
Date: Dec 29, 2010
One-line Summary: fixed "cam3" and "cam3/aquaplanet" to be more
consistent with CAM3.1 release. Adjusted cam4
and cam5 aquaplanet namelist parameters.
Purpose of changes: fix cam3 and aquaplanet runs. Added cldfrc tuning
parameters to namelist. 1 minor bugfix
List of changes to have the current "cam3" replicate CAM3.1:
1) models/atm/cam/src/physics/cam/stratiform.F90: Remove extra cldfrc call after pcond call
2) models/atm/cam/src/physics/cam/zm_conv_intr.F90: Shut off convective momentum transport
3) models/atm/cam/src/physics/cam/zm_conv.F90: Call buoyan instead of buoyan_dilute
4) models/atm/cam/bld/namelist_files/use_cases/cam3.xml: cldfrc_freeze_dry set to .false.
5) models/atm/cam/bld/namelist_files/use_cases/cam3.xml:
a) tuning parameters are set back to their CAM3.1 levels
b) orb_iyear, solar_const, GHG values all set back to CAM3.1 values
Additional changes to have the current "cam3" aquaplanet replicate CAM3.1 aquaplanet:
1) models/atm/cam/bld/namelist_files/use_cases/aquaplanet_cam3.xml and
models/atm/cam/src/control/physconst.F90:
modify freezing point of water from 273.15 --> 273.16
2) copy models/csm_share/shr/shr_flux_mod.F90 to
models/atm/cam/src/utils/cam_aqua/shr_flux_mod.F90 and modify:
a) originally uses T rather than virtual T to compute stability functions.
Changed back to virtual T for aqua_planet.
b) "umin" set back to 1. from 0.5
3) models/atm/cam/src/physics/cam/radlw.F90: initialization of, "cldp", using
"ntoplw" instead of "ntopcld"
4) models/atm/cam/src/chemistry/utils/prescribed_ozone.F90: for unit conversion of APE ozone,
change universal constants:
from:
- real(R8), parameter :: SHR_CONST_MWDAIR = 28.96623324623746_R8
- real(r8), parameter :: molmass = 47.9981995_r8
to:
- real(r8), parameter :: amass = 28.9644_r8
- real(r8), parameter :: molmass = 48._r8
5) models/atm/cam/bld/namelist_files/use_cases/aquaplanet_cam3.xml:
a) solar_const, GHG values set back to CAM3.1 aquaplanet levels:
b) orbit set to circular/equinox settings.
c) dtime set to 5 minutes physics timestep (dtime = 150 for Eulerian dycore)
NOTE --> All tuning parameters for cam3 aquaplanet are set to standard CAM3.1 T85
settings no matter the dycore or resolution. DTIME set to 5 minutes
physics timestep. User must change in namelist if other settings are
desired.
Bugs fixed (include bugzilla ID): State variables now output time-averaged
by default when "budget_history" is true.
Describe any changes made to build system: csm_share/shr and csm_share/dshr
moved to bottom of Filepath - has
no effect on answers
Describe any changes made to the namelist:
When running full model, "cam3" physics, must use "-use_case cam3" in build-namelist call
(numerous tuning parameter changes to be consistent with CAM3.1 release)
When running aquaplanet, "cam3" physics, must use "-use_case aquaplanet_cam3" in build-namelist call
(replaces aquaplanet_cam3_cam4.xml)
When running aquaplanet, "cam4" physics, must use "-use_case aquaplanet_cam4" in build-namelist call
(replaces aquaplanet_cam3_cam4.xml)
When running aquaplanet, "cam5" physics, must use "-use_case aquaplanet_cam5" in build-namelist call
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: not checked
Code reviewed by: self, eaton
List all subroutines eliminated:
models/atm/cam/bld/namelist_files/use_cases/aquaplanet_cam3_cam4.xml
List all subroutines added and what they do:
models/atm/cam/bld/namelist_files/use_cases/aquaplanet_cam3.xml
namelist defaults for cam3 aquaplanet
models/atm/cam/bld/namelist_files/use_cases/aquaplanet_cam4.xml
namelist defaults for cam4 aquaplanet
models/atm/cam/bld/namelist_files/use_cases/cam3.xml
namelist defaults for cam3 full model runs
models/atm/cam/src/utils/cam_aqua/shr_flux_mod.F90
copy of "models/csm_share/shr/shr_flux_mod.F90" for use
when running aqua_planet. Modified to replicate cam3 aquaplanet
results.
models/atm/cam/test/system/config_files/f10c3aqdm
regression test added to test cam3 aquaplanet
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/build-namelist
- added cldfrc tuning parameters
models/atm/cam/bld/configure
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- moved some namelist defaults to "cam3.xml", added cldfrc tuning params
models/atm/cam/bld/namelist_files/namelist_definition.xml
- cldfrc tuning params
models/atm/cam/bld/namelist_files/use_cases/aquaplanet_cam5.xml
- set some new aqua_planet defaults
models/atm/cam/src/chemistry/utils/prescribed_ozone.F90
- changed physical constants for cam3 aquaplanet
models/atm/cam/src/control/physconst.F90
- set tmelt to be initialized at runtime
models/atm/cam/src/physics/cam/cam_diagnostics.F90
- made state variables time averaged rather than instantaneous
on output when "history_budget" is .true.
models/atm/cam/src/physics/cam/cldwat2m_macro.F90
models/atm/cam/src/physics/cam/cldwat2m_micro.F90
- set tmax_fsnow, tmin_fsnow to be initialized at runtime
models/atm/cam/src/physics/cam/cldwat.F90
- set tmax_fsnow, tmin_fsnow, tmax_fice, tmin_fice to be initialized at runtime
models/atm/cam/src/physics/cam/cloud_fraction.F90
- made cldfrc tuning params namelist parameters
models/atm/cam/src/physics/cam/microp_driver.F90
- made state variables time averaged rather than instantaneous
on output when "history_budget" is .true.
models/atm/cam/src/physics/cam/physpkg.F90
- call inimc from here rather than from stratiform
models/atm/cam/src/physics/cam/radlw.F90
- code change to fix cam3 physics
models/atm/cam/src/physics/cam/stratiform.F90
- made state variables time averaged rather than instantaneous
on output when "history_budget" is .true.
- Remove extra cldfrc call after pcond call (cam3 physics change)
models/atm/cam/src/physics/cam/zm_conv.F90
- Call buoyan instead of buoyan_dilute (cam3 physics change)
models/atm/cam/src/physics/cam/zm_conv_intr.F90
- Shut off convective momentum transport (cam3 physics change)
models/atm/cam/src/physics/rrtmg/ext/rrtmg_lw/rrtmg_lw_init.f90
- set physics constants from "physconst" rather than "shr_const_mod"
models/atm/cam/src/physics/rrtmg/ext/rrtmg_sw/rrtmg_sw_init.f90
- set physics constants from "physconst" rather than "shr_const_mod"
models/atm/cam/test/system/input_tests_master
- use aquaplanet .xml files for all aquaplanet regression tests
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
All pass except the following baseline tests which change due to changes in aquaplanet
006 bl133 TBL.sh e48c4paqdm aqpgro+aquaplanet_cam4 3s .....................FAIL! rc= 5 at Tue Dec 28 14:14:56 MST 2010
013 bl330 TBL.sh f4c5paqdh aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Tue Dec 28 14:42:34 MST 2010
023 bl333 TBL.sh f4c4paqdh aqpgro+aquaplanet_cam4 3s ......................FAIL! rc= 5 at Tue Dec 28 15:06:39 MST 2010
063 bl731 TBL.sh h16c3aqdh aqua+aquaplanet_cam3 9s ........................FAIL! rc= 5 at Tue Dec 28 18:19:21 MST 2010
066 bl751 TBL.sh h30c4aqdm outfrq3s+aquaplanet_cam4 9s ....................FAIL! rc= 5 at Tue Dec 28 18:31:44 MST 2010
edinburgh/lf95:
All pass except the following baseline tests which change due to changes in aquaplanet and/or cam3
011 bl113 TBL.sh e8c5paqdm aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Wed Dec 29 11:36:29 MST 2010...
032 bl313 TBL.sh f10c3aqdm outfrq3s+aquaplanet_cam3 3s ....................FAIL! rc= 5 at Wed Dec 29 12:48:00 MST 2010...
038 bl315 TBL.sh f10c3dm outfrq3s+cam3 9s .................................FAIL! rc= 5 at Wed Dec 29 13:22:18 MST 2010...
040 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 7 at Wed Dec 29 13:33:49 MST 2010...
edinburgh/pgi or jaguar/pgi:
All pass except the following baseline tests which change due to changes in aquaplanet and/or cam3
011 bl113 TBL.sh e8c5paqdm aqpgro+aquaplanet_cam5 3s ......................FAIL! rc= 7 at Wed Dec 29 11:36:01 MST 2010...
032 bl313 TBL.sh f10c3aqdm outfrq3s+aquaplanet_cam3 3s ....................FAIL! rc= 5 at Wed Dec 29 11:52:15 MST 2010...
038 bl315 TBL.sh f10c3dm outfrq3s+cam3 9s .................................FAIL! rc= 5 at Wed Dec 29 12:09:40 MST 2010...
040 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 7 at Wed Dec 29 12:16:19 MST 2010...
049 bl734 TBL.sh hn16c5aqdm outfrq3s+aquaplanet_cam5 9s ...................FAIL! rc= 7 at Wed Dec 29 13:09:07 MST 2010...
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB except > RO changes in aquaplanet and any cam3 runs
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_0_34
Originator(s): Jim Edwards
Date: Dec 21, 2010
One-line Summary: fix mpi issues
Purpose of changes: bug fixes
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: self
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
mo_lightning.F90 : needed to add optional commm parameter to repro_sum calls
metdata.F90 : needed to move pio_closefile call out of mastertask only clause.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: all pass
edinburgh/lf95: all pass
edinburgh/pgi or jaguar/pgi: all pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate): None
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_0_33
Originator(s): mvr, jenkay
Date: Thu Dec 16 2010
One-line Summary: extended cosp code to work with cam5 physics; update
to latest cosp vendor tag
Purpose of changes:
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: not checked
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do:
A models/atm/cam/test/system/config_files/f10c5cdm
- new configuration options test file for testing cosp with cam5 physics
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/tests_pretag_edinburgh_pgi
- added test for cosp/cam5 to default suite of tests on edinburgh/pgi
- removed test for cosp/cam4 from default suite (still remains in edinburgh/lahey)
M models/atm/cam/test/system/input_tests_master
- added definition of news tests for cosp/cam5
M models/atm/cam/bld/configure
- no longer dies for cosp/cam5 configuration
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- entry for cosp_nradsteps; correction to cosp_histfile_num's type
M models/atm/cam/src/physics/cam/initindx.F90
- added call to new method, conv_water_register
M models/atm/cam/src/physics/cam/radiation.F90
M models/atm/cam/src/physics/cam/stratiform.F90
M models/atm/cam/src/physics/cam/cosp_share.F90
M models/atm/cam/src/physics/cam/convect_shallow.F90
M models/atm/cam/src/physics/cam/cospsimulator_intr.F90
M models/atm/cam/src/physics/cam/microp_driver.F90
M models/atm/cam/src/physics/cam/cldwat2m_micro.F90
M models/atm/cam/src/physics/cam/conv_water.F90
M models/atm/cam/src/physics/cam/uwshcu.F90
M models/atm/cam/src/physics/rrtmg/radiation.F90
- local mods needed to work with new cosp vendor code to enable cosp
to work with cam5 physics
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
M models/atm/cam
- update to latest cosp vendor tag
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: ALL PASS
edinburgh/lf95:
043 bl318 TBL.sh f10c4cdm outfrq3s 9s .....................................FAIL! rc= 7 at Thu Dec 16 11:10:47 MST 2010
- expected failure for cosp using cam4 due to changes in external cosp code
edinburgh/pgi:
043 bl317 TBL.sh f10c5cdm outfrq3s 9s .....................................FAIL! rc= 7 at Wed Dec 15 13:57:42 MST 2010
- new test for cosp using cam5 failed because baseline code did not recognize test configuration
CAM tag used for the baseline comparison tests if different than previous
tag:
**note: additional tests were performed for cosp using both cam4 and cam5 where results
were compared to the cosp branch tag, cosp08_cam5_0_18...results were b4b
Summarize any changes to answers: BFB
======================================================================
===============================================================
Tag name: cam5_0_32
Originator(s): Edwards, Vitt
Date: 12-10-10
One-line Summary: Generalize chemistry data input for unstructured grids
Purpose of changes: Extend unstructured grid functionality to chemistry/ cam5
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
Changed default drydep_method to xactive_lnd except when modal chemistry is activated
Added variable drydep_srf_file to chem_inparm namelist, this allows drydep surface values
to be interpolated to the cam grid offline by the new tool mkatmsrffile
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Edwards, Vitt, Kinnison, Lamarque, Marsh, Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
Added atm/cam/tools/mkatmsrffile to create fraction_landuse and soilw fields off line
List all existing files that have been modified, and describe the changes:
models/atm/cam/tools/interpic_new/README
extended instructions for creating homme dycore input data files
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
added defualts for drydep_method and drydep_srf_file
models/atm/cam/bld/namelist_files/namelist_definition.xml
added drydep_srf_file
models/atm/cam/src/chemistry/utils/mo_flbc.F90
models/atm/cam/src/chemistry/mozart/mo_lightning.F90
models/atm/cam/src/chemistry/mozart/mo_chemini.F90
models/atm/cam/src/chemistry/mozart/mo_airplane.F90
models/atm/cam/src/chemistry/mozart/mo_fstrat.F90
models/atm/cam/src/chemistry/mozart/mo_photo.F90
models/atm/cam/src/chemistry/mozart/mo_strato_sad.F90
changed IO and interpolation methods
models/atm/cam/src/chemistry/utils/tracer_data.F90
removed unused use statement
models/atm/cam/src/chemistry/mozart/mo_setext.F90
moved airplane specific code to mo_airplane.F90
models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
updated interfaces to chemistry code with new IO and
interpolation methods
models/atm/cam/src/chemistry/mozart/chemistry.F90
added support for new namelist variable drydep_srf_file
removed hard coded string lengths and replaced with shr_kind_cl
models/atm/cam/src/chemistry/mozart/mo_drydep.F90
added support for reading fraction_landuse and soilw as generated offline
models/atm/cam/src/chemistry/mozart/mz_aerosols_intr.F90
fraction_landuse changed from lat/lon grid to model grid
models/atm/cam/src/chemistry/mozart/mo_waccm_hrates.F90
changed interface to set_ub_col
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
All pass except the following baseline tests which change due to changes in interpolation
033 bl336 TBL.sh wm4c4h outfrq3s+waccm_1850_cam4 9s
045 bl373 TBL.sh fm1.9c4dh outfrq3s 9s
050 bl374 TBL.sh fs1.9c4dh outfrq3s+1850-2005_cam4_super_fast_llnl 9s
060 bl379 TBL.sh wm1.9c4h outfrq24h+waccm_1850_cam4 2d
edinburgh/lf95: All pass
edinburgh/pgi or jaguar/pgi: All pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
cam4 with chemistry
what platforms/compilers: All
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
Larger than roundoff changes in interpolation of input fields most pronounced near the
poles due to changes in interpolation method.
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
Tag name: cam5_0_31
Originator(s): mvr, eaton, sungsu, gettleman
Date: Mon Dec 06 2010
One-line Summary: corrected the default settings for compiler type on linux
systems; fix to unwanted side-effect of using do_circulation_diags; mod to
enable running cam5 physics without chemistry; default settings added for
divergence damping and velocity diffusion as well as the number of vertical
re-mapping timesteps per physics timestep; bug fix in shallow convection
scheme; added a few new tests, changed the names of a few others
Purpose of changes:
The previous tag broke cam testing on the crays, where the test scripts were
relying on default settings for compiler type...changes to Makefile.in were
applied to address this problem
We discovered that the CAM use case files that set do_circulation_diags=.true.
were having the undesirable side effect of writing the 3D field TH into the
daily avg h1 file.
Modifications were added to get cam5 with -chem none option to produce a
climate 'similar' to standard cam5. This also enabled running the homme
dycore with cam5 physics - tests were added to exercise these. (gettleman)
Default setting added for divergence damping and velocity diffusion at 1deg
for cam5 (was 2, now 4)
Default settings added for the number of vertical re-mapping timesteps per
physics timestep: 2 for 0.5deg and 0.25deg and 1 otherwise
The shallow convection scheme needed a fix to prevent division by zero in
computing cumulus top height (sungsu).
edinburgh testing will now test separate suites for lahey and pgi compilers
Scam test using cam5 will now have chemistry enabled.
Default homme testing now done at higher resolution (h30np4) on bluefire
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: none
Describe any changes made to the namelist:
- add_default for nspltvrm (number of vertical re-mapping timesteps per
physics timestep)
- corrected default settings for compiler type on linux systems
List any changes to the defaults for the boundary datasets:
- new inic file for homme using cam5 physics
Describe any substantial timing or memory changes: not checked
Code reviewed by: eaton
List all subroutines eliminated: none
D models/atm/cam/test/system/config_files/f10c5pdm
D models/atm/cam/test/system/config_files/e48c4pdm
D models/atm/cam/test/system/config_files/e8c5pdm
D models/atm/cam/test/system/config_files/h16x4c3dh
D models/atm/cam/test/system/config_files/e64c5pm
D models/atm/cam/test/system/config_files/s8c5pdm
D models/atm/cam/test/system/config_files/s32c5pdh
D models/atm/cam/test/system/config_files/f4c5pdh
D models/atm/cam/test/system/config_files/f4c4pdh
- these configuration options files (used by test scripts) were renamed
D models/atm/cam/test/system/config_files/h16c4aqdm
D models/atm/cam/test/system/tests_pretag_edinburgh
- removed these obsolete test files
List all subroutines added and what they do:
A + models/atm/cam/test/system/config_files/e64c5paqm
A + models/atm/cam/test/system/config_files/s32c5paqdh
A + models/atm/cam/test/system/config_files/f4c4paqdh
A + models/atm/cam/test/system/config_files/f10c5paqdm
A + models/atm/cam/test/system/config_files/f4c5paqdh
A + models/atm/cam/test/system/config_files/e8c5paqdm
A + models/atm/cam/test/system/config_files/s8c5paqdm
A + models/atm/cam/test/system/config_files/h16c3aqdh
A + models/atm/cam/test/system/config_files/e48c4paqdm
- these are newly named configuration options files (used by test scripts)
A models/atm/cam/test/system/config_files/h30c4aqdm
A models/atm/cam/test/system/config_files/hn16c5aqdm
A models/atm/cam/test/system/config_files/fn10c5dm
- these are new configuration options files needed for new tests
A models/atm/cam/test/system/tests_pretag_edinburgh_lahey
A models/atm/cam/test/system/tests_pretag_edinburgh_pgi
- will now have separate default test suites depending on the edinburgh compiler used
- new tests added to pgi suite
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/tests_pretag_bluefire
- default homme testing now done at higher resolution on bluefire
M models/atm/cam/test/system/config_files/scm64c4bfbiop
M models/atm/cam/test/system/config_files/e64c4bfbiop
- removed redundant setting of '-chem none' from cam4 scam test
- speeding up global 3d portion of test to run with mpi-only rather than serial
M models/atm/cam/test/system/config_files/scmc5armiop
M models/atm/cam/test/system/config_files/scm64c5bfbiop
M models/atm/cam/test/system/config_files/e64c5bfbiop
- removed setting of '-chem none' to test default chemistry as part of cam5 scam test
- speeding up global 3d portion of test to run with mpi-only rather than serial
M models/atm/cam/test/system/test_driver.sh
- will now have separate default test suites depending on the edinburgh compiler used
M models/atm/cam/test/system/input_tests_master
- changes to test definitions due to filename changes with configuration options
- several new tests added, cleanup
M models/atm/cam/bld/Makefile.in
- on linux systems, made the default fc_type pgi and set the default C compiler to
cc when the fortran compiler is ftn
- added the same optimization to the CFLAGS for pgi that is set in the
CESM macros (for consistency)
M models/atm/cam/bld/config_files/definition.xml
- The parameters cflags, fflags, and ldflags have all had the list="1" attribute added,
to allow adding flag values that contain commas.
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- new inic file for homme using cam5 physics
- new default settings for the number of vertical re-mapping timesteps per physics timestep
- new default setting for divergence damping and velocity diffusion for 1deg cam5
M models/atm/cam/bld/namelist_files/use_cases/waccm_2000_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1850-2005_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp45.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1955-2005_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1974-2005_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_refb1.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1850_cam4.xml
- Add the fields from the removed add_default calls (VTH2d, WTH2d, UV2d,
UW2d, TH, and MSKtem) explicitly to the fincl2 specification in *all*
WACCM use cases
M models/atm/cam/bld/build-namelist
- add_default added for nspltvrm
M models/atm/cam/src/physics/cam/microp_aero.F90
- modifications to get cam5 with -chem none to produce a climate 'similar' to
standard cam5 (gettleman)
M models/atm/cam/src/physics/cam/uwshcu.F90
- fix to prevent division by zero in computing cumulus top height in shallow convection
scheme (sungsu)
M models/atm/cam/src/dynamics/fv/ctem.F90
- remove the add_default calls for VTH2d, WTH2d, UV2d, UW2d, TH, and MSKtem
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
006 bl133 TBL.sh e48c4paqdm aqpgro 3s .....................................FAIL! rc= 7 at Mon Dec 6 09:46:12 MST 2010
013 bl330 TBL.sh f4c5paqdh aqpgro 3s ......................................FAIL! rc= 7 at Mon Dec 6 09:46:13 MST 2010
023 bl333 TBL.sh f4c4paqdh aqpgro 3s ......................................FAIL! rc= 7 at Mon Dec 6 09:46:51 MST 2010
063 bl731 TBL.sh h16c3aqdh aqua 9s ........................................FAIL! rc= 7 at Mon Dec 6 11:19:38 MST 2010
066 bl751 TBL.sh h30c4aqdm outfrq3s 9s ....................................FAIL! rc= 7 at Mon Dec 6 11:31:15 MST 2010
- bl751 is a new test, the other failures were due to name changes of the configuration options files
edinburgh/lf95:
011 bl113 TBL.sh e8c5paqdm aqpgro 3s ......................................FAIL! rc= 7 at Mon Dec 6 09:42:07 MST 2010
032 bl313 TBL.sh f10c5paqdm aqpgro 3s .....................................FAIL! rc= 7 at Mon Dec 6 10:37:41 MST 2010
040 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 7 at Mon Dec 6 10:59:10 MST 2010
- failures were due to name changes of the configuration options files
edinburgh/pgi:
011 bl113 TBL.sh e8c5paqdm aqpgro 3s ......................................FAIL! rc= 7 at Mon Dec 6 09:30:14 MST 2010
032 bl313 TBL.sh f10c5paqdm aqpgro 3s .....................................FAIL! rc= 7 at Mon Dec 6 09:56:54 MST 2010
040 bl316 TBL.sh f10c5paqdm aqpgro+aquaplanet_cam5 3s .....................FAIL! rc= 7 at Mon Dec 6 10:05:49 MST 2010
046 bl319 TBL.sh fn10c5dm outfrq3s 9s .....................................FAIL! rc= 7 at Mon Dec 6 10:14:43 MST 2010
049 bl734 TBL.sh hn16c5aqdm outfrq3s 9s ...................................FAIL! rc= 7 at Mon Dec 6 10:35:24 MST 2010
- bl734 and bl319 are new tests, the other failures were due to name changes of the configuration options files
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
======================================================================
======================================================================
Tag name: cam5_0_30
Originator(s): Diane Feddema
Date: Dec 1 2010
One-line Summary: Add new command line option to CAM's configure, fc_type. Also code cleanup.
Purpose of changes: CAM's configure and Makefile now allow the user to specify which compiler they want when
they specify the wrapper script ftn as their compiler. fc_type is the underlying compiler that
will be invoked, valid options are pgi, intel, pathscale, lahey.
Bugs fixed (include bugzilla ID): None
Describe any changes made to build system: CAM's configure has a new commandline option, fc_type.
Describe any changes made to the namelist:None
List any changes to the defaults for the boundary datasets:None
Describe any substantial timing or memory changes: None
Code reviewed by:Brian Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/config_files/definition.xml
- added fc_type
M models/atm/cam/bld/configure
- added fc_type command line option
M models/atm/cam/bld/Makefile.in
- added support for fc_type command line option
M models/atm/cam/src/physics/cam/convect_shallow.F90
- removed duplicate declarations
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: None
edinburgh/lf95: None
edinburgh/pgi or jaguar/pgi: None
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
All tests are BFB
If bitwise differences were observed, how did you show they were no worse
than roundoff?
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_0_29
Originator(s): Andrew Gettelman and Cheryl Craig
Date: Nov 23 2010
One-line Summary: Separate microp/macrop physics
Purpose of changes: Separate microphysics and macrophysics drivers in CAM5 code
Bugs fixed (include bugzilla ID): Fix a bug in call to conv_water_4rad that would
lead to erroneous results with conv_water_in_rad=2 option if invoked.
Describe any changes made to build system:None
Describe any changes made to the namelist:None
List any changes to the defaults for the boundary datasets:None
Describe any substantial timing or memory changes: None
Code reviewed by:Brian Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
A models/atm/cam/src/physics/cam/macrop_driver.F90
- File which contains the macrop physics processing which
had resided in stratiform
A models/atm/cam/src/physics/cam/microp_driver.F90
- File which contains the microp physics processing which
had resided in stratiform
A models/atm/cam/src/physics/cam/conv_water.F90
- File which contains the convective water processing which
had resided in stratiform
List all existing files that have been modified, and describe the changes:
M models/atm/cam/src/physics/cam/microp_aero.F90
- changed to support separation of microphysics and macrophysics and
concentrate aerosol information there
M models/atm/cam/src/physics/cam/param_cldoptics.F90
- Changed file to support separation of the macro and micro physics code
- Fixed the bug found when conv_water_in_rad = 2 in the call to conv_water_4rad
M models/atm/cam/src/physics/cam/initindx.F90
M models/atm/cam/src/physics/cam/physpkg.F90
M models/atm/cam/src/physics/cam/stratiform.F90
M models/atm/cam/src/physics/cam/tphysbc.F90
M models/atm/cam/src/dynamics/sld/inidat.F90
M models/atm/cam/src/dynamics/eul/inidat.F90
M models/atm/cam/src/dynamics/homme/inidat.F90
M models/atm/cam/src/dynamics/fv/inidat.F90
- Changed files to support separation of the macro and micro physics code
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: None
edinburgh/lf95: None
edinburgh/pgi or jaguar/pgi: None
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
All tests are BFB
If bitwise differences were observed, how did you show they were no worse
than roundoff?
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_0_28
Originator(s): mvr,jenkay,jedwards,hannay
Date: Mon Nov 22 2010
One-line Summary: pio namelist changes; mods to handle variable orbital settings;
time_manager mods for calendar type; diagnostics added; added support to test
scripts for additional machines; updated externals
Purpose of changes:
- pio now controlled by settings in pio_inparm of the drv namelist
- now resetting orbital parameters at run phase
- time_manager to use share constants for calendar type, new method for cf standards
- new diagnostics for inversion parameters and visible cloud optical depth (jenkay),
additional history_aerosol diagnostics (hannay)
- support was added to test scripts for jaguarpf and lynx_pgi
- external definitions updated for drv,clm,docn,cice,share,mct,scripts,pio to
match those intended for cesm1_0_beta11
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets:
- fix to settings for initial conditions files for cam5 at half and one degree fv
- updated to use the gx1v6 version of both the 1/4 deg focndomain and fatmlndfrc files
Describe any substantial timing or memory changes: not checked
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do:
A models/atm/cam/bld/run-lynx.csh
- template run script for ncar cray, lynx
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/tests_pretag_bluefire
- reinstated tests er731 and bl731 to default suite on bluefire
M models/atm/cam/test/system/test_driver.sh
M models/atm/cam/test/system/CAM_runcmnd.sh
- added support to test scripts for jaguarpf, lynx_pgi
M models/atm/cam/test/system/nl_files/fvvp_lb2
M models/atm/cam/test/system/nl_files/fvvp_lb0
- modified test definition to utilize more of the requested resources
M models/atm/cam/bld/Makefile.in
- added -DNO_MPI2 to all serial builds
M models/atm/cam/bld/build-namelist
- namelist variable changed from orb_iyear_ad to orb_iyear
M models/atm/cam/bld/namelist_files/use_cases/aquaplanet_cam5.xml
M models/atm/cam/bld/namelist_files/use_cases/aquaplanet_cam3_cam4.xml
- now specifying orb_mode=fixed_parameters in use_cases where any orbital info is specified
M models/atm/cam/src/chemistry/utils/mo_solar_parms.F90
- replaced PIO_NOCLOBBER with PIO_NOWRITE to enable running on lynx, which uses NETCDF 4
MM models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
- additional diagnostics output if history_aerosol
M models/atm/cam/bld/configure
- removed references to USE_ESMF_LIB
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- fix to settings for initial conditions files for cam5 at half and one degree fv
- namelist variable changed from orb_iyear_ad to orb_iyear
- updated to use the gx1v6 version of both the 1/4 deg focndomain and fatmlndfrc files
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- mods to reflect changes in the pio namelist
- namelist variable changed from orb_iyear_ad to orb_iyear
- new entry for orb_mode
M models/atm/cam/src/utils/cam_pio_utils.F90
M models/atm/cam/src/chemistry/mozart/mo_srf_emissions.F90
- mods needed in connection with pio changes
M models/atm/cam/src/control/runtime_opts.F90
- removed unneeded broadcasts of orbital parameters
M models/atm/cam/src/control/cam_control_mod.F90
- cleanup of parameter iyear_ad
M models/atm/cam/src/control/cam_history.F90
- replaced use of get_calendar with get_calendar_cf to retrieve cf standard calendar name
- mods needed in connection with pio changes
M models/atm/cam/src/utils/time_manager.F90
- now using share values for noleap and gregorian calendar names
- removed method get_calendar
- new method, timemgr_is_caltype, returns true if incoming calendar type string
matches actual calendar type in use
- new method, timemgr_get_calendar_cf, returns cf standard for calendar type
- removed references to USE_ESMF_LIB
M models/atm/cam/src/utils/cam_csim4/ice_time_manager.F90
M models/atm/cam/src/utils/cam_dom/ocn_time_manager.F90
- now using share values for noleap and gregorian calendar names
- removed method get_calendar
- new method, timemgr_is_caltype, returns true if incoming calendar type string
matches actual calendar type in use
M models/atm/cam/src/cpl_mct/atm_comp_mct.F90
- now resetting orbital parameters with each run phase, rather than just at init
M models/atm/cam/src/physics/cam/radiation.F90
M models/atm/cam/src/physics/cam/radsw.F90
- new visible cloud optical depth diagnostics for cam4 (from jenkay)
M models/atm/cam/src/physics/cam/cam_diagnostics.F90
- new inversion diagnostics (from jenkay)
M models/atm/cam/src/physics/rrtmg/radiation.F90
- new visible cloud optical depth diagnostics for cam5 (from jenkay)
M models/atm/cam/src/physics/cam/cospsimulator_intr.F90
M models/atm/cam/src/chemistry/utils/tracer_data.F90
- removed references to get_calendar method in time_manager.F90
M models/atm/cam/src/chemistry/mozart/spedata.F90
M models/atm/cam/src/dynamics/fv/metdata.F90
- replaced use of get_calendar with timemgr_is_caltype and shr constants
M models/atm/cam/src/cpl_esmf/atm_comp_esmf.F90
- added code to load orbital parameters
M models/atm/cam/src/control/cam_comp.F90
- cleanup of comments referring to orbital parameters
M SVN_EXTERNAL_DIRECTORIES
M .
- updated external definitions of component tags to those meant for cesm1_0_beta11
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
063 bl731 TBL.sh h16x4c3dh aqua 9s ........................................FAIL! rc= 7 at Mon Nov 22 00:57:53 MST 2010
- failure due to test being reinstated and therefore not recognized by baseline scripts
**Note that sm731 had been failing up until this tag. It is unclear what exactly
was responsible for it now passing. Through isolating each of the modifications
going in with this tag, it appears to have something to do with the changes made
in time_manager.F90, where one method was removed and two were added.
edinburgh/lf95: All PASS
edinburgh/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_0_27
Originator(s): eaton
Date: Thu Nov 18 14:55:18 MST 2010
One-line Summary: Remove dependence of modal optics calc on prognostic MAM.
Purpose of changes:
. The first step in providing a prescribed modal aerosol capability is to
make the current prognostic MAM interact with the radiative heating
calculation via the rad_constituents module. This commit is just part of
the first step; the dependence of the modal_aer_opt module on the
prognostic MAM code has been removed. A side effect of this refactoring
in modal_aer_opt was an extensive code clean up including removal of the
inactive mie code (which can be reintroduced later if desired).
The major structural change in modal_aer_opt is to reorder the loops so
that the loop over modes is the outermost loop. This is in preparation
for doing diagnostic forcing calculations in which some modes are
excluded from having a radiative effect. The ability to do these
diagnostic calculations is not yet in place.
The rad_constituents interface is used to access the physical properties
and mass mixing ratios of the individual species in each mode. The
rad_constituents interface requires the names of the species (as stored
in the state or physics buffer) to access properties. The names of the
species in each mode (which is hardcoded in the prognostic MAM code) were
added to the modal optics data file to make them accessible for either
prognostic or prescribed modal optics calculations.
The existing rad_constituent interfaces used list indexes to access the
individual aerosols. The interface was intended to be used in a context
where the aerosols were all considered to be bulk aerosols whose optical
properties would be externally mixed. But in the modal context where the
species are internally mixed we would like to be able to access
individual species by name. So generic methods have been added to the
rad_constituents interfaces that allow requesting the physical properties
and the mass mixing ratios either by index or by name.
The namelist setting for rad_climate was updated to include all the
3-mode species. This is required to use the rad_constituents interface
to retrieve properties and mmr for all the species contributing to the
modes.
The species refractive indices were being read from the physprops files,
but the species densities were hardcoded and the densities in the
physprops files did not match the hardcoded values. So the physprops
files were updated to contain the hardcoded values and the prognostic MAM
code was changed to set the densities from the files.
The phys_prop module had been hacked to get the refractive index of water
from the physprops files of the first aerosol, and the refractive index
of water data was duplicated in all the aerosol species files. Remove
this hack and put water refractive index in it's own file. Read the data
in the modal_aer_opt module since that's where it's used. Give
modal_aer_opt its own namelist for reading the name of both the water
refindex file and the modal optics file.
Refactor phys_prop and rad_constituents interfaces to pass complex form
of refractive index.
Refactor phys_prop to read refractive index data if it's present. Don't
depend on MODAL_AERO cpp token which is only defined for prognostic MAM.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist:
. rad_climate updated for trop_mam3 to include all species that contribute
to the modes.
List any changes to the defaults for the boundary datasets:
. The modal optics file has been updated to include the metadata that
describes the modes, i.e., the number of species in each mode, the names
of the species in each mode, and the types of the species in each mode.
The original file, modal_optics_3mode.nc, was updated and renamed
modal_optics_3mode_c100507.nc.
. The following files have the "density" variables updated to match the
hardcoded values in the trop_mam3 code.
bcpho_rrtmg_c090310.nc --> bcpho_rrtmg_c100508.nc
ocpho_rrtmg_c100528.nc --> ocpho_rrtmg_c101112.nc
ocphi_rrtmg_c080918.nc --> ocphi_rrtmg_c100508.nc
ssam_rrtmg_c080918.nc --> ssam_rrtmg_c100508.nc
. A new file was constructed that contains the refractive index data for
water. The file was created by extracting the data from
ocpho_rrtmg_c080918.nc. It's named water_refindex_rrtmg_c080910.nc
Describe any substantial timing or memory changes: not checked
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/build-namelist
. add all 3-mode species to the rad_climate specifier
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. update the modal_optics file
. update the physprop files
. change modal_optics to modal_optics_file and change the attribute from
chem to aero which is set in build-namelist depending on whether 3mode or
7mode aerosols are used. But it's no longer dependent on particular
chemistry scenarios in anticipation of a prescribed modal aerosol
capability.
. add water_refindex_file
models/atm/cam/bld/namelist_files/namelist_definition.xml
. add water_refindex_file
. change modal_optics to modal_optics_file and move out of rad_cnst_nl and
into modal_aer_opt_nl.
models/atm/cam/src/chemistry/modal_aero/modal_aero_initialize_data.F90
. use rad_constituents interface to set specdens_amode
. update the rad_cnst_get_aer_props interfaces to get complex refractive
index and copy to local storage
models/atm/cam/src/control/filenames.F90
. remove modal_optics
models/atm/cam/src/control/runtime_opts.F90
. remove modal_optics
. add call to modal_aer_opt_readnl
models/atm/cam/src/physics/cam/aer_rad_props.F90
. rename aer_rad_props_get_clim_{sw,lw} to aer_rad_props_{sw,lw}
. remove loop over sw bands around call to modal_aero_sw
. move the call to modal_size_parameters inside modal_aero_sw.
models/atm/cam/src/physics/cam/modal_aer_opt.F90
. remove unused AEROCOM declarations
. remove calc_optics option - currently hardcoded to not operate
. remove miefit, fitcurv, fitcurvlin, chebft, write_modal_optics
. remove BACKSCAT which was only used for mie calcs
. remove "use mpishorthand" -- not referenced
. remove module variables nrefr, nrefi -- they're not used as variables, so
just use the parameters prefr, prefi
. move the loop over sw bands inside modal_aero_sw
. move the call to modal_size_parameters inside modal_aero_sw.
. modal_aero_sw
- change args to match aer_rad_props_sw (except diagnostic capability not
implemented yet)
- pull some initializations outside the loop over sw bands
- switch the order of the outer level and mode do loops putting the mode
loop on the outside. This causes roundoff diffs in AODABS, AODVIS,
SSAVIS. These are diagnostic fields that are summed over both level
and mode, so interchanging that sum introduces a roundoff difference.
- move the initializations of the diagnostic output outside the sw loop.
Need to put the burden calc inside a savaervis conditional.
- move the mode loop outside the sw band loop. The diagnostic output
moves outside all the loops since the arrays used for output are only
set for the visible band.
- replace direct use of state%q by call to rad_cnst_get_aermmr in the
aerosol species loop
- refactor dust aod diagnostic to look for xname_spectype='dust'
- remove the else clause in the computation of the dustaodmode
calculation. This is a bug. The calc is accumulating contributions to
dustaodmode inside a loop over levels. Setting the whole result to
zero upon encountering a layer with wetvol<=1.e-40 is zeroing the
entire accumulation over levels to that point rather than just zeroing
the contribution from that level.
- use rad_cnst_get_aer_props to get species densities
. modal_aero_lw::
- change args to match aer_rad_props_lw (except diagnostic capability not
implemented yet)
- repeat the same refactoring that was done in modal_aero_sw to reorder
loops and obtain the species mmr, density, and refractive index from
the rad_constituent interfaces.
. read_modal_optics
- modes arg removed. Will instead set module variable ntot_amode with
the number of modes.
- remove all args that are just being used to set module data, and
instead set it directly.
- read new variables that contain the number of species in each mode, and
the species names and types from optics file.
- move sigma_logr_aer from local to module data so that it can be used
instead of modal_aero_data::sigmag_amode
. add subroutine modal_aer_opt_readnl to read modal_optics_file and
water_refindex_file.
. add new subroutine read_water_refindex and use it to set the module data
for complex water refractive index
. remove the call to rad_cnst_get_clim_aer_prop to access water refractive
index data
. remove unused wavelength band arguments from modal_aer_opt_init
. access species refractive index via rad_cnst_get_aer_props
. add alnsg_amode as module data that is set using the sigma_logr_aer field
read from the modal optics file, rather than being set from the hardcoded
values of sigmag_amode in modal_aero_data
. allocate/deallocate arrays with sizes that depend on the number of modes.
. remove ifdef MODAL_AERO -- the code is now independent of prognostic MAM
models/atm/cam/src/physics/cam/phys_prop.F90
. remove reading of water refractive index.
. refindex_aer_init - get rid of the ifdef MODAL_AERO by adding checks that
the refindex variables are present in the file before trying to read
them. Only allocate space if the variable is present.
. supply the complex version of the refractive index data directly rather
than as real and imaginary parts.
models/atm/cam/src/physics/cam/rad_constituents.F90
. implement generic rad_cnst_get_aer_mmr
. implement generic rad_cnst_get_props
. remove optional args that provide water refindex via the
rad_cnst_get_aer_props interfaces.
. update the rad_cnst_get_aer_props interfaces to pass complex refractive
index
models/atm/cam/src/physics/cam/radiation.F90
. rename aer_rad_props_get_clim_{sw,lw} to aer_rad_props_{sw,lw}
models/atm/cam/src/physics/rrtmg/radiation.F90
. rename aer_rad_props_get_clim_{sw,lw} to aer_rad_props_{sw,lw}
. remove wavelength band args from call to modal_aer_opt_init
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Fri Nov 12 13:02:18 MST 2010
021 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...............................FAIL! rc= 7 at Fri Nov 12 13:29:39 MST 2010
036 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ......................................FAIL! rc= 7 at Fri Nov 12 14:12:22 MST 2010
040 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .....................................FAIL! rc= 7 at Fri Nov 12 14:32:34 MST 2010
047 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...........................FAIL! rc= 7 at Fri Nov 12 15:22:29 MST 2010
061 sm731 TSM.sh h16x4c3dh aqua 9s ........................................FAIL! rc= 8 at Fri Nov 12 18:31:10 MST 2010
edinburgh/lf95: All PASS except:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Fri Nov 12 13:01:04 MST 2010
009 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Fri Nov 12 13:45:01 MST 2010
015 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Fri Nov 12 14:30:07 MST 2010
022 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Fri Nov 12 15:43:19 MST 2010
030 bl312 TBL.sh f10c5dm ghgrmp 9s ........................................FAIL! rc= 7 at Fri Nov 12 16:28:54 MST 2010
edinburgh/pgi: All PASS except:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Fri Nov 12 11:41:25 MST 2010
009 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Fri Nov 12 11:51:29 MST 2010
015 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Fri Nov 12 12:02:37 MST 2010
022 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Fri Nov 12 12:19:57 MST 2010
030 bl312 TBL.sh f10c5dm ghgrmp 9s ........................................FAIL! rc= 7 at Fri Nov 12 12:33:58 MST 2010
The baseline failures are all due to roundoff diffs introduced into the
diagnostic fields AODABS, AODVIS, and SSAVIS.
sm731 is a pre-existing failure.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB except roundoff changes in some
diagnostic fields for trop_mam3.
===============================================================
===============================================================
Tag name: cam5_0_26
Originator(s): jet
Date: Wed Nov 03 2010
One-line Summary:Fix bug with new dynamics namelists,update external tags for high res support and Flow control
Purpose of changes: There was a group name error for the spectral namelists that prevented user modifications of the dynamics variables via the namelist. Support for high resolution
Bugs fixed (include bugzilla ID):Fixed namelist bug created during last commit.
Describe any changes made to build system:Updated SVN_EXTERNALS to reflect new tags for high res support in the externals. Also brought remaining externals up to beta 10.
Describe any changes made to the namelist:Updated namelist defaults for high res support.
List any changes to the defaults for the boundary datasets:Added default datasets for high res.
Describe any substantial timing or memory changes:
Code reviewed by:jet
List all subroutines eliminated:
List all subroutines added and what they do:
Added a diagnostic test that checks that a variable in the
spectral namelist can be successfully changed.
A models/atm/cam/test/system/TDD.sh
List all existing files that have been modified, and describe the changes:
new diagnostic test that checks that a variable in the spectral
namelist can be successfully changed.
M models/atm/cam/test/system/input_tests_master
M models/atm/cam/test/system/tests_pretag_edinburgh
M models/atm/cam/test/system/nl_files/ttrac
***********************
additional support for high res models - default tuning parameters added
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
***********************
Fixed bug with new dynamics namelist reads
M models/atm/cam/src/dynamics/sld/sld_control_mod.F90
M models/atm/cam/src/dynamics/eul/eul_control_mod.F90
***********************
Updated externals to a minimum of ccsm beta10 or the
latest tags with updates for add high res support
M SVN_EXTERNAL_DIRECTORIES
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
004 bl132 TBL.sh e48c4dh ghgrmp+1850_cam4 9s ....FAIL!
011 bl137 TBL.sh e48c5h outfrq24h 2d ............FAIL!
017 bl331 TBL.sh f4c4dh co2rmp+1850_cam4 9s .....FAIL!
021 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s .....FAIL!
033 bl336 TBL.sh wm4c4h outfrq3s+waccm_1850_cam4 9s ..........FAIL!
036 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ............FAIL!
040 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s ...........FAIL!
Failure for test 4 was expected. I added a new test of
the dynamics namelist to a pre-existing test. Because
of the namelist bug a new dynamics variable was not
updated correctly in the previous tag. In the new tag
the changed namelist variable is used and produces different
answers
The rest of the failures were expected due to an update to the
external libraries. The only field that failed the bfb tests
was U10 which was added to the trunk by updating the externals
to ccsm_beta10.
edinburgh/lf95:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .............FAIL!
009 bl112 TBL.sh e8c5dm ghgrmp 9s ...............FAIL!
015 bl114 TBL.sh e8c5dm co2rmp 9s ...............FAIL!
022 bl311 TBL.sh f10c5t5mdm ttrac 9s ............FAIL!
030 bl312 TBL.sh f10c5dm ghgrmp 9s ..............FAIL!
035 bl314 TBL.sh wg10c4dm outfrq3s 9s ...........FAIL!
038 bl315 TBL.sh f10c3dm outfrq3s 9s ............FAIL!
043 bl318 TBL.sh f10c4cdm outfrq3s 9s ...........FAIL!
All expected as explained above.
edinburgh/pgi or jaguar/pgi:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .............FAIL!
009 bl112 TBL.sh e8c5dm ghgrmp 9s ...............FAIL!
015 bl114 TBL.sh e8c5dm co2rmp 9s ...............FAIL!
022 bl311 TBL.sh f10c5t5mdm ttrac 9s ............FAIL!
030 bl312 TBL.sh f10c5dm ghgrmp 9s ..............FAIL!
035 bl314 TBL.sh wg10c4dm outfrq3s 9s ...........FAIL!
038 bl315 TBL.sh f10c3dm outfrq3s 9s ............FAIL!
043 bl318 TBL.sh f10c4cdm outfrq3s 9s ...........FAIL!
All expected as explained above.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
BFB with the exception of diagnostic field U10
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_0_25
Originator(s): jet
Date: Fri Oct 29 2010
One-line Summary:clean up dynamics, add eulerian subcycling, support for t341
Purpose of changes:Commit of code modifications from Kate Evans,
Mark Taylor and Pat Worley. These modifications add support
running the T341 eulerian model. Mark Taylor's mods to allow
eulerian subcycling were also added to correct an instability in
the cloud/convection parameterization at higher resolutions.
As part of merging Mark Taylor's mods in correctly some clean
up of the dynamics routines was required.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:Kate Evan's added support for
T341 resolutions
Describe any changes made to the namelist:Moved dynamic specific variables
out of the cam_inparm namelist into a dynamics namelist. Also
split the spmd and dynamics variables into separate namelists.
List any changes to the defaults for the boundary datasets: Added boundary datasets for t341 support.
Describe any substantial timing or memory changes:
Code reviewed by:jet,eaton
List all subroutines eliminated:
D models/atm/cam/src/control/decompinit.F90
D models/atm/cam/src/control/fv_control_mod.F90
D models/atm/cam/src/physics/cam/spmd_phys.F90
List all subroutines added and what they do:
models/atm/cam/src/dynamics/fv/fv_control_mod.F90
runtime opts was cleaned up, moving the dynamics namelist
initializations into their own init routines under dynamics.
After this was done fv_control_mod was moved into the fv dynamics
directory.
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/config_files/horiz_grid.xml
M models/atm/cam/bld/build-namelist
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
M models/atm/cam/bld/namelist_files/namelist_definition.xml
M models/atm/cam/doc/ChangeLog
M models/atm/cam/src/control/readinitial.F90
M models/atm/cam/src/control/runtime_opts.F90
M models/atm/cam/src/control/cam_control_mod.F90
M models/atm/cam/src/control/startup_initialconds.F90
M models/atm/cam/src/control/cam_restart.F90
M models/atm/cam/src/utils/spmd_utils.F90
M models/atm/cam/src/physics/cam/phys_grid.F90
M models/atm/cam/src/dynamics/sld/spegrd.F90
M models/atm/cam/src/dynamics/sld/dyn_comp.F90
M models/atm/cam/src/dynamics/sld/hordif.F90
M models/atm/cam/src/dynamics/sld/spmd_dyn.F90
M models/atm/cam/src/dynamics/sld/hdinti.F90
M models/atm/cam/src/dynamics/eul/getinterpnetcdfdata.F90
M models/atm/cam/src/dynamics/eul/inital.F90
M models/atm/cam/src/dynamics/eul/iop.F90
M models/atm/cam/src/dynamics/eul/dyn.F90
M models/atm/cam/src/dynamics/eul/grcalc.F90
M models/atm/cam/src/dynamics/eul/dynpkg.F90
M models/atm/cam/src/dynamics/eul/stepon.F90
M models/atm/cam/src/dynamics/eul/courlim.F90
M models/atm/cam/src/dynamics/eul/restart_dynamics.F90
M models/atm/cam/src/dynamics/eul/tfilt_massfix.F90
M models/atm/cam/src/dynamics/fv/dyn_comp.F90
M models/atm/cam/src/dynamics/fv/spmd_dyn.F90
M models/atm/cam/src/dynamics/fv/inital.F90
M models/atm/cam/src/dynamics/fv/restart_dynamics.F90
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
edinburgh/lf95:
edinburgh/pgi or jaguar/pgi:
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_0_24
Originator(s): dfeddema
Date: Mon Oct 25 2010
One-line Summary: CAM cleanup for code which stores and retrieves pbuf indices. These changes 1) save pbuf index as local module data when pbuf_add is called 2) move pbuf_get_fld_idx calls from run routines to init routines.
Purpose of changes: readability, consistency and performance
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: not checked
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
[...]/models/atm/cam/src/physics/rrtmg/slingo.F90
[...]/models/atm/cam/src/physics/rrtmg/radiation.F90
[...]/models/atm/cam/src/physics/rrtmg/oldcloud.F90
[...]/models/atm/cam/src/physics/rrtmg/ebert_curry.F90
[...]/models/atm/cam/src/physics/cam/zm_conv_intr.F90
[...]/models/atm/cam/src/physics/cam/vertical_diffusion.F90
[...]/models/atm/cam/src/physics/cam/stratiform.F90
[...]/models/atm/cam/src/physics/cam/sslt_rebin.F90
[...]/models/atm/cam/src/physics/cam/radiation.F90
[...]/models/atm/cam/src/physics/cam/physpkg.F90
[...]/models/atm/cam/src/physics/cam/param_cldoptics.F90
[...]/models/atm/cam/src/physics/cam/convect_shallow.F90
[...]/models/atm/cam/src/physics/cam/convect_deep.F90
[...]/models/atm/cam/src/physics/cam/cloud_fraction.F90
[...]/models/atm/cam/src/physics/cam/check_energy.F90
[...]/models/atm/cam/src/physics/cam/cam_diagnostics.F90
[...]/models/atm/cam/src/physics/cam/mo_gas_phase_chemdr.F90
[...]/models/atm/cam/src/chemistry/mozart/iondrag.F90
[...]/models/atm/cam/src/chemistry/mozart/chemistry.F90
[...]/models/atm/cam/src/chemistry/bulk_aero/aerosol_intr.F90
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
061 sm731 TSM.sh h16x4c3dh aqua 9s ........................................FAIL! rc= 8 at Sun Oct 24 14:54:04 MDT 2010
- sm731 failure is pre-existing
edinburgh/lf95: All PASS
edinburgh/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_0_23
Originator(s): mvr,mvertens
Date: Thu Oct 21 2010
One-line Summary: mods for processing u10 (10m wind speed) which is now
being received from drv code; externals updated for new drv, cice and pio tags
Purpose of changes:
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: not checked
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
M models/atm/cam/src/control/camsrfexch_types.F90
M models/atm/cam/src/cpl_mct/atm_comp_mct.F90
M models/atm/cam/src/physics/cam/cam_diagnostics.F90
M models/atm/cam/src/cpl_esmf/atm_comp_esmf.F90
- mods for processing u10 now being received from the driver
M models/atm/cam/src/utils/cam_csim4/cpl_mct/ice_comp_mct.F90
- removed references to variable sicthk
M SVN_EXTERNAL_DIRECTORIES
M .
- externals updated for new drv, cice and pio tags
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
061 sm731 TSM.sh h16x4c3dh aqua 9s ........................................FAIL! rc= 8 at Wed Oct 20 22:26:59 MDT 2010
- sm731 failure is pre-existing
edinburgh/lf95: All PASS
edinburgh/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_0_22
Originator(s): mvr,hannay
Date: Wed Oct 13 2010
One-line Summary: corrected the default settings for condensate-
to-rain autoconversion coefficients for cam5 configurations
Purpose of changes:
cam5 runs (other than those at 1.9x2.5) were using values meant
for cam4 configurations
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: none
Describe any changes made to the namelist:
- new default values for condensate-to-rain autoconversion coefficients
at cam5 configurations
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes: not checked
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Tue Oct 12 20:51:50 MDT 2010
013 bl330 TBL.sh f4c5pdh aqpgro 3s ........................................FAIL! rc= 7 at Tue Oct 12 20:59:06 MDT 2010
021 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...............................FAIL! rc= 7 at Tue Oct 12 21:18:23 MDT 2010
036 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ......................................FAIL! rc= 7 at Tue Oct 12 21:56:41 MDT 2010
061 sm731 TSM.sh h16x4c3dh aqua 9s ........................................FAIL! rc= 8 at Wed Oct 13 00:34:44 MDT 2010
- sm731 failure is pre-existing
- baseline failures were expected at cam5 configurations (other than 1.9x2.5) with changes to tuning parameters
edinburgh/lf95: All PASS except:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Tue Oct 12 17:01:21 MDT 2010
008 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Tue Oct 12 17:44:55 MDT 2010
010 bl113 TBL.sh e8c5pdm aqpgro 3s ........................................FAIL! rc= 7 at Tue Oct 12 18:19:12 MDT 2010
014 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Tue Oct 12 18:25:07 MDT 2010
021 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Tue Oct 12 19:33:09 MDT 2010
029 bl312 TBL.sh f10c5dm ghgrmp 9s ........................................FAIL! rc= 7 at Tue Oct 12 20:16:18 MDT 2010
031 bl313 TBL.sh f10c5pdm aqpgro 3s .......................................FAIL! rc= 7 at Tue Oct 12 20:41:53 MDT 2010
039 bl316 TBL.sh f10c5pdm aqpgro+aquaplanet_cam5 3s .......................FAIL! rc= 7 at Tue Oct 12 21:47:57 MDT 2010
- baseline failures were expected at cam5 configurations (other than 1.9x2.5) with changes to tuning parameters
edinburgh/pgi: All PASS except:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Tue Oct 12 16:28:44 MDT 2010
008 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Tue Oct 12 16:37:59 MDT 2010
010 bl113 TBL.sh e8c5pdm aqpgro 3s ........................................FAIL! rc= 7 at Tue Oct 12 16:45:03 MDT 2010
014 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Tue Oct 12 16:47:05 MDT 2010
021 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Tue Oct 12 17:03:29 MDT 2010
029 bl312 TBL.sh f10c5dm ghgrmp 9s ........................................FAIL! rc= 7 at Tue Oct 12 17:15:03 MDT 2010
031 bl313 TBL.sh f10c5pdm aqpgro 3s .......................................FAIL! rc= 7 at Tue Oct 12 17:22:05 MDT 2010
039 bl316 TBL.sh f10c5pdm aqpgro+aquaplanet_cam5 3s .......................FAIL! rc= 7 at Tue Oct 12 17:35:58 MDT 2010
- baseline failures were expected at cam5 configurations (other than 1.9x2.5) with changes to tuning parameters
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB, except for cam5 configurations other than 1.9x2.5
===============================================================
===============================================================
Tag name: cam5_0_21
Originator(s): mvr
Date: Tue Oct 12 2010
One-line Summary: new tuning mods added as defaults for T31; mods to
test scripts to reflect changes in the cice decomposition formulation
Purpose of changes:
T31 tuning mods...
cldfrc_rhminh = 0.50
cldwat_icritc = 1.e-6
cldwat_r3lcrit = 1.e-6 (and moved into a namelist variable)
do_tms = true
tms_orocnst = 1.0
tms_z0fac = 0.1 (and moved into a namelist variable)
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: none
Describe any changes made to the namelist:
- new tuning mods added as defaults for T31
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes: not checked
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/TCB.sh
- mods to utilize fixes made to cice decomp script
M models/atm/cam/test/system/CAM_decomp.sh
- bug fix to test script for properly altering task count for hybrid tests
M models/atm/cam/test/system/TBR.sh
M models/atm/cam/test/system/TER.sh
M models/atm/cam/test/system/TSM.sh
- test scripts now utilize new nl option for cice distribution type, roundrobin
M models/atm/cam/test/system/test_driver.sh
- bluefire testing returned to use all 64procs of node with fixes to cice decomp
M models/atm/cam/bld/build-namelist
- cleaned up handling of tms variables, added code for new namelist parameters
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- added default values for T31 (see description above)
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- added entries for new namelist variables
M models/atm/cam/src/physics/cam/trb_mtn_stress.F90
M models/atm/cam/src/physics/cam/phys_control.F90
M models/atm/cam/src/physics/cam/cldwat.F90
M models/atm/cam/src/physics/cam/vertical_diffusion.F90
- changes to reflect new handling of tms namelist variables
M SVN_EXTERNAL_DIRECTORIES
M .
- updated to new cice tag with fixes to its decomposition formulation
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
004 bl132 TBL.sh e48c4dh ghgrmp+1850_cam4 9s ..............................FAIL! rc= 7 at Tue Oct 12 08:48:07 MDT 2010
006 bl133 TBL.sh e48c4pdm aqpgro 3s .......................................FAIL! rc= 7 at Tue Oct 12 08:48:08 MDT 2010
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Tue Oct 12 08:48:15 MDT 2010
061 sm731 TSM.sh h16x4c3dh aqua 9s ........................................FAIL! rc= 8 at Tue Oct 12 08:50:49 MDT 2010
- sm731 failure is pre-existing
- baseline failures were expected at T31 with changes to tuning parameters
edinburgh/lf95: All PASS
edinburgh/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB, except at T31
===============================================================
===============================================================
Tag name: cam5_0_20
Originator(s): jet
Date: Mon Oct 4, 2010
One-line Summary: Updated SVN_EXTERNALS to include workaround for intel compile bug when compiling shr_scam_mod.
Purpose of changes: Updated SVN_EXTERNALS to include workaround for intel compile bug when compiling shr_scam_mod.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: None
Describe any changes made to the namelist: None
List any changes to the defaults for the boundary datasets: None
Describe any substantial timing or memory changes: None
Code reviewed by:jet,tcraig
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/src/chemistry/mozart/tgcm_forcing.F90
M SVN_EXTERNAL_DIRECTORIES
tgcm_forcing had what looked to be debug io that was
incompatible with single column mode. For now that write
statement was commented out.
bluefire: All PASS except:
061 sm731 TSM.sh h16x4c3dh aqua 9s ........................................FAIL! rc= 8 at Thu Sep 30 11:50:18 MDT 2010
Expected failure from previous tags.
edinburgh/lf95: BFB scam runs completed on tramhill with lf95
edinburgh/pgi or jaguar/pgi: BFB scam runs completed on tramhill with pfg90
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff? bfb
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_0_19
Originator(s): mvr,eaton
Date: Thr Sep 30 2010
One-line Summary: update component externals to those of cesm1_0_beta08;
corrected ghg dataset for cam5 use_case; added default ic file for 0.5deg cam5
Purpose of changes:
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets:
- new default initial conditions file for 0.5deg cam5
Describe any substantial timing or memory changes: not checked
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- added listing for default initial conditions file at 0.5deg cam5
M models/atm/cam/bld/namelist_files/use_cases/1850-2005_cam5.xml
- corrected the ghg dataset being used
M SVN_EXTERNAL_DIRECTORIES
M .
- new external definitions for drv, clm, cice, csm_share and scripts
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
061 sm731 TSM.sh h16x4c3dh aqua 9s ........................................FAIL! rc= 8 at Thu Sep 30 11:50:18 MDT 2010
047 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...........................FAIL! rc= 7 at Thu Sep 30 14:15:47 MDT 2010
- sm731 failure is pre-existing
- bl375 baseline comparison was expected to fail due to the change made to the ghg dataset used
edinburgh/lf95: All PASS
edinburgh/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_0_18
Originator(s): pworley, jedwards, eaton
Date: Sat Sep 25 11:20:58 MDT 2010
One-line Summary: Mods for COSP; phys_grid column pair fix
Purpose of changes:
. Changes were made to configure and to the COSP source tree to enable the
build of COSP from the CESM scripts as part of the atm library (the
standalone CAM build builds COSP as a separate library).
. The COSP interface module was incurring an unnecessary memory overhead
when the COSP simulator was not being used. This has been remedied
(fixes from Jim Edwards).
. Modify configure to remove restrictions against building for spmd or smp
modes with Darwin OS.
Bugs fixed (include bugzilla ID):
. Fix to column pairing algorithm in phys_grid. From Pat Worley:
> The error does not break the
> code, it just erroneously disables the wrap map of the columns as
> well as the column pairing. This essentially disables all load
> balancing when the namelist parameter phys_twin_algorithm == 0 or
> when using HOMME and not setting phys_twin_algorithm = 1.
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes:
. Replace static allocation of atrain arrays in the COSP interface by
dynamic allocation to remove the overhead when COSP isn't being used.
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/SVN_EXTERNAL_DIRECTORIES
. update cosp to cosp_v1_3_100924
models/atm/cam/bld/configure
. only write cosp Makefile if doing a cam standalone build
. modify the write_filepath_ccsm method to include the cosp filepaths if
running with cosp enabled
. remove checks that disallow darwin builds with SPMD or SMP
models/atm/cam/src/physics/cam/cospsimulator_intr.F90:
. mpibcast of cosp_histfile_num, atrainday, atrainhr, atrainmin, atrainsec
had wrong type
. change large atrain orbit data arrays from static to allocatable
models/atm/cam/src/physics/cam/phys_grid.F90
. fix for twin algorithm
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
061 sm731 TSM.sh h16x4c3dh aqua 9s ........................................FAIL! rc= 8 at Fri Sep 24 23:36:01 MDT 2010
sm731 failure is pre-existing
edinburgh/lf95: All PASS
edinburgh/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_0_17
Originator(s): mvr,jek,eaton, pworley, bardeenc
Date: 16 Sep 2010
One-line Summary: update/fixes made to COSP code; add a test to test_driver
for exercising COSP; Supply a missing initialization in the reproducible sum code
Purpose of changes:
1) additional logic so that cosp outputs that require both radar and
lidar simulator are saved only when both simulators are run
2) double counting of the convective cloud water in cosp inputs. this
required defining some new variables for the stratiform cloud water
within cospsimulator_intr.F90.
3) Pressures and heights not defined correctly in cosp inputs.
Yuying: "Based on TYPE COSP_GRIDBOX (line 282-286) in cosp_types.F90,
zlev & p are height and pressure at model levels,
and zlev_half & ph are values at bottom (interface) of model layer.
The definition in cospsimulator_intr.F90 (line 1581-1584) shows that
gbx%p & gbx%zlev are interface values while
gbx%ph & gbx%zlev_half are at model levels."
4) convection *_flxprc are for rain+snow while stratiform *_flxprc was
for rain only. I am changing ls_flxprc so that it includes
both rain and snow in cldwat.F90, ~ line 946, rkflx = precab instead of
rkflx = precab - snowab. It is confusing to have things labeled
*_flxprc when some include rain and others include rain+snow. I also
made name mods in stratiform.F90. I modified cospsimulator.F90
accordingly -- see lines 1447 - 1469.
5) Jim wanted a namelist variable so that you can specify which CAM
history tape COSP outputs are written to
I used Jerry's mods for history_budget_histfile_num as a template.
I added cosp_histfile_num as a namelist variable by making mods to both
namelist_definition.xml and to cospsimulator_intr.F90.
6) circa line 4288, dimension name wrong --
"dimen4a(3)=cosp_prstau_modis_dim" should be
"dimen4m(3)=cosp_prstau_modis_dim"
7) for cfad_sr532_cal. nbze_cosp = nsr_cosp = 15. this leads to if
statement problems because cam_history.F90 only checks flags and
dimensions sizes. i set the flag_**dbze** = .false when flag_cosphtsrlev
= .true circa line 1281
8) similar problem for fisccp1_cosp and isccp simulator flags. fix circa
line 1293
Bugs fixed (include bugzilla ID):
. fix for missing initializer in repro_sum_mod module (from Chuck Bardeen
and Pat Worley)
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do:
A models/atm/cam/test/system/config_files/f10c4cdm
- new configuration options test file for testing of cosp
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/tests_pretag_edinburgh
- added new cosp test to default set of required pretag tests on edinburgh
M models/atm/cam/test/system/TER.sh
- modified the exact restart test to not change number of tasks/threads upon
restart for any test running cosp
M models/atm/cam/test/system/input_tests_master
- added definition of new cosp tests to master list
M models/atm/cam/test/system/CAM_runcmnd.sh
- change to test scripts to set run command on edinburgh while accounting for
potential tests of both mpi-only or hybrid
M models/atm/cam/src/utils/repro_sum_mod.F90
- Supply a missing initialization in the reproducible sum code. The code
was not producing wrong answers, but could possibly produce a runtime
failure due to a bad argument (fix from Bardeen, Worley).
M models/atm/cam/bld/namelist_files/namelist_definition.xml
M models/atm/cam/src/control/cam_history.F90
M models/atm/cam/src/physics/cam/stratiform.F90
M models/atm/cam/src/physics/cam/cospsimulator_intr.F90
M models/atm/cam/src/physics/cam/cldwat.F90
- cosp changes (see description above)
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
061 sm731 TSM.sh h16x4c3dh aqua 9s ........................................FAIL! rc= 8 at Thu Sep 16 00:23:05 MDT 2010
- existing failure
edinburgh/lf95:
042 bl318 TBL.sh f10c4cdm outfrq3s 9s .....................................FAIL! rc= 5 at Wed Sep 15 19:38:59 MDT 2010
- new test failed because it is not defined in baseline code
edinburgh/pgi:
042 bl318 TBL.sh f10c4cdm outfrq3s 9s .....................................FAIL! rc= 5 at Wed Sep 15 16:13:34 MDT 2010
- new test failed because it is not defined in baseline code
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate): b4b
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_0_16
Originator(s): Francis Vitt
Date: 15 Sep 2010
One-line Summary: Added capability to overwrite the cam surface GHG
values with data from waccm lower boundary file.
Purpose of changes:
This makes the co2vmr value used in the WACCM nlte_fomichev module
consistent with the value in the WACCM lower boundary file. This
capability has been extended to non WACCM configurations so that
a WACCM LBC file can be used set the CAM GHG surface values.
Also included is a change in the age-of-air tracers to use values
from the initial conditions file as initial values. To override
this default behavior set aoa_read_from_ic_file namelist variable
to false.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
D models/atm/cam/src/chemistry/mozart/mo_flbc.F90
D models/atm/cam/src/chemistry/mozart/m_types.F90
- files moved to the models/atm/cam/src/chemistry/utils directory so
that they are available in non-chemistry configurations
List all subroutines added and what they do:
A models/atm/cam/src/chemistry/utils/mo_flbc.F90
A models/atm/cam/src/chemistry/utils/m_types.F90
- files moved from the models/atm/cam/src/chemistry/mozart/ directory
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/build-namelist
- if a LBC file is defined then do not set default settings for
bndtvghg, co2vmr, ch4vmr, n2ovmr, f11vmr, f12vmr
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- aoa_read_from_ic_file defaults to TRUE
- more explanation is added to ext_frc_specifier namelist description
- flbc_* namelist variables have been moved to "chem_srfvals_nl" namelist
M models/atm/cam/bld/namelist_files/use_cases/waccm_2000_cam4.xml
- removed the co2vmr setting since this is now set by the LBC file
M models/atm/cam/src/physics/cam/phys_gmean.F90
- added a subroutine to compute global mean of a single field
M models/atm/cam/src/physics/cam/chem_surfvals.F90
- changes to read flbc_* namelist variables
- changes to overwrite the GHG surface values with global means from
the mo_flbc module
M models/atm/cam/src/physics/cam/aoa_tracers.F90
- changes the default ICs to be set from the CAM initial conditions file
M models/atm/cam/src/chemistry/utils/tracer_data.F90
- set initial value of list_cycled to .false.
- if data path is empty do not append "/filename" to it
M models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
M models/atm/cam/src/chemistry/mozart/mo_chemini.F90
M models/atm/cam/src/chemistry/mozart/chemistry.F90
M models/atm/cam/src/chemistry/mozart/mo_srf_emissions.F90
- changes needed to make mo_flbc available to non-chemistry configurations
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
033 bl336 TBL.sh wm4c4h outfrq3s+waccm_1850_cam4 9s .......................FAIL! rc= 7 at Wed Sep 15 10:13:08 MDT 2010
045 bl373 TBL.sh fm1.9c4dh outfrq3s 9s ....................................FAIL! rc= 7 at Wed Sep 15 10:13:48 MDT 2010
060 bl379 TBL.sh wm1.9c4h outfrq24h+waccm_1850_cam4 2d ....................FAIL! rc= 7 at Wed Sep 15 10:45:34 MDT 2010
061 sm731 TSM.sh h16x4c3dh aqua 9s ........................................FAIL! rc= 8 at Wed Sep 15 10:46:35 MDT 2010
The WACCM baseline tests bl336 and bl379 failures are due to the change in the co2vmr
surface value used in the nlte_fomichev module and the change in initial conditions
of the age-of-air tracers
The trop_mozart baseline test bl379 failure is due to the change in GHG values used in
the radiation which are now based on the chemistry lower conditions file
Test sm731 failure is pre-existing
edinburgh/lf95:
034 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7 at Mon Sep 13 17:51:19 MDT 2010
This failure is due to the change in initial conditions of the age-of-air tracers
edinburgh/pgi:
034 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 7 at Mon Sep 13 15:09:27 MDT 2010
This failure is due to the change in initial conditions of the age-of-air tracers
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_0_15
Originator(s): mirin, eaton
Date: Wed Sep 8 11:46:04 MDT 2010
One-line Summary: Fix for FV div4 filter; refactor pertlim code; rcp6.0 update
Purpose of changes:
. Fixes to indices for ghost regions in FV div4 filter (from Art Mirin).
. Refactor the application of a perturbation to the initial T field so that
it is distributed. Previously the 3D T field was read to the master
task and then scattered after the perturbation was applied.
. Modify the rcp6.0 use case to use the datm version of the prescribed
aerosol deposition fluxes.
Bugs fixed (include bugzilla ID):
. The results of the FV dycore with the div4 filtering on (only on by
default for 1/2 deg and higher resolution) where found to depend on the
decomposition. The fixes to ghost region indexing has resolved this.
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets:
. change aerosol deposition dataset in use case 2005-2100_cam4_rcp60
Describe any substantial timing or memory changes:
. The change to how an initial temperature perturbation is applied
substantially reduces memory requirements (eliminating the need for a
global 3D field) so that perturbation growth tests may be done at higher
resolutions.
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/namelist_files/use_cases/2005-2100_cam4_rcp60.xml
. modify aerodep_flx_file to use the datm dataset.
models/atm/cam/src/dynamics/fv/dyn_comp.F90
. remove debug write to all procs
models/atm/cam/src/dynamics/fv/inidat.F90
. remove reading 3D T field to masterproc and scattering when pertlim is
specified.
. add code that uses the global column number to seed the random number
generator. This makes setting the perturbation independent of
decomposition.
models/atm/cam/src/dynamics/fv/dynamics_vars.F90
models/atm/cam/src/dynamics/fv/sw_core.F90
. Fixes for indices in ghost regions of FV div4 filter.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
061 sm731 TSM.sh h16x4c3dh aqua 9s ........................................FAIL! rc= 8 at Tue Sep 7 21:14:08 MDT 2010
sm731 failure is pre-existing.
edinburgh/lf95: All PASS
edinburgh/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB except FV run that use the div4
filter produces larger than roundoff diffs. This filter is the default
for 1/2 deg and higher resolutions.
===============================================================
===============================================================
Tag name: cam5_0_14
Originator(s): mvr
Date: Wed Sep 01
One-line Summary: use_cases added for waccm, rcp6.0; test added to default test
suite; removed reference to shr_date_mod; updated externals to cesm1_0_beta07
Purpose of changes:
- a test was added to the test suite to ensure a run of stand-alone-cam vs
an "F" case via the ccsm scripts was b4b
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: not checked
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do:
A models/atm/cam/test/system/nl_files/fcase
A models/atm/cam/test/system/nl_files/user_nl_cam
-new test files required for new "stand-alone-cam vs fcase" test
A models/atm/cam/bld/namelist_files/use_cases/2005-2100_cam4_rcp60.xml
A models/atm/cam/bld/namelist_files/use_cases/waccm_2005-2100_cam4_rcp45.xml
A models/atm/cam/bld/namelist_files/use_cases/waccm_1955-2005_cam4.xml
-new use_cases added
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/tests_pretag_bluefire
M models/atm/cam/test/system/tests_posttag_jaguar_cb
M models/atm/cam/test/system/tests_posttag_jaguar
M models/atm/cam/test/system/input_tests_master
-added new "stand-alone-cam vs fcase" test to master and default test lists
M models/atm/cam/test/system/TCB_ccsm.sh
-configure-time mods needed for new "stand-alone-cam vs fcase" test
M models/atm/cam/test/system/TBR.sh
M models/atm/cam/test/system/TEQ_ccsm.sh
M models/atm/cam/test/system/TEQ.sh
-correction to output messages of test scripts
M models/atm/cam/test/system/test_driver.sh
M models/atm/cam/test/system/TER_ccsm.sh
M models/atm/cam/test/system/TSM_ccsm.sh
-mods to enable testing of ccsm scripts via cam test suite on jaguar
M models/atm/cam/src/physics/cam/tidal_diag.F90
-removed the usage of the module shr_date_mod
M SVN_EXTERNAL_DIRECTORIES
M .
-updated external definitions for drv, csm_share and scripts to represent those
of beta07 as closely as possible
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
007 sm731 TSM.sh h16x4c3dh aqua 9s ........................................FAIL! rc= 8 at Tue Aug 31 17:14:51 MDT 2010
016 bl992 TBL_ccsm.sh f19_g16 E 2d ........................................FAIL! rc= 7 at Tue Aug 31 17:15:06 MDT 2010
-bl992 failure is due to change in the setting for co2_ppmv
-sm731 failure is pre-existing
edinburgh/lf95: All PASS
edinburgh/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_0_13
Originator(s): eaton
Date: Fri Aug 20 12:37:04 MDT 2010
One-line Summary: fix HOMME for repro_sum change; mods to high-res FV config; logfile output; COSP updates
Purpose of changes:
. Fix HOMME dycore to supply the new "commid" argument in the repro_sum
calls.
. Changes to FV filtering default for high resolution grids (1/2 deg and
finer).
. Modify logfile output to include lists of all history fields in the
master and the active lists, including long_name.
. Remove the echo of the GHG and Solar forcing data to the IC file (it's
still in the history files). It's never read from the IC file, so it's
confusing to have it there.
. Modify waccm_2000_cam4 use case with updated IC files.
. Makefile updates for the BGP and the Linux/Intel builds.
. Updates to COSP simulator code:
- fix mpi bug
- output cosp coordinate values in variables rather than as attributes
- output cosp coordinate bounds in CF compliant variables rather than as
attributes
- mods to save COSP CFMIP outputs with different frequencies on separate
history files.
. Modify pretag tests for HOMME on bluefire
Bugs fixed (include bugzilla ID):
. fix mpi bug in cospsimulator (missing mpibcast call for namelist vars)
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: not checked
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do:
models/atm/cam/test/system/config_files/h16c4aqdm
. new config: homme, ne16np4, cam4, aquaplanet, debug, pure mpi
List all existing files that have been modified, and describe the changes:
models/atm/cam/SVN_EXTERNAL_DIRECTORIES
. update cosp externals to cosp_v1_3_100813
models/atm/cam/bld/Makefile.in
. add an include path for the BGP build
. add special rules to the ifort section for compiling 2 rrtmg modules
and shr_scam_mod.F90 without optimization
models/atm/cam/bld/build-namelist
. remove the manual setting of the ape attribute (comes automatically from
the cache file now that setting -ocn aquaplanet is a build time setting)
. add default for filtcw (since it now depends on resolution)
models/atm/cam/bld/namelist_files/use_cases/waccm_2000_cam4.xml
. update initial files
models/atm/cam/bld/configure
. remove unused -notest option
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. make div24del2flag=42 the default for 1/2 deg and finer FV grids
. make filtcw=1 the default for 1/4 deg and finer FV grids
. replace the attribute ape="1" by ocn="aquaplanet" which comes
automatically from the config_cache.xml file
models/atm/cam/bld/namelist_files/namelist_definition.xml
. fix documentation of fft_flt
. prepend "cosp_" to all cosp namelist variables
models/atm/cam/bld/namelist_files/use_cases/waccm_2000_cam4.xml
. update IC files
models/atm/cam/src/control/cam_history.F90
. modify output for master and active field lists
. mods to remove GHG/Solar forcing data from IC file.
. mods to cosp vertical coordinates for CF conformance
models/atm/cam/src/control/camsrfexch_types.F90
. fix comment on snowhice to indicate that it is actual snow height, not
the liquid water equivalent.
models/atm/cam/src/control/runtime_opts.F90
. add comment about physconst_readnl
models/atm/cam/src/dynamics/homme/external/global_norms_mod.F90
models/atm/cam/src/dynamics/homme/external/prim_driver_mod.F90
. supply the optional arg "commid" in the repro_sum calls (it's really only
optional if the correct communicator is MPI_COMM_WORLD).
models/atm/cam/src/physics/cam/cosp_share.F90
. add new 2D bounds variables for CF conforming output
models/atm/cam/src/physics/cam/cospsimulator_intr.F90
. add missing mpibcast calls for namelist variables
. prepend "cosp_" to all cosp namelist variables
. modify output for CFMIP experiments
models/atm/cam/test/system/config_files/h16c4aqdm
. homme, ne16np4, cam4, aquaplanet, debug, pure mpi
models/atm/cam/test/input_tests_master
. add sm733, er733, bl733 to test h16c4aqdm, outfrq3s
models/atm/cam/test/system/nl_files/aqpgro
models/atm/cam/test/system/nl_files/aqua
. remove the aqua_planet and aqua_planet_sst settings -- they are redundant
since aquaplanet is now a build time setting.
models/atm/cam/test/tests_pretag_bluefire
. add sm733, er733, bl733
. remove er731, bl731
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
061 sm731 TSM.sh h16x4c3dh aqua 9s ........................................FAIL! rc= 8 at Thu Aug 19 20:44:26 MDT 2010
064 bl733 TBL.sh h16c4aqdm outfrq3s 9s ....................................FAIL! rc= 5 at Thu Aug 19 20:44:27 MDT 2010
The failed test sm731 of HOMME in hybrid mode appears to be a system
problem. It is only when the executable is built with threads enabled that
the failure occurs. The error message points to the following location:
Illegal instruction (privileged operation) in cam_history.__cam_history_NMOD__&&_cam_history.__cam_history_NMOD_h_define at line 3350 in file "/ptmp/eaton/src/cam5_0_12_trb/models/atm/cam/src/control/cam_history.F90" ($t1)
3350 if(restart) then
I'm guessing that the illegal instruction message is a result of the
variable "restart" being corrupted. This configuration runs successfully
under PGI on edinburgh. On bluefire the configuration runs successfully if
the debug flags are removed. It also runs successfully with debug in
serial and pure mpi modes. The code fails in a pure OMP mode in with an
identical error message as above. Running a debug/OMP executable in
totalview with memoryscape enabled is successful, apparently due to
totalview's control of memory management. When run in totalview without
memory tracing enabled the reported error is the same as above. The
reported error is not in a threaded code region. I added a pure mpi
version of the test so that HOMME would continue to be tested, and left
just the smoke test version of the debug/hybrid configuration that's
failing to keep this problem on the radar.
edinburgh/lf95: All PASS
edinburgh/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_0_12
Originator(s): jenkay, eaton
Date: Tue Aug 10 14:02:00 MDT 2010
One-line Summary: Add COSP simulator.
Purpose of changes:
. add ability to produce diagnostic output from the COSP simulator as an
option. The option is enabled by adding the argument "-cosp" to the
configure commandline.
. The Makefile.in template was refactored to facilitate passing the Fortran
compiler flags to a sub-Make. The COSP simulator is built as a separate
library because it needs to be compiled using autopromotion flags.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system:
. add -cosp argument to configure to enable cosp simulator
Describe any changes made to the namelist:
. New variables for COSP simulator:
docosp:
If true, the COSP cloud simulator is run.
COSP will not run unless this is set to .true. in the namelist!
Turn on the desired simulators using lXXX_sim namelist vars
If no specific simulators are specified, all of the simulators
are run on all columns and all output is saved. (useful for testing).
This option available with CAM4 physics only.
This default logical is set in cospsimulator_intr.F90.
Default: FALSE
lradar_sim:
If true, COSP radar simulator will be run and all output
will be saved.
Default: FALSE
llidar_sim:
If true, COSP lidar simulator will be run and all output
will be saved
Default: FALSE
lisccp_sim:
If true, COSP ISCCP simulator will be run and all output
will be saved. ISCCP simulator is run on only daylight
columns.
Default: FALSE
lmisr_sim:
If true, MISR simulator will be run and all output
will be saved. MISR simulator is run on only daylight
columns.
Default: FALSE
lmodis_sim:
If true, MODIS simulator will be run and all output
will be saved.
Default: FALSE
cfmip_3hr:
If true, the COSP cloud simulator is run for CFMIP 3-hourly
experiments.
This default logical is set in cospsimulator_intr.F90
Default: FALSE
cfmip_da:
If true, the COSP cloud simulator is run for CFMIP daily
experiments.
This default logical is set in cospsimulator_intr.F90
Default: FALSE
cfmip_off:
If true, the COSP cloud simulator is run for CFMIP off-line
monthly experiments.
This default logical is set in cospsimulator_intr.F90
Default: FALSE
cfmip_mon:
If true, the COSP cloud simulator is run for CFMIP monthly
experiments.
This default logical is set in cospsimulator_intr.F90
Default: FALSE
ncolumns:
Number of subcolumns in SCOPS
This default logical is set in cospsimulator_intr.F90
Default: 50
sample_atrain:
Turns on sampling along a-train orbit for radar and lidar simulators.
This default logical is set in cospsimulator_intr.F90
Default: TRUE
atrainorbitdata:
Path for the Atrain orbit data file provided by CFMIP.
There is no default for this, but sample_atrain = .true. will not work
if this namelist variable is undefined.
currently /project/cms/jenkay/SCAM/repository/cloudsat_orbit_08921_14250.nc
Default: no default set now, need to specify in namelist
lfrac_out:
Turns on sub-column output from COSP.
If both the isccp/misr simulators and the lidar/radar simulators
are run, lfrac_out is from the isccp/misr simulators columns.
This default logical is set in cospsimulator_intr.F90
Default: FALSE
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes:
. the simulator is not on by default
. performance and memory implications of the simulator in operation have
not been evaluated
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do:
models/atm/cam/src/physics/cam/cosp_share.F90
models/atm/cam/src/physics/cam/cospsimulator_intr.F90
. CAM specific interfaces to simulator code
models/atm/cam/src/physics/cosp/*
models/atm/cam/src/physics/icarus-scops/*
. simulator code is linked in as an svn external
List all existing files that have been modified, and describe the changes:
models/atm/cam/SVN_EXTERNAL_DIRECTORIES
. link cosp source via externals
models/atm/cam/bld/Makefile.in
. add rules to build cosp library
. refactor how FFLAGS are set to facilitate passing the fortran compiler
flags to the sub-Make used to build cosp
models/atm/cam/bld/build-namelist
. set docosp=.true. when CAM is built with the cosp simulator
models/atm/cam/bld/config_files/definition.xml
models/atm/cam/bld/configure
. add -cosp argument to enable cosp simulator
models/atm/cam/bld/namelist_files/namelist_definition.xml
. add namelist variables to control cosp simulator
models/atm/cam/src/control/cam_history.F90
. add output for cosp simulator -- the main issue is all the new level
dimensions and coordinate variables
. remove the call to check_accum from subroutine hbuf_accum_inst. This
subroutine checks that either all values in a column are fill values, or
none are. This breaks when the vertical dimension is actually a
collection of subcolumns (stacked on top of one another) some of which
contain fill values and others which don't. An assumption of the
accumulator for time averaging is that a column (the vertical dimension)
is all fill value or no fill value. These hybrid vertical dimensions
break that assumption. As a consequence they should only be output as
instantaneous fields.
. Don't call the h_normalize subroutine for instantaneous fields. For
fields with fill values it enforces the constraint that either all values
in the vertical dimension are the fill value, or none are. This is not
true for cosp output. It should be up to the diagnostic code to enforce
this constraint where it is appropriate, not the generic h_normalize
code.
models/atm/cam/src/control/runtime_opts.F90
. call cospsimulator_intr_readnl method
models/atm/cam/src/physics/cam/cldwat.F90
. add diagnostic fields needed by cosp simulator as intent(out) args
models/atm/cam/src/physics/cam/radiation.F90
. add calls to init and run the simulator
models/atm/cam/src/physics/cam/convect_shallow.F90
models/atm/cam/src/physics/cam/stratiform.F90
models/atm/cam/src/physics/cam/zm_conv_intr.F90
. add fields to physics buffer for driving cosp simulator
models/atm/cam/src/utils/cam_pio_utils.F90
. add flags for new level coordinates for cosp output
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS
edinburgh/lf95: All PASS
edinburgh/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_0_11
Originator(s): Patrick Worley / Art Mirin
Date: Sun Aug 8 2010
One-line Summary: FV memory optimization; repro_sum update; spmd_utils and mod_comm communication protocol updates.
Purpose of changes:
. significant decrease in memory requirements in FV mod_comm
layer (Mirin)
. modification to implementation of handshaking communication
logic in spmd_utils and mod_comm to improve robustness.
Problems have not been observed with previous implementation,
and these changes are simply to avoid potential future problems.
. modification to repro_sum logic, improving accuracy in extreme
cases, enabling bit-for-bit agreement whether using 4-byte or
8-byte integers in implementation, and changing CPP token used
to determine which length integers to use
. modification to repro_sum implementation to remove
dependencies in mpishorthand, spmd_utils, cam_iulog,
and abortutils, as first step to moving repro_sum_mod to
models/csm_share or models/utils (for use by other components)
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
. Changed CPP token that forces use of 4-byte integers in repro_sum
algorithm from "BGP" to "noI8".
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
. significant decrease in FV memory requirements, no longer
allocating memory for undecomposed 3D field (only undecomposed
2D field) when modcomm_gatscat /= 0
Code reviewed by: Worley
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
. removal of dependence on spmd_utils and cam_iulog required new
formal parameters for specifying log unit number and "masterproc" in
repro_sum_setopts routine
control/runtime_opts.F90
utils/repro_sum_mod.F90
. removal of dependence on cam_iulog required new formal parameter
for specifying log unit number in repro_sum_tol_exceeded routine
utils/repro_sum_mod.F90
dynamics/fv/d2a3dikj.F90
dynamics/fv/fv_prints.F90
dynamics/fv/mean_module.F90
dynamics/fv/p_d_adjust.F90
dynamics/fv/par_xsum.F90
physics/cam/phys_gmean.F90
. removal of dependence on spmd_utils required addition of exisiting
optional commid parameter in calls to repro_sum
physics/cam/polar_avg.F90
. reimplementation of communication handshaking logic so that each
handshaking message is received into a different memory location
(rather than all going to/from the same location); also
add nonblocking handshaking message logic to mp_swaptrirr.
utils/pilgrim/mod_comm.F90
utils/spmd_utils.F90
. new CCP token to indicate use of 4-byte integers instead of
8-byte integers (original workaround for compiler bug on BG/P -
no longer necessary?): "noI8" replacing "BGP"; modified algorithm
termination logic so that get full accuracy and 4-byte integer
and 8-byte integers options now
utils/repro_sum_mod.F90
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: none
edinburgh/lf95: none
edinburgh/pgi: none
CAM tag used for the baseline comparison tests if different than previous
tag: cam5_0_10
Summarize any changes to answers: NONE
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_0_10
Originator(s): Diane Feddema
Date: 4, Aug, 2010
One-line Summary:Fixed CAM memory leak with work-around for xlf90 compiler bad code generation bug(IBM PMR 21571.003.000)
Purpose of changes: Fixed memory leak in CAM by adding a work-around to CAM for a bad code generation bug in IBM's xlf90 comipler.
Bugs fixed (include bugzilla ID):
. Fix memory leak in CAM caused by IBM xlf90 compiler error which does not free memory for an automatic array (which is the return value for an called function) before the calling function exits. The calling function is pbuf_get_fld_idx. When the calling function is exited memory for the automatic array is not freed - and is never freed. Diane submitted a failing testcase to IBM and they filed IBM internal bug report PMR 21571.003.000.
. Fixed misplaced t_startf call in aircraft_emit.F90 which results in mismatched timer calls when aircraft_emit.F90 results in mismatched timer calls when aircraft emission code isn't active. (Also removed embedded space from teh event tag).
describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: memory leak fixed
Code reviewed by: Brian Eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
./models/atm/cam/src/physics/cam/phys_buffer.F90
Modified routine pbuf_get_fld_idx so that it is case sensitive. This modication removes a check (call to to_upper from inside an if-block that contains a return statement within the if-block) that caused a memory leak, due to a bug in the IBM xlf90 compiler (see IBM bug report PMR 21571.003.000).
./models/atm/cam/src/physics/cam/stratiform.F90
Made pbuf_get_fld_idx calls case sensitive by using arguments with the same case for field name that was used to create the pbuf entry.
ifld = pbuf_get_fld_idx('KVH') was changed to ifld = pbuf_get_fld_idx('kvh')
ifld = pbuf_get_fld_idx('TKE') was changed to ifld = pbuf_get_fld_idx('tke')
ifld = pbuf_get_fld_idx('TURBTYPE') was changed to ifld = pbuf_get_fld_idx('turbtype')
ifld = pbuf_get_fld_idx('SMAW') was changed to ifld = pbuf_get_fld_idx('smaw')
./models/atm/cam/src/physics/cam/tphysac.F90
Made pbuf_get_fld_idx call case sensitive by using argument with the same case for field name that was used to create the pbuf entry.
ifld = pbuf_get_fld_idx('KVH') was chaned to ifld = pbuf_get_fld_idx('kvh')
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: none
edinburgh/lf95: none
edinburgh/pgi : none
CAM tag used for the baseline comparison tests if different than previous
tag: cam5_0_09
Summarize any changes to answers, i.e., NONE
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_0_09
Originator(s): Cheryl Craig and Jack Chen
Date: 27 July, 2010
One-line Summary: Added the ability to read in FAA aircraft files
Added latitude weighted interpolation and conserve column capabilities
Purpose of changes: To allow FAA aircraft data to be used in CESM
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: Created aircraft_emit_nl: contains aircraft_spcecifier and aircraft_type
aircraft_specifier contains a list of "aircraft species -> filename of file list for that species"
aircraft_type contains "CYCLICAL_LIST" which says the file list should be cycled
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: Brian Eaton, Francis Vitt
List all subroutines eliminated: none
List all subroutines added and what they do:
A models/atm/cam/src/chemistry/utils/horizontal_interpolate.F90: NEW MODULE
! This module computes and uses weighting functions to map a variable of (im1,jm1) resolution to (im2,jm2) resolution
! weight_x(im2,im1) is the weighting function for zonal interpolation
! weight_y(jm2,jm1) is the weighting function for meridional interpolation
models/atm/cam/src/chemistry/utils/tracer_data.F90
- advance_file: Advances to the next file
- vert_interp_mixrat: performs vertical interpolation of mixing ratios
models/atm/cam/src/chemsitry/utils/aircraft_emit.F90: NEW MODULE
- aircraft_emit_register: adds the aircraft emission data to the physics buffer
- aircraft_emit_init: Initialize the aircraft emission data handling
- aircraft_emit_adv: Advance to the next aircraft data
- aircraft_emit_readnl: Reads the aircraft emission namelist
- function get-aircraft_ndx: returns the index of the aircraft data being requested
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- added aircraft_specifier and aircraft_type
M models/atm/cam/src/physics/cam/initindx.F90
- added call to new routine aircraft_emit_register
M models/atm/cam/src/physics/cam/physpkg.F90
- added call to new routine aircraft_emit_init
M models/atm/cam/src/physics/cam/advnce.F90
- added call to new aircraft_emit_adv
M models/atm/cam/src/chemistry/utils/tracer_data.F90
- added the following new basic functionalities
- stepTime: time can "step" (no interpolation is done in time)
- CYCLICAL_LIST: added capability of cycling of filelists
- weight_by_lat: if true, then performs interpolations weighted by lat
- conserve_column: if tue, then interpolations preserve the total column
- introduced new routine advance_file: Copy of section in check_files which is
now called in two places.
M models/atm/cam/src/control/runtime_opts.F90
- added call to aircraft_emit_readnl
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: none
calgary/lf95:
calgary/pgi or jaguar/pgi:
CAM tag used for the baseline comparison tests if different than previous
tag: camdev_cam3_6_23_tags/camdev55_cam3_6_72
Summarize any changes to answers, i.e., NONE
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff? NONE
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_0_08
Originator(s): Diane Feddema
Date: 26 Jul 2010
One-line Summary: Renamed two derived types: surface_state became cam_out_t, srfflx_state became cam_in_t.
Purpose of changes: To clarify the intended use of these derived types.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Brian Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
models/atm/cam/src/control/camsrfexch_types.F90
derived type name "surface_state" was changed to "cam_out_t"
derived type name "srfflx_state" was changed to "cam_in_t"
models/atm/cam/src/control/srfxfer.F90
derived type name "surface_state" was changed to "cam_out_t"
models/atm/cam/src/control/cam_restart.F90
derived type name "surface_state" was changed to "cam_out_t"
models/atm/cam/src/control/cam_comp.F90
derived type name "surface_state" was changed to "cam_out_t"
derived type name "srfflx_state" was changed to "cam_in_t"
models/atm/cam/src/cpl_mct/atm_comp_mct.F90
derived type name "surface_state" was changed to "cam_out_t"
derived type name "srfflx_state" was changed to "cam_in_t"
models/atm/cam/src/physics/cam/advnce.F90
derived type name "surface_state" was changed to "cam_out_t"
models/atm/cam/src/physics/cam/iop_surf.F90
derived type name "srfflx_state" was changed to "cam_in_t"
models/atm/cam/src/physics/cam/phys_debug.F90
derived type name "srfflx_state" was changed to "cam_in_t"
models/atm/cam/src/physics/cam/radiation.F90
derived type name "surface_state" was changed to "cam_out_t"
derived type name "srfflx_state" was changed to "cam_in_t"
models/atm/cam/src/physics/cam/restart_physics.F90
derived type name "surface_state" was changed to "cam_out_t"
models/atm/cam/src/physics/cam/cam_diagnostics.F90
derived type name "surface_state" was changed to "cam_out_t"
derived type name "srfflx_state" was changed to "cam_in_t"
models/atm/cam/src/physics/cam/chemistry.F90
derived type name "surface_state" was changed to "cam_out_t"
derived type name "srfflx_state" was changed to "cam_in_t"
models/atm/cam/src/physics/cam/flux_avg.F90
derived type name "srfflx_state" was changed to "cam_in_t"
models/atm/cam/src/physics/cam/physpkg.F90
derived type name "surface_state" was changed to "cam_out_t"
derived type name "srfflx_state" was changed to "cam_in_t"
models/atm/cam/src/physics/cam/tphysbc.F90
derived type name "surface_state" was changed to "cam_out_t"
derived type name "srfflx_state" was changed to "cam_in_t"
models/atm/cam/src/physics/cam/tphysac.F90
derived type name "surface_state" was changed to "cam_out_t"
derived type name "srfflx_state" was changed to "cam_in_t"
models/atm/cam/src/physics/rrtmg/radiation.F90
derived type name "surface_state" was changed to "cam_out_t"
derived type name "srfflx_state" was changed to "cam_in_t"
models/atm/cam/src/chemistry/utils/aerodep_flx.F90
derived type name "surface_state" was changed to "cam_out_t"
models/atm/cam/src/chemistry/modal_aero/modal_aero_deposition.F90
derived type name "surface_state" was changed to "cam_out_t"
models/atm/cam/src/chemistry/bulk_aero/dust_intr.F90
derived type name "surface_state" was changed to "cam_out_t"
models/atm/cam/src/chemistry/bulk_aero/aerosol_intr.F90
derived type name "surface_state" was changed to "cam_out_t"
derived type name "srfflx_state" was changed to "cam_in_t"
models/atm/cam/src/chemistry/mozart/chemistry.F90
derived type name "surface_state" was changed to "cam_out_t"
derived type name "srfflx_state" was changed to "cam_in_t"
models/atm/cam/src/chemistry/mozart/mo_drydep.F90
derived type name "srfflx_state" was changed to "cam_in_t"
models/atm/cam/src/chemistry/mozart/mo_lightning.F90
derived type name "srfflx_state" was changed to "cam_in_t"
models/atm/cam/src/chemistry/mozart/mz_aerosols_intr.F90
derived type name "surface_state" was changed to "cam_out_t"
models/atm/cam/src/chemistry/mozart/spedata.F90
comment line changed ("srfflx_state" was changed to "cam_in" in comment)
models/atm/cam/src/dynamics/eul/stepon.F90
derived type name "surface_state" was changed to "cam_out_t"
models/atm/cam/src/dynamics/homme/stepon.F90
derived type name "surface_state" was changed to "cam_out_t"
models/atm/cam/src/dynamics/fv/metdata.F90
derived type name "srfflx_state" was changed to "cam_in_t"
models/atm/cam/src/dynamics/fv/stepon.F90
derived type name "surface_state" was changed to "cam_out_t"
models/atm/cam/src/dynamics/fv/fv_prints.F90
derived type name "surface_state" was changed to "cam_out_t"
models/atm/cam/src/cpl_esmf/atm_comp_esmf.F90
derived type name "surface_state" was changed to "cam_out_t"
derived type name "srfflx_state" was changed to "cam_in_t"
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: all pass
edinburgh/lf95: all pass
edinburgh/pgi or jaguar/pgi: all pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_0_07
Originator(s): Francis Vitt
Date: 21 Jul 2010
One-line Summary: Bug fixes to FV dycore offline driver and the chemistry customization option in CAM configure.
Purpose of changes:
Fix a bug in the FV dycore offline driver that was introduced when the
metdata module used by the FV dycore offline driver was modified to use
parallel IO utilities.
Fix a bug in the CAM configure utility to allow the user to customize
the trop_bam, trop_ghg, and waccm_ghg chemistry packages via the "-usr_mech_infile"
configure option.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Brian Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/config_files/fmo1.9c4dh
Use the new -offline_dyn configure option
M models/atm/cam/bld/configure
M models/atm/cam/bld/config_files/definition.xml
Added "-offline_dyn" option
M models/atm/cam/bld/config_files/defaults_trop_bam.xml
M models/atm/cam/bld/config_files/defaults_trop_ghg.xml
M models/atm/cam/bld/config_files/defaults_waccm_ghg.xml
Removed the "prog_species" specification to allow the user to customize
the chemical mechanism. The prog_species specification is not needed
here since these chemistry packages are pre-defined, i.e., the preprocessor
generated source code for the chem packages are already included in the
standard source code distribution.
M models/atm/cam/bld/build-namelist
Adjustments needed for trop_bam, trop_ghg, and waccm_ghg since prog_species
is not set for these chem packages.
Included the ver=>'fixed_ox' specification to tracer_cnst* default namelist
settings.
M models/atm/cam/src/dynamics/fv/metdata.F90
Corrected problems with the reading of prescribed meteorology fields that
were introduced when this module was modified to use parallel IO utilities
in cam4_1_12.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: all pass
edinburgh/lf95: all pass
edinburgh/pgi: all pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam5_0_06
Originator(s): andrew, eaton
Date: Mon Jul 19 19:10:10 MDT 2010
One-line Summary: separate aerosol activation code from microphysics code
Purpose of changes:
. Separate the aerosol activation code from the microphysics code.
. Change waccm_ghg test to use cam4 physics.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: not checked
Code reviewed by: eaton
List all subroutines eliminated:
models/atm/cam/test/system/config_files/wg10c5dm
. this file was moved to wg10c4dm
List all subroutines added and what they do:
models/atm/cam/src/physics/cam/microp_aero.F90
. aerosol activation code previously in cldwat2m_micro.F90
models/atm/cam/test/system/config_files/wg10c4dm
. was wg10c4dm -- change cam5 to cam4
List all existing files that have been modified, and describe the changes:
models/atm/cam/src/physics/cam/cldwat2m_micro.F90
. Move aerosol activation code to new microp_aero module.
models/atm/cam/src/physics/cam/stratiform.F90
. add calls to microp_aero subroutines
models/atm/cam/test/system/input_tests_master
. wg10c5dm changed to wg10c4dm
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Wed Jul 14 19:14:39 MDT 2010
021 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...............................FAIL! rc= 7 at Wed Jul 14 19:40:24 MDT 2010
036 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ......................................FAIL! rc= 7 at Wed Jul 14 20:20:16 MDT 2010
040 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .....................................FAIL! rc= 7 at Wed Jul 14 20:40:42 MDT 2010
047 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...........................FAIL! rc= 7 at Wed Jul 14 21:31:32 MDT 2010
These baseline failures are due to roundoff changes associated with the
code refactoring.
edinburgh/lf95:
034 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 5 at Mon Jul 19 18:45:06 MDT 2010
edinburgh/pgi:
034 bl314 TBL.sh wg10c4dm outfrq3s 9s .....................................FAIL! rc= 5 at Mon Jul 19 18:26:03 MDT 2010
bl314 fails because it is a new test -- no baseline available for comparison.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: cam4 is BFB. cam5 has roundoff changes
due to the code refactoring. Andrew Gettelman verified the roundoff
changes.
===============================================================
===============================================================
Tag name: cam5_0_05
Originator(s): olson
Date: Fri Jun 30 2010
One-line Summary: added capability to specify which history file to place heat/moisture budget terms
Purpose of changes:
Added the namelist variable "history_budget_histfile_num" to specify which
history file to place the heat/moisture budget terms (and instantaneous state fields).
Previously the only option was the primary history file. When "history_budget" is .false.,
"history_budget_histfile_num" is ignored. Default is "1" (primary).
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: added "history_budget_histfile_num"
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/namelist_files/namelist_definition.xml
M models/atm/cam/src/physics/cam/phys_control.F90
- added "history_budget_histfile_num" to namelist and control infrastructure
M models/atm/cam/src/dynamics/eul/dyn_comp.F90
M models/atm/cam/src/dynamics/fv/dyn_comp.F90
M models/atm/cam/src/physics/cam/convect_shallow.F90
M models/atm/cam/src/physics/cam/gw_drag.F90
M models/atm/cam/src/physics/cam/radiation.F90
M models/atm/cam/src/physics/cam/vertical_diffusion.F90
M models/atm/cam/src/physics/cam/zm_conv_intr.F90
M models/atm/cam/src/physics/rrtmg/radiation.F90
- replaced the primary history file designation ("1") with "history_budget_histfile_num"
in the call to "add_default"
M models/atm/cam/src/physics/cam/cam_diagnostics.F90
M models/atm/cam/src/physics/cam/stratiform.F90
- replaced the primary history file designation ("1") with "history_budget_histfile_num"
in the call to "add_default". Also, made the state fields instantaneous for that
history file when "history_budget" is .true.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS
edinburgh/lf95: All PASS
edinburgh/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_0_04
Originator(s): mvr, jedwards
Date: Fri Jun 25 2010
One-line Summary: various fixes made to test scripts; pnetcdf bug fix and
io cleanup from release branch
Purpose of changes:
The test scripts place an explicit call to generate_cice_decomp.pl (the cice
decomposition script) so that cice can be built in a manner that allows reuse
of the resulting binary during restart tests, where the number of mpi tasks
(and possible smp threads) are changed. Until now, this script was not being
called with the proper inputs, but still managed to provide an executable
that could be reused for restart tests.
However, we discovered that this was the cause of a failed test in the jaguar
posttag testing, where the pgi compiler had a problem with zero-length arrays
being created as a by-product. To fix that problem, changes were made to properly
call the decomp script and make the necessary changes to the rest of the test
scripts to deal with some of its constraints. For example, testing on bluefire
was moved to 48 processors (under-utilizing one bluefire node) because many of its
tests are done at fv 4x5, which does not have a cartesian solution with 64 procs.
This is viewed as a short-term solution. Long-term we intend to remove the call
to generate_cice_decomp.pl from the test scripts and have cam's configure figure
out the parameters to send to the cice build.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
D models/atm/cam/test/system/config_files/fma1.9c5m
D models/atm/cam/test/system/config_files/fmgpa1.9c5dm
D models/atm/cam/test/system/config_files/fmo1.9c5dh
- config options test files using cam5 were replaced with those using cam4
D models/atm/cam/test/system/nl_files/off2x2.5
- removed obsolete test file
List all subroutines added and what they do:
A models/atm/cam/test/system/config_files/fmgpa1.9c4dm
A models/atm/cam/test/system/config_files/fma1.9c4m
A models/atm/cam/test/system/config_files/fmo1.9c4dh
- config options test files using cam5 were replaced with those using cam4
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/TCB.sh
- modification of the call to generate_cice_decomp.pl
M models/atm/cam/test/system/test_driver.sh
- bluefire testing now done on 48 processors instead of 64; no longer
using 'module purge' prior to jaguar testing
M models/atm/cam/test/system/nl_files/off1.9x2.5
- namelist options test file now will use inputfile with 26 levels (needed for cam4 testing)
M models/atm/cam/test/system/nl_files/fvvp_lb2
- namelist options file modified to fit into fewer processors on bluefire
M models/atm/cam/test/system/input_tests_master
- changes to test definitions to reflect some tests moving from cam5 to cam4
M models/atm/cam/bld/run-pc.csh
M models/atm/cam/bld/run-ibm.csh
M models/atm/cam/bld/run-cray.csh
- run template scripts now execute build-namelist from rundir rather than blddir
M models/atm/cam/src/utils/cam_pio_utils.F90
- bug fix for pnetcdf (from jedwards)
M models/atm/cam/src/utils/time_manager.F90
M models/atm/cam/src/utils/cam_csim4/ice_filenames.F90
M models/atm/cam/src/utils/cam_csim4/cpl_mct/ice_comp_mct.F90
M models/atm/cam/src/utils/cam_csim4/ice_data.F90
M models/atm/cam/src/utils/cam_csim4/ice_time_manager.F90
M models/atm/cam/src/utils/cam_csim4/ice_srf.F90
M models/atm/cam/src/utils/cam_csim4/ice_comp.F90
M models/atm/cam/src/physics/cam/rayleigh_friction.F90
M models/atm/cam/src/physics/cam/phys_gmean.F90
M models/atm/cam/src/physics/cam/modal_aer_opt.F90
M models/atm/cam/src/physics/cam/ndrop.F90
M models/atm/cam/src/physics/cam/convect_shallow.F90
M models/atm/cam/src/physics/cam/cldwat2m_macro.F90
M models/atm/cam/src/physics/cam/uwshcu.F90
M models/atm/cam/src/physics/cam/vertical_diffusion.F90
M models/atm/cam/src/physics/rrtmg/cloud_rad_props.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_wateruptake.F90
M models/atm/cam/src/chemistry/bulk_aero/progseasalts_intr.F90
M models/atm/cam/src/chemistry/mozart/lin_strat_chem.F90
M models/atm/cam/src/chemistry/mozart/wei96.F90
M models/atm/cam/src/dynamics/sld/restart_dynamics.F90
M models/atm/cam/src/dynamics/eul/restart_dynamics.F90
M models/atm/cam/src/dynamics/fv/dyn_comp.F90
M models/atm/cam/src/dynamics/fv/restart_dynamics.F90
- io cleanup pulled in from release branch
M models/atm/cam/src/dynamics/homme/restart_dynamics.F90
- bug fix for homme restart on 48 processors (from jedwards)
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS
edinburgh/lf95: All PASS
edinburgh/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_0_03
Originator(s): eaton
Date: Mon Jun 21 09:08:32 MDT 2010
One-line Summary: code workaround for bluefire threading problem; update externals
Purpose of changes:
. A system problem on bluefire has been causing threaded CAM runs to either
crash badly (bringing down the node) when running cam5, or just seg fault
when running with cam4. A code modification has been made in the
tropopause module which allows cam4 to run without seg faulting, but does
not fix the problem running cam5. A firmware fix for bluefire will be
applied shortly, and at that time we may back out this workaround.
. Update to cesm1_0_beta05 externals.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets:
. add default jan 01 IC for 1 deg cam5
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
SVN_EXTERNAL_DIRECTORIES
. drvseq3_1_30 -> drvseq3_1_32
clm4_0_08 -> clm4_0_09
cice4_0_20100607 -> cice4_0_20100611
share3_100607 -> share3_100616
scripts4_100608 -> scripts4_100617
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. add cami-mam3_0000-01-01_0.9x1.25_L30_c100618.nc
models/atm/cam/src/physics/cam/tropopause.F90
. Make the variables kap, faktor, and ka1 module variables, and remove the
local declarations in routine twmo.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS
edinburgh/lf95: All PASS
edinburgh/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_0_02
Originator(s): aconley, fvitt
Date: 16 June 2010
One-line Summary: RRTMG source code clean up.
Purpose of changes:
Rename the "conley" liquid cloud optics method to "gammadist"
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
RRTMG liqcloudoptics method "conley" changed to "gammadist"
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Brian Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
U models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
U models/atm/cam/bld/namelist_files/namelist_definition.xml
U models/atm/cam/src/physics/rrtmg/radiation.F90
- Changed "conley" liquid cloud optics method to "gammadist"
U models/atm/cam/src/physics/rrtmg/ebert_curry.F90
U models/atm/cam/src/physics/rrtmg/slingo.F90
- Fixed bug for wavelengths greater the 2.38 micrometers.
These methods are not used in the default configurations so this
bug fix had no effect on the base-line regression tests.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: all PASS
edinburgh/lf95: all PASS
edinburgh/pgi: all PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, BFB
===============================================================
===============================================================
Tag name: cam5_0_01
Originator(s): eaton
Date: Tue Jun 15 08:21:45 MDT 2010
One-line Summary: back out change to restart file
Purpose of changes:
. The change to the restart file made in cam4_9_11 which moved the variable
rst_calendar to a global attribute has been backed out. The change is
not backwards compatible with older restart files which prevents the
model from doing branch runs from older ccsm4 runs.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/src/utils/time_manager.F90
. revert rst_calendar from a global attribute back to being a variable
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: not done
edinburgh/lf95: not done
edinburgh/pgi or jaguar/pgi: not done
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam5_0_00
Originator(s): fvitt, eaton
Date: Sun Jun 13 09:30:22 MDT 2010
One-line Summary: branch point for CAM5 release; hist output mods
Purpose of changes:
. The CAM5 release branch will be made off of this tag.
. Modify addfld/add_default calls so that fields relevent to cam5 physics
don't show up by default when using cam4 physics.
. Updates to chemistry namelist documentation.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/namelist_files/namelist_definition.xml
. update documentation
models/atm/cam/src/chemistry/mozart/lin_strat_chem.F90
. don't execute init method unless needed
models/atm/cam/src/physics/cam/ndrop.F90
. add the addfld/add_default calls for
WTKE,LCLOUD,NDROPMIX,NDROPSRC,NDROPSNK,NDROPCOL
models/atm/cam/src/physics/cam/stratiform.F90
. remove addfld/add_default calls for
WTKE,LCLOUD,NDROPMIX,NDROPSRC,NLSED,NDROPSNK,NDROPCOL
- there was no outfld call for NLSED, the other diags belong in ndrop.
models/atm/cam/src/physics/waccm/qbo.F90
. remove extraneous log messages
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS
edinburgh/lf95: All PASS
edinburgh/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam4_9_12
Originator(s): eaton et al.
Date: Sat Jun 12 12:25:34 MDT 2010
One-line Summary: update externals; build-namelist mod for cam5; change
adiabatic mode dry mass; fix cam5 budget output; cleanup for release
Purpose of changes:
. update externals to be consistent with cesm1_0_beta03
. modify build-namelist to call the cice build-namelist with cam5=.true. for
cam5 runs. This enables the cice build-namelist to provide the following
cam5 specific settings: r_snw=1.75, dt_mlt_in=1.0, rsnw_melt_in=1000.
Note that these changes were already in place for runs using the ccsm
scripts.
. updates to namelist defaults for HOMME (from Mark Taylor)
. new run script template for scam (from John Truesdale)
. updates to aquaplanet use case (from John Truesdale)
. change how the dry atmosphere mass is fixed in adiabatic mode (from
Christianne Jablonowski)
. cleanup comments in stratiform (from Andrew Gettelman)
. add missing CAM5 budget diagnostics (from Jerry Olson)
. remove unsupported KE deep convection scheme
. remove unsupported water tracer code
Bugs fixed (include bugzilla ID): none
. fix bug in creation of cam.buildexe.csh (wouldn't allow switching between
mct and esmf interfaces after the script was created) This is only
relevent to running with ccsm scripts. (from Tony Craig)
Describe any changes made to build system: none
Describe any changes made to the namelist:
. modify the atm_dep_flux namelist variable so that setting it to false
explicitly sets the deposition fluxes sent to the coupler to zero.
Previously setting it to false was supposed to signal to CLM and CICE to
read the deposition fluxes from an external file. But that functionality
has been removed from CICE, and both CLM and CICE now expect CAM to
always provide these fluxes. Explicitly setting them to zero provides a
way to turn of the climate feedbacks from prognostically determined
fluxes.
List any changes to the defaults for the boundary datasets:
. update the 20th century use cases to use the 1850-2008 AMIP SST datasets
rather than 1850 climatology. (This affects CAM standalone runs only,
not runs done using ccsm scripts.)
Describe any substantial timing or memory changes: not checked
Code reviewed by: eaton
List all subroutines eliminated:
models/atm/cam/src/physics/cam/convect_ke.F90
models/atm/cam/src/physics/cam/convect_ke_intr.F90
. remove KE deep convection
models/atm/cam/src/control/wtrc_flxoce.F90
models/atm/cam/src/physics/cam/water_isotopes.F90
models/atm/cam/src/physics/cam/water_tracers.F90
. remove water tracer code
List all subroutines added and what they do:
models/atm/cam/bld/run-scam.csh
. new run script template for scam
List all existing files that have been modified, and describe the changes:
SVN_EXTERNAL_DIRECTORIES
. clm4_0_07 -> clm4_0_08
cice4_0_20100602 -> cice4_0_20100607
scripts4_100603a -> scripts4_100608
models/atm/cam/bld/build-namelist
. add "cam5=.true." to the cice namelist arg when cam5 physics used.
. remove setting of faerdep in clm namelist
models/atm/cam/bld/cam.cpl7.template
. fix bug in creation of cam.buildexe.csh (wouldn't allow switching between
mct and esmf interfaces after the script was created)
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. update defaults for HOMME (parameters and ne240np4 IC file for jan 01)
. remove faerdep defaults since no longer asking CLM to read deposition file
models/atm/cam/bld/namelist_files/namelist_definition.xml
. remove KE as valid option for deep_scheme
. remove wisotope, tracer_water
models/atm/cam/bld/namelist_files/use_cases/aquaplanet_cam3_cam4.xml
. update tuning parameters
models/atm/cam/bld/namelist_files/use_cases/1850-2005_cam4_bgc.xml
models/atm/cam/bld/namelist_files/use_cases/1850-2005_cam4_super_fast_llnl.xml
models/atm/cam/bld/namelist_files/use_cases/1850-2005_cam4.xml
models/atm/cam/bld/namelist_files/use_cases/1850-2005_cam5.xml
models/atm/cam/bld/namelist_files/use_cases/waccm_1850-2005_cam4.xml
. update the 20th century use cases to use the 1850-2008 AMIP SST datasets
rather than 1850 climatology.
models/atm/cam/src/control/runtime_opts.F90
. remove trace_water, wisotope variables
. remove water_isotopes, water_tracers
models/atm/cam/src/cpl_esmf/atm_comp_esmf.F90
models/atm/cam/src/cpl_mct/atm_comp_mct.F90
. hardcode the setting of atm_aero=.true. rather than listening to the
atm_dep_flux variable since the ability of the cice code to read the
deposition fluxes from an external file has been removed.
models/atm/cam/src/dynamics/eul/inidat.F90
models/atm/cam/src/dynamics/sld/inidat.F90
. initialize tmass0 from the initial pressure field for adiabatic runs
models/atm/cam/src/physics/cam/cam_diagnostics.F90
. in subroutine diag_surf set the deposition fluxes to zero if
atm_dep_flux=.false.
models/atm/cam/src/physics/cam/cldwat.F90
. remove deep_scheme='KE'
models/atm/cam/src/physics/cam/cloud_fraction.F90
. remove unused deep_scheme variable
models/atm/cam/src/physics/cam/convect_deep.F90
. remove deep_scheme='KE'
models/atm/cam/src/physics/cam/initindx.F90
models/atm/cam/src/physics/cam/physpkg.F90
. remove water_tracers
models/atm/cam/src/physics/cam/stratiform.F90
. cleanup comments
. remove deep_scheme from inimc calling args.
. add diagnostics:
'DPDLFLIQ ','DPDLFICE ','SHDLFLIQ ','SHDLFICE ','DPDLFT ','SHDLFT '
. remove water_tracers
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
009 bl134 TBL.sh e48adh adia 9s ...........................................FAIL! rc= 7 at Fri Jun 11 18:10:39 MDT 2010
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Wed Jun 9 15:36:44 MDT 2010
021 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...............................FAIL! rc= 7 at Wed Jun 9 16:03:07 MDT 2010
036 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ......................................FAIL! rc= 7 at Wed Jun 9 16:39:49 MDT 2010
040 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .....................................FAIL! rc= 7 at Wed Jun 9 16:59:21 MDT 2010
045 bl373 TBL.sh fm1.9c4dh outfrq3s 9s ....................................FAIL! rc= 7 at Wed Jun 9 17:28:04 MDT 2010
047 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...........................FAIL! rc= 7 at Wed Jun 9 17:46:41 MDT 2010
050 bl374 TBL.sh fs1.9c4dh outfrq3s+1850-2005_cam4_super_fast_llnl 9s .....FAIL! rc= 7 at Wed Jun 9 18:06:49 MDT 2010
054 bl387 TBL.sh f1.9c4dm outfrq3s+1850-2005_cam4 9s ......................FAIL! rc= 7 at Wed Jun 9 18:22:54 MDT 2010
057 bl388 TBL.sh f1.9c4m outfrq24h+1850_cam4 2d ...........................FAIL! rc= 7 at Wed Jun 9 18:34:21 MDT 2010
060 bl379 TBL.sh wm1.9c4h outfrq24h+waccm_1850_cam4 2d ....................FAIL! rc= 7 at Wed Jun 9 19:14:19 MDT 2010
068 bl992 TBL_ccsm.sh f19_g16 E 2d ........................................FAIL! rc= 7 at Wed Jun 9 19:55:26 MDT 2010
edinburgh/lf95: All PASS except:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Wed Jun 9 15:13:25 MDT 2010
008 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Wed Jun 9 15:34:55 MDT 2010
014 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Wed Jun 9 15:48:02 MDT 2010
021 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Wed Jun 9 16:26:15 MDT 2010
029 bl312 TBL.sh f10c5dm ghgrmp 9s ........................................FAIL! rc= 7 at Wed Jun 9 17:02:19 MDT 2010
034 bl314 TBL.sh wg10c5dm outfrq3s 9s .....................................FAIL! rc= 4 at Wed Jun 9 17:10:32 MDT 2010
edinburgh/pgi: All PASS except:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Wed Jun 9 14:47:02 MDT 2010
008 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Wed Jun 9 14:52:42 MDT 2010
014 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Wed Jun 9 14:58:31 MDT 2010
021 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Wed Jun 9 15:08:45 MDT 2010
029 bl312 TBL.sh f10c5dm ghgrmp 9s ........................................FAIL! rc= 7 at Wed Jun 9 15:16:42 MDT 2010
034 bl314 TBL.sh wg10c5dm outfrq3s 9s .....................................FAIL! rc= 7 at Wed Jun 9 15:27:47 MDT 2010
The baseline failures are due to the following:
. The bugfix to snow hydrology in CLM changes answers in all configurations
that use CLM.
. The update to build-namelist changes answers for all configurations using
CAM5 physics and CICE due to changed parameter values.
. The update to how the dry mass is initialized in adiabatic mode causes
answer changes in all adiabatic tests.
. The update to use the AMIP SST datasets for all transient tests changes
answers in runs using the 1850-2005 use cases.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: cam is BFB, but answers change due to new
externals and boundary datasets.
===============================================================
===============================================================
Tag name: cam4_9_11
Originator(s): Jim Edwards, Art Mirin, Mark Taylor
Date: 6/9/2010
One-line Summary: Update Homme Dynamics
Purpose of changes: Development/ bug fixes
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Jim
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
models/atm/cam/doc/ChangeLog_template
removed calgary added edinburgh
models/atm/cam/test/system/input_tests_master
Renamed models/atm/cam/test/system/config_files/h16x4c3dm
to
models/atm/cam/test/system/config_files/h16x4c3dh
following the convention because OpenMP was added to the test.
models/atm/cam/test/system/test_driver.sh
updated pnetcdf version for bluefire.
models/atm/cam/bld/config_files/horiz_grid.xml
added a very low resolution homme grid
models/atm/cam/bld/namelist_files/namelist_definition.xml
added qsplit to homme ctl_nl
models/atm/cam/src/utils/time_manager.F90
made the calendar string an attribute rather than a variable
models/atm/cam/src/dynamics/sld/inidat.F90
models/atm/cam/src/dynamics/eul/inidat.F90
Fixed a bug in reading the PHIS variable when the from_hires attribute
is not defined.
models/atm/cam/src/dynamics/homme/external/filter_mod.F90
models/atm/cam/src/dynamics/homme/external/diffusion_mod.F90
models/atm/cam/src/dynamics/homme/external/mass_matrix_mod.F90
models/atm/cam/src/dynamics/homme/external/prim_advance_mod.F90
models/atm/cam/src/dynamics/homme/external/parallel_mod.F90
models/atm/cam/src/dynamics/homme/external/prim_state_mod.F90
models/atm/cam/src/dynamics/homme/external/global_norms_mod.F90
models/atm/cam/src/dynamics/homme/external/control_mod.F90
models/atm/cam/src/dynamics/homme/external/prim_driver_mod.F90
models/atm/cam/src/dynamics/homme/external/element_mod.F90
models/atm/cam/src/dynamics/homme/external/viscosity_mod.F90
models/atm/cam/src/dynamics/homme/external/prim_advection_mod.F90
models/atm/cam/src/dynamics/homme/external/edge_mod.F90
models/atm/cam/src/dynamics/homme/external/dof_mod.F90
models/atm/cam/src/dynamics/homme/external/solver_mod.F90
models/atm/cam/src/dynamics/homme/external/namelist_mod.F90
models/atm/cam/src/dynamics/homme/dp_coupling.F90
models/atm/cam/src/dynamics/homme/stepon.F90
models/atm/cam/src/dynamics/homme/inidat.F90
models/atm/cam/src/dynamics/homme/dyn_comp.F90
models/atm/cam/src/dynamics/homme/restart_dynamics.F90
models/atm/cam/src/dynamics/homme/spmd_dyn.F90
Fixed OpenMP implementation. Ported recent development from homme standalone
repository.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: all pass except
063 bl731 TBL.sh h16x4c3dh aqua 9s expected due to dycore
development and change of test from MPI to hybrid.
edinburgh/lf95: all pass
edinburgh/pgi : all pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam4_9_10
Originator(s): fvitt, eaton
Date: Tue Jun 8 08:15:34 MDT 2010
One-line Summary: add BGC and WACCM use cases; update externals
Purpose of changes:
. add use cases 1850-2005_cam4_bgc and waccm_2000_cam4
. update csm_share and driver externals
. modify run script templates to explicitly run fv, 1.9x2.5
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: not checked
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do:
models/atm/cam/bld/namelist_files/use_cases/1850-2005_cam4_bgc.xml
models/atm/cam/bld/namelist_files/use_cases/waccm_2000_cam4.xml
. new use case files
List all existing files that have been modified, and describe the changes:
SVN_EXTERNAL_DIRECTORIES
. drvseq3_1_29 -> drvseq3_1_30
. share3_100515 -> share3_100607
models/atm/cam/bld/namelist_files/namelist_definition.xml
. fix category of co2_cycle_rad_passive namelist var
models/atm/cam/bld/run-cray.csh
models/atm/cam/bld/run-ibm.csh
models/atm/cam/bld/run-pc.csh
. configure for fv, 1.9x2.5
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS
edinburgh/lf95: All PASS
edinburgh/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers; BFB
===============================================================
===============================================================
Tag name: cam4_9_09
Originator(s): hannay, gettelman, morrison, mvr
Date:
One-line Summary: finalize CAM5
Purpose of changes:
* bugfix in microphysics and major (non-answer changing) cleanup of code,
* physics changes (scale sea-salt, change in solubility factor, emissions numbers for black carbon and organic carbon, rhminl)
* Change testing of mozart : fm1.9c5 => fm1.9c4
Bugs fixed (include bugzilla ID):
Bug Fix in microphysics: calculation of pgam changed to c1 * (ncic/1e.-6 * rho) + c2 [bug was dividing by rho]
Describe any changes made to build system:
Describe any changes made to the namelist:
* Changes to output by default the MG microphysics variables for AMWG package
List any changes to the defaults for the boundary datasets:
* Changes in emissions numbers for black carbon and organic carbon
* Changes in hygroscopicity "B" of POA from 0 to 0.1
Describe any substantial timing or memory changes:
* efficiency improvements due to major non-answer changing cleanup of microphysics code
Code reviewed by: hannay, eaton
List all subroutines eliminated:
* models/atm/cam/test/system/config_files/fm1.9c5dh
* models/atm/cam/test/system/config_files/fm1.9c5m
Delete testing: fm1.9c5 (testing mozart in cam5)
List all subroutines added and what they do:
* models/atm/cam/test/system/config_files/fm1.9c4dh
* models/atm/cam/test/system/config_files/fm1.9c4m
Add testing: fm1.9c4 (testing mozart in cam4)
List all existing files that have been modified, and describe the changes:
* models/atm/cam/test/system/input_tests_master
Change testing of mozart: fm1.9c5 => fm1.9c4
* models/atm/cam/src/physics/cam/cldwat2m_micro.F90
Major non-answer changing cleanup of code, includes efficiency improvements
Bug Fix: calculation of pgam changed to c1 * (ncic/1e.-6 * rho) + c2 [bug was dividing by rho]
Sub-grid qc removed from fallspeed calculation (unc, umc)
For t > tmelt, use esl (saturation vapor pressure over liquid)(2-4 are answer changing)
* models/atm/cam/src/chemistry/bulk_aero/progseasalts_intr.F90
scale sea-salts: increase by 35%
* models/atm/cam/src/chemistry/mozart/mz_aerosols_intr.F90
change in the solubility factor: sol_factic = 0.5->0.4
* models/atm/cam/src/physics/cam/phys_control.F90
output by default the MG microphysics variables for AMWG package
* models/atm/cam/bld/build-namelist
Changes to output by default the MG microphysics variables for AMWG package
* models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
Change in rhminl: 0.905->0.89
Changes in emissions numbers for black carbon and organic carbon
Changes in hygroscopicity "B" of POA from 0 to 0.1 (atm/cam/physprops/ocpho_rrtmg_c100528.nc)
* models/atm/cam/bld/namelist_files/use_cases/1850-2005_cam5.xml
Changes in emissions numbers for black carbon and organic carbon
* models/atm/cam/bld/namelist_files/use_cases/1850_cam5.xml
Changes in emissions numbers for black carbon and organic carbon
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Mon Jun 7 11:10:45 MDT 2010
013 bl330 TBL.sh f4c5pdh aqpgro 3s ........................................FAIL! rc= 7 at Mon Jun 7 11:10:48 MDT 2010
021 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...............................FAIL! rc= 7 at Mon Jun 7 11:10:59 MDT 2010
036 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ......................................FAIL! rc= 7 at Mon Jun 7 11:11:09 MDT 2010
040 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .....................................FAIL! rc= 7 at Mon Jun 7 11:11:34 MDT 2010
047 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...........................FAIL! rc= 7 at Mon Jun 7 11:23:02 MDT 2010
calgary/lf95:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Sat Jun 5 16:09:08 MDT 2010
008 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Sat Jun 5 16:22:41 MDT 2010
010 bl113 TBL.sh e8c5pdm aqpgro 3s ........................................FAIL! rc= 7 at Sat Jun 5 16:26:37 MDT 2010
014 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Sat Jun 5 16:32:33 MDT 2010
021 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Sat Jun 5 16:59:28 MDT 2010
029 bl312 TBL.sh f10c5dm ghgrmp 9s ........................................FAIL! rc= 7 at Sat Jun 5 17:27:04 MDT 2010
031 bl313 TBL.sh f10c5pdm aqpgro 3s .......................................FAIL! rc= 7 at Sat Jun 5 17:32:17 MDT 2010
034 bl314 TBL.sh wg10c5dm outfrq3s 9s .....................................FAIL! rc= 7 at Sat Jun 5 17:32:19 MDT 2010
039 bl316 TBL.sh f10c5pdm aqpgro+aquaplanet_cam5 3s .......................FAIL! rc= 7 at Sat Jun 5 17:35:00 MDT 2010
- any baseline tests where cam5 was used were expected to fail becasue changes in the cam5 physics
calgary/pgi or jaguar/pgi:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 7 at Sat Jun 5 15:57:09 MDT 2010
008 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 7 at Sat Jun 5 16:03:12 MDT 2010
010 bl113 TBL.sh e8c5pdm aqpgro 3s ........................................FAIL! rc= 7 at Sat Jun 5 16:07:29 MDT 2010
014 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 7 at Sat Jun 5 16:09:19 MDT 2010
021 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 7 at Sat Jun 5 16:19:37 MDT 2010
029 bl312 TBL.sh f10c5dm ghgrmp 9s ........................................FAIL! rc= 7 at Sat Jun 5 16:28:00 MDT 2010
031 bl313 TBL.sh f10c5pdm aqpgro 3s .......................................FAIL! rc= 7 at Sat Jun 5 16:32:30 MDT 2010
034 bl314 TBL.sh wg10c5dm outfrq3s 9s .....................................FAIL! rc= 7 at Sat Jun 5 16:38:44 MDT 2010
039 bl316 TBL.sh f10c5pdm aqpgro+aquaplanet_cam5 3s .......................FAIL! rc= 7 at Sat Jun 5 16:43:06 MDT 2010
- any baseline tests where cam5 was used were expected to fail becasue changes in the cam5 physics
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
Tag name: cam4_9_08
Originator(s): mvr, hannay
Date: Fri Jun 4 2010
One-line Summary: updated to recent versions of external component code; Turn on
volcanic aerosols in CAM5 for the 1850-2005 use case.
Purpose of changes:
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: changes needed to implement volcanic
aerosols in cam5
List any changes to the defaults for the boundary datasets: new setting for volcanics
Describe any substantial timing or memory changes: not checked
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do:
A models/atm/cam/test/system/nl_files/volc
- new namelist options test file for testing volcanics
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/TBR.sh
M models/atm/cam/test/system/TER.sh
M models/atm/cam/test/system/TSM.sh
- added dumping of missing namelists to test log output
M models/atm/cam/test/system/input_tests_master
- change in test definition for test of cam5 volcanics
M models/atm/cam/bld/run-pc.csh
M models/atm/cam/bld/run-ibm.csh
M models/atm/cam/bld/run-cray.csh
- fix to run template scripts needed to utilize new ocn.stream.txt info
M models/atm/cam/bld/configure
- removed ntr_pond and ntr_iage settings from cice configure call
M models/atm/cam/bld/build-namelist
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
M models/atm/cam/bld/namelist_files/use_cases/1850-2005_cam5.xml
- changes required to implement volcanic aerosols in cam5
M SVN_EXTERNAL_DIRECTORIES
M .
- updated external definitions for drv,clm,cice,pio,scripts
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
050 bl374 TBL.sh fs1.9c4dh outfrq3s+1850-2005_cam4_super_fast_llnl 9s .....FAIL! rc= 7 at Fri Jun 4 12:26:23 MDT 2010
- failure expected due to bug fix in drydep code of new clm tag
002 bl375 TBL.sh f1.9c5m volc+1850-2005_cam5 2d ...........................FAIL! rc= 5 at Fri Jun 4 14:57:52 MDT 2010
- failure due to new test definition not recognized by baseline code, but would have failed anyway due to change
in volcanics in cam5 physics
edinburgh/lf95: All PASS
edinburgh/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam4_9_07
Originator(s): eaton
Date: Wed Jun 2 16:33:56 MDT 2010
One-line Summary: mods for BGC runs; mods to super_fast_llnl use cases;
misc cleanup
Purpose of changes:
. Add logic to build-namelist for the desired co2_cycle behavior:
If co2_cycle is true then:
- don't read the co2 fossil fuel fluxes when sim_year=1850
- read the co2 fossil fuel fluxes when sim_year=1850-2000 unless the user
explicitly sets co2_readflux_fuel=.false.
- make the prognostic CO2 radiatively active unless the user explicitly
sets co2_cycle_rad_passive=.true.
. mods for output when co2_cycle is active
. mods in super_fast_llnl use cases
. make the setting of npr_yz in the cam.buildnml.csh script more robust
. remove the compiler patch args from the AIX section.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist:
. namelist variables for the co2_cycle module were put into their own
namelist group co2_cycle_nl
. namelist variables for the chem_surfvals module were put into their own
namelist group chem_surfvals_nl
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: not checked
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/build-namelist
. add logic for desired co2_cycle behavior: when co2_cycle is true then
- don't read the co2 fossil fuel fluxes when sim_year=1850
- read the co2 fossil fuel fluxes when sim_year=1850-2000 unless the user
explicitly sets co2_readflux_fuel=.false.
- make the prognostic CO2 radiatively active unless the user explicitly
sets co2_cycle_rad_passive=.true.
models/atm/cam/bld/cam.cpl7.template
. make the setting of npr_yz in the cam.buildnml.csh script more robust
models/atm/cam/bld/Makefile.in
. Remove the compiler patch args from the AIX section.
models/atm/cam/bld/namelist_files/namelist_definition.xml
. change group of co2_cycle namelist variables to co2_cycle_nl
. put the chem_surfvals namelist variables into the chem_surfvals_nl group.
. remove unused variable prognostic_sulfur
. add co2_cycle_rad_passive
. updates to documentation
models/atm/cam/bld/namelist_files/use_cases/1850-2005_cam4_super_fast_llnl.xml
. adjust h0 fields
models/atm/cam/bld/namelist_files/use_cases/1850_cam4_super_fast_llnl.xml
. turn on clm drydep
models/atm/cam/src/physics/cam/aoa_tracers.F90
. put log output inside if(masterproc) condition
models/atm/cam/src/control/cam_history.F90
. replace chem_surfvals_get('CO2VMR') by chem_surfvals_co2_rad(vmr_in=.true.)
models/atm/cam/src/control/runtime_opts.F90
. remove co2_{default,set}opts routines
. add call to co2_cycle_readnl
. replace chem_surfvals_{default,set}opts calls by chem_surfvals_readnl
models/atm/cam/src/cpl_esmf/atm_comp_esmf.F90
models/atm/cam/src/cpl_mct/atm_comp_mct.F90
. remove outfld calls for SFCO2*
models/atm/cam/src/physics/cam/cam_diagnostics.F90
. add outfld calls for SFCO2*, CO2*_BOT
models/atm/cam/src/physics/cam/chem_surfvals.F90
. modify chem_surfvals_co2_rad with an optional argument to allow it to
return co2 as vmr (default is to return it as mmr).
. add chem_surfvals_readnl for reading namelist
models/atm/cam/src/physics/cam/co2_cycle.F90
. add addfld calls for CO2*_BOT
. add add_default calls for TMCO2*
models/atm/cam/src/physics/cam/co2_cycle.F90
. add co2_cycle_readnl routine to read namelist
. remove old co2_{default,set}opts routines
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS
edinburgh/lf95: All PASS
edinburgh/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
Note: answers will change when using the 1850_cam4_super_fast_llnl use case.
===============================================================
===============================================================
Tag name: cam4_9_06
Originator(s): mvr
Date: 27 May 2010
One-line Summary: cleanup of test scripts
Purpose of changes:
the filename of the configure options file used by each test did not
always contain a string indicating which physics package was being
exercised - filenames in these instances were changed
removed the cppdef -DNO_VEXP from the test scripts as it is no longer
being used by the model
removed test of waccm using cam5 physics until the configuration has been
validated
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
D models/atm/cam/test/system/config_files/f4h
D models/atm/cam/test/system/config_files/fmo1.9dh
D models/atm/cam/test/system/config_files/fm1.9m
D models/atm/cam/test/system/config_files/e64m
D models/atm/cam/test/system/config_files/s8dm
D models/atm/cam/test/system/config_files/e48m
D models/atm/cam/test/system/config_files/e64o
D models/atm/cam/test/system/config_files/scm64bfbiop
D models/atm/cam/test/system/config_files/e64bfbiop
D models/atm/cam/test/system/config_files/f4dm
D models/atm/cam/test/system/config_files/s48dh
D models/atm/cam/test/system/config_files/s32pdh
D models/atm/cam/test/system/config_files/wm4dh
D models/atm/cam/test/system/config_files/s32dh
D models/atm/cam/test/system/config_files/f10pdm
D models/atm/cam/test/system/config_files/wm1.9m
D models/atm/cam/test/system/config_files/scmarmiop
D models/atm/cam/test/system/config_files/e8dm
D models/atm/cam/test/system/config_files/f10t5mdm
D models/atm/cam/test/system/config_files/wg10dm
D models/atm/cam/test/system/config_files/scm64bfbiop_4
D models/atm/cam/test/system/config_files/fma1.9h
D models/atm/cam/test/system/config_files/e64bfbiop_4
D models/atm/cam/test/system/config_files/fm1.9dh
D models/atm/cam/test/system/config_files/fma1.9m
D models/atm/cam/test/system/config_files/e8pdm
D models/atm/cam/test/system/config_files/e8t5mdm
D models/atm/cam/test/system/config_files/s64h
D models/atm/cam/test/system/config_files/f1.9h
D models/atm/cam/test/system/config_files/s64m
D models/atm/cam/test/system/config_files/f1.9m
D models/atm/cam/test/system/config_files/e64pm
D models/atm/cam/test/system/config_files/h16x4dm
D models/atm/cam/test/system/config_files/s64o
D models/atm/cam/test/system/config_files/s8pdm
D models/atm/cam/test/system/config_files/f1.9o
D models/atm/cam/test/system/config_files/fmgpa1.9dh
D models/atm/cam/test/system/config_files/s8t5mdm
D models/atm/cam/test/system/config_files/fmgpa1.9dm
D models/atm/cam/test/system/config_files/f10dm
D models/atm/cam/test/system/config_files/s48pdh
List all subroutines added and what they do:
A models/atm/cam/test/system/config_files/f10c5pdm
A models/atm/cam/test/system/config_files/scmc5armiop
A models/atm/cam/test/system/config_files/f10c5t5mdm
A models/atm/cam/test/system/config_files/e8c5dm
A models/atm/cam/test/system/config_files/wg10c5dm
A models/atm/cam/test/system/config_files/fm1.9c5dh
A models/atm/cam/test/system/config_files/e8c5pdm
A + models/atm/cam/test/system/config_files/scm64c4bfbiop
A + models/atm/cam/test/system/config_files/e64c4bfbiop
A models/atm/cam/test/system/config_files/wm1.9c4h
A models/atm/cam/test/system/config_files/e8c5t5mdm
A models/atm/cam/test/system/config_files/s64c5h
A models/atm/cam/test/system/config_files/s64c5m
A + models/atm/cam/test/system/config_files/h16x4c3dm
A models/atm/cam/test/system/config_files/e64c5pm
A models/atm/cam/test/system/config_files/s8c5pdm
A models/atm/cam/test/system/config_files/s8c5t5mdm
A models/atm/cam/test/system/config_files/fmgpa1.9c5dm
A models/atm/cam/test/system/config_files/s32c5pdh
A models/atm/cam/test/system/config_files/f10c5dm
A models/atm/cam/test/system/config_files/s32c5dh
A models/atm/cam/test/system/config_files/fm1.9c5m
A models/atm/cam/test/system/config_files/fma1.9c5m
A models/atm/cam/test/system/config_files/s8c5dm
A models/atm/cam/test/system/config_files/e64c5m
A models/atm/cam/test/system/config_files/e48c5m
A models/atm/cam/test/system/config_files/scm64c5bfbiop
A models/atm/cam/test/system/config_files/e64c5bfbiop
A models/atm/cam/test/system/config_files/fmo1.9c5dh
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/tests_pretag_bluefire
M models/atm/cam/test/system/config_files/e48c4pdm
M models/atm/cam/test/system/config_files/wm1.9c4m
M models/atm/cam/test/system/config_files/f4c5dm
M models/atm/cam/test/system/config_files/wm4c4h
M models/atm/cam/test/system/config_files/f4c4pdh
M models/atm/cam/test/system/tests_posttag_jaguar_cb
M models/atm/cam/test/system/tests_posttag_jaguar
M models/atm/cam/test/system/input_tests_master
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
045 bl373 TBL.sh fm1.9c5dh outfrq3s 9s ....................................FAIL! rc= 5 at Thu May 27 11:16:50 MDT 2010
060 bl379 TBL.sh wm1.9c4h outfrq24h+waccm_1850_cam4 2d ....................FAIL! rc= 5 at Thu May 27 12:49:01 MDT 2010
063 bl731 TBL.sh h16x4c3dm aqua 9s ........................................FAIL! rc= 5 at Thu May 27 12:53:53 MDT 2010
- failures were due to the change in name of the respective configuration options file, where it was not recognized
by the baseline scripts
edinburgh/lf95:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 5 at Wed May 26 11:18:09 MDT 2010
008 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 5 at Wed May 26 11:30:34 MDT 2010
010 bl113 TBL.sh e8c5pdm aqpgro 3s ........................................FAIL! rc= 5 at Wed May 26 11:33:53 MDT 2010
014 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 5 at Wed May 26 11:38:42 MDT 2010
021 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 5 at Wed May 26 12:04:26 MDT 2010
029 bl312 TBL.sh f10c5dm ghgrmp 9s ........................................FAIL! rc= 5 at Wed May 26 12:29:01 MDT 2010
031 bl313 TBL.sh f10c5pdm aqpgro 3s .......................................FAIL! rc= 5 at Wed May 26 12:32:56 MDT 2010
034 bl314 TBL.sh wg10c5dm outfrq3s 9s .....................................FAIL! rc= 5 at Wed May 26 12:54:55 MDT 2010
039 bl316 TBL.sh f10c5pdm aqpgro+aquaplanet_cam5 3s .......................FAIL! rc= 5 at Wed May 26 13:02:29 MDT 2010
- failures were due to the change in name of the respective configuration options file, where it was not recognized
by the baseline scripts
edinburgh/pgi:
004 bl111 TBL.sh e8c5t5mdm ttrac 9s .......................................FAIL! rc= 5 at Wed May 26 11:07:28 MDT 2010
008 bl112 TBL.sh e8c5dm ghgrmp 9s .........................................FAIL! rc= 5 at Wed May 26 11:10:34 MDT 2010
010 bl113 TBL.sh e8c5pdm aqpgro 3s ........................................FAIL! rc= 5 at Wed May 26 11:12:41 MDT 2010
014 bl114 TBL.sh e8c5dm co2rmp 9s .........................................FAIL! rc= 5 at Wed May 26 11:14:03 MDT 2010
021 bl311 TBL.sh f10c5t5mdm ttrac 9s ......................................FAIL! rc= 5 at Wed May 26 11:21:38 MDT 2010
029 bl312 TBL.sh f10c5dm ghgrmp 9s ........................................FAIL! rc= 5 at Wed May 26 11:26:50 MDT 2010
031 bl313 TBL.sh f10c5pdm aqpgro 3s .......................................FAIL! rc= 5 at Wed May 26 11:29:08 MDT 2010
034 bl314 TBL.sh wg10c5dm outfrq3s 9s .....................................FAIL! rc= 5 at Wed May 26 11:32:59 MDT 2010
039 bl316 TBL.sh f10c5pdm aqpgro+aquaplanet_cam5 3s .......................FAIL! rc= 5 at Wed May 26 11:37:13 MDT 2010
- failures were due to the change in name of the respective configuration options file, where it was not recognized
by the baseline scripts
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: b4b
===============================================================
Tag name: cam4_9_05
Originator(s): sungsu, aconley, hannay, eaton
Date: Wed May 26 09:24:56 MDT 2010
One-line Summary: "final" cam5 version
Purpose of changes:
. Mods to CAM5 physics package:
- change to the initialization for the radius in prescribed_volcaero.F90
- correction to the lookup table in aer_rad_props.F90
- fixes in the wet scavenging of aerosols
- solubility factor = 0.4 instead of 0.6
- increase SOAG by 1.5
- reduction in black & organic carbon number by a factor of 3
This is expected to be the final version of the CAM5 physics for the
CESM-1.0 release except for bug fixes and code cleanup.
Bugs fixed (include bugzilla ID):
. bug fix in modal_aero_lw (makes refractive index vary with wavelength)
. bugfix in aerosol activation
. bugfix in wet deposition
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets:
- new emission datasets for soag and black carbon numbers (num_a1)
Describe any substantial timing or memory changes: not checked
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
models/atm/cam/bld/namelist_files/use_cases/1850-2005_cam5.xml
models/atm/cam/bld/namelist_files/use_cases/1850_cam5.xml
models/atm/cam/bld/namelist_files/use_cases/1850_cam5_super_fast_llnl.xml
models/atm/cam/src/chemistry/bulk_aero/aerosol_intr.F90
models/atm/cam/src/chemistry/bulk_aero/dust_intr.F90
models/atm/cam/src/chemistry/bulk_aero/progseasalts_intr.F90
models/atm/cam/src/chemistry/bulk_aero/sulchem.F90
models/atm/cam/src/chemistry/bulk_aero/wetdep.F90
models/atm/cam/src/chemistry/mozart/mz_aerosols_intr.F90
models/atm/cam/src/chemistry/utils/prescribed_volcaero.F90
models/atm/cam/src/physics/cam/aer_rad_props.F90
models/atm/cam/src/physics/cam/cldwat2m_micro.F90
models/atm/cam/src/physics/cam/cloud_fraction.F90
models/atm/cam/src/physics/cam/convect_deep.F90
models/atm/cam/src/physics/cam/convect_shallow.F90
models/atm/cam/src/physics/cam/eddy_diff.F90
models/atm/cam/src/physics/cam/modal_aer_opt.F90
models/atm/cam/src/physics/cam/ndrop.F90
models/atm/cam/src/physics/cam/stratiform.F90
models/atm/cam/src/physics/cam/tphysbc.F90
models/atm/cam/src/physics/cam/uwshcu.F90
models/atm/cam/src/physics/cam/vertical_diffusion.F90
models/atm/cam/src/physics/cam/zm_conv_intr.F90
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Tue May 25 15:23:22 MDT 2010
013 bl330 TBL.sh f4c5pdh aqpgro 3s ........................................FAIL! rc= 7 at Tue May 25 15:30:27 MDT 2010
021 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...............................FAIL! rc= 7 at Tue May 25 15:49:40 MDT 2010
036 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ......................................FAIL! rc= 7 at Tue May 25 16:27:50 MDT 2010
040 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .....................................FAIL! rc= 7 at Tue May 25 16:47:45 MDT 2010
045 bl373 TBL.sh fm1.9dh outfrq3s 9s ......................................FAIL! rc= 7 at Tue May 25 17:23:04 MDT 2010
047 bl375 TBL.sh f1.9c5m outfrq24h+1850-2005_cam5 2d ......................FAIL! rc= 7 at Tue May 25 17:40:45 MDT 2010
edinburgh/lf95: All PASS except:
004 bl111 TBL.sh e8t5mdm ttrac 9s .........................................FAIL! rc= 7 at Tue May 25 15:08:15 MDT 2010
008 bl112 TBL.sh e8dm ghgrmp 9s ...........................................FAIL! rc= 7 at Tue May 25 15:31:59 MDT 2010
010 bl113 TBL.sh e8pdm aqpgro 3s ..........................................FAIL! rc= 7 at Tue May 25 15:38:27 MDT 2010
014 bl114 TBL.sh e8dm co2rmp 9s ...........................................FAIL! rc= 7 at Tue May 25 15:44:34 MDT 2010
021 bl311 TBL.sh f10t5mdm ttrac 9s ........................................FAIL! rc= 7 at Tue May 25 16:24:29 MDT 2010
029 bl312 TBL.sh f10dm ghgrmp 9s ..........................................FAIL! rc= 7 at Tue May 25 17:02:39 MDT 2010
031 bl313 TBL.sh f10pdm aqpgro 3s .........................................FAIL! rc= 7 at Tue May 25 17:10:39 MDT 2010
034 bl314 TBL.sh wg10dm outfrq3s 9s .......................................FAIL! rc= 7 at Tue May 25 17:47:25 MDT 2010
039 bl316 TBL.sh f10pdm aqpgro+aquaplanet_cam5 3s .........................FAIL! rc= 7 at Tue May 25 17:55:54 MDT 2010
edinburgh/pgi: All PASS except:
004 bl111 TBL.sh e8t5mdm ttrac 9s .........................................FAIL! rc= 7 at Tue May 25 14:44:21 MDT 2010
008 bl112 TBL.sh e8dm ghgrmp 9s ...........................................FAIL! rc= 7 at Tue May 25 14:49:54 MDT 2010
010 bl113 TBL.sh e8pdm aqpgro 3s ..........................................FAIL! rc= 7 at Tue May 25 14:54:09 MDT 2010
014 bl114 TBL.sh e8dm co2rmp 9s ...........................................FAIL! rc= 7 at Tue May 25 14:55:46 MDT 2010
021 bl311 TBL.sh f10t5mdm ttrac 9s ........................................FAIL! rc= 7 at Tue May 25 15:05:59 MDT 2010
029 bl312 TBL.sh f10dm ghgrmp 9s ..........................................FAIL! rc= 7 at Tue May 25 15:13:48 MDT 2010
031 bl313 TBL.sh f10pdm aqpgro 3s .........................................FAIL! rc= 7 at Tue May 25 15:18:22 MDT 2010
034 bl314 TBL.sh wg10dm outfrq3s 9s .......................................FAIL! rc= 7 at Tue May 25 15:25:17 MDT 2010
039 bl316 TBL.sh f10pdm aqpgro+aquaplanet_cam5 3s .........................FAIL! rc= 7 at Tue May 25 15:29:38 MDT 2010
All failures are baseline tests that are expected to fail due to changed
answers with cam5 physics.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: cam4 is BFB. cam5 has new climate.
===============================================================
===============================================================
Tag name: cam4_9_04
Originator(s): jet,eaton,mvertens
Date: 5-20-2010
One-line Summary: Added aquaplanet as component model and make docn default for cam
Purpose of changes: aquaplanet is now a supported configuration. It is no longer
a namelist option but is now an -ocn aquaplanet option to the configure script.
The configure -ocn aquaplanet and build-namelist calls will produce a cam
executable that uses the default analytic sst pattern #1. I also included two
new use cases that allow building namelists which have specific settings for the
ape experiments (resetting of model constants, special boundary datasets, etc.)
There is a use case for cam3/cam4 called aquaplanet_cam3_cam4 and one for cam5
physics called aquaplanet_cam5. The cam3/cam4 use-case turns off aerosol forcing
and the cam5 use-case uses most of the cam5 default configuration which includes
prognostic aerosols.
DOCN is now the default ocn component for CAM. DOM is still an option to configure.
The DOCN model uses a slightly different interpolation algorithm and is roundoff
compared to DOM.
Cleaned up the use of shr_const in the model. Replace most of the shr_const parameters
with their physconst analogs. This allows a consistent use and derivation of physical
constants for when running aquaplanet experiments.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
DOCN is the new default ocn component
DOM must now be specified via -ocn dom setting of configure
aquaplanet is now an configure option -ocn aquaplanet
Describe any changes made to the namelist:
=====================================================================
Brought some additional resolution tuning parameters into the namelist
=====================================================================
Cldwat
cldwat_icritc !threshold for autoconversion of cold ice
cldwat_icritw !threshold for autoconversion of warm ice
cldwat_conke !tunable constant for evaporation of precip
hkconv
hkconv_cmftau !characteristic adjustment time scale
hkconv_c0 !rain water autoconversion coefficient
zmconv
zmconv_ke !Tunable evaporation efficiency
=====================================================================
Added docn parameters to namelist
=====================================================================
restfilm !Full pathname of docn restart file.
ocn_in !ocn type
decomp !1d
=====================================================================
Added namelist option to pick analytic sst pattern for aquaplanet
=====================================================================
aqua_planet_sst !Set the sst to a particular analytic solution.
=====================================================================
Added 6 physical constants to namelist to support APE experiments
=====================================================================
sday ! sec in siderial day ~ sec
rearth ! radius of earth (m)
gravit ! gravitational acceleration (m/s**2)
mwdry ! molecular weight dry air
mwh2o ! molecular weight h2o
cpwv ! specific heat of water vapor (J/K/kg)
These values default to their shr_const values if not reset in the namelist
List any changes to the defaults for the boundary datasets:
added new solar and ozone datasets for aquaplanet experiments.
Describe any substantial timing or memory changes:
Code reviewed by: jet eaton
List all subroutines eliminated:
Moved physconst to control
D models/atm/cam/src/physics/cam/physconst.F90
Replace shr_const values with physconst equivalents
D models/atm/cam/src/dynamics/fv/dynconst.F90
D models/atm/cam/src/dynamics/sld/dynconst.F90
D models/atm/cam/src/dynamics/eul/dynconst.F90
List all subroutines added and what they do:
New Use Cases for Aquaplanet
A models/atm/cam/bld/namelist_files/use_cases/aquaplanet_cam5.xml
A models/atm/cam/bld/namelist_files/use_cases/aquaplanet_cam3_cam4.xml
Moved physconst from physics to control
A + models/atm/cam/src/control/physconst.F90
Added new utils directory for aquaplanet support
A models/atm/cam/src/utils/cam_aqua
A models/atm/cam/src/utils/cam_aqua/ocn_types.F90
A models/atm/cam/src/utils/cam_aqua/cpl_mct
A models/atm/cam/src/utils/cam_aqua/cpl_mct/ocn_comp_mct.F90
A models/atm/cam/src/utils/cam_aqua/cpl_esmf
A models/atm/cam/src/utils/cam_aqua/ocn_comp.F90
List all existing files that have been modified, and describe the changes:
Aquaplanet support
M models/atm/cam/test/system/config_files/e48c4pdm
M models/atm/cam/test/system/config_files/f10c3dm
M models/atm/cam/test/system/config_files/s32pdh
M models/atm/cam/test/system/config_files/f10pdm
M models/atm/cam/test/system/config_files/f4c5pdh
M models/atm/cam/test/system/config_files/e8pdm
M models/atm/cam/test/system/config_files/e48adh
M models/atm/cam/test/system/config_files/s8pdm
M models/atm/cam/test/system/config_files/h16x4dm
M models/atm/cam/test/system/config_files/e64pm
M models/atm/cam/test/system/config_files/f4c4pdh
New diagnostic tests for aquaplanet use case
M models/atm/cam/test/system/tests_posttag_edinburgh
M models/atm/cam/test/system/tests_pretag_edinburgh
Added support for docn branch
M models/atm/cam/test/system/TBR.sh
Aquaplanet/docn support
M models/atm/cam/test/system/nl_files/aqua
M models/atm/cam/test/system/nl_files/aqpgro
M models/atm/cam/test/system/input_tests_master
M models/atm/cam/bld/configure
M models/atm/cam/bld/config_files/definition.xml
Added new namelist parameters that control tuning parameters and docn component
M models/atm/cam/bld/build-namelist
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
M models/atm/cam/bld/namelist_files/namelist_definition.xml
M models/atm/cam/src/physics/cam/cloud_fraction.F90
M models/atm/cam/src/physics/cam/zm_conv.F90
M models/atm/cam/src/physics/cam/cldwat.F90
M models/atm/cam/src/physics/cam/hk_conv.F90
M models/atm/cam/src/control/runtime_opts.F90
Code cleanup
M models/atm/cam/src/control/ncdio_atm.F90
M models/atm/cam/src/utils/cam_csim4/cpl_mct/ice_comp_mct.F90
M models/atm/cam/src/utils/cam_csim4/print_coverage.F90
M models/atm/cam/src/utils/cam_dom/sst_data.F90
M models/atm/cam/src/utils/cam_dom/cpl_mct/ocn_comp_mct.F90
M models/atm/cam/src/utils/cam_dom/ocn_comp.F90
M models/atm/cam/src/control/cam_history.F90
updated to latest docn7/csm_share/drv components to allow support for docn and aquaplanet
M SVN_EXTERNAL_DIRECTORIES
shr_const parameters replaced with physconst equivalents
M models/atm/cam/src/physics/cam/tropopause.F90
M models/atm/cam/src/physics/cam/boundarydata.F90
M models/atm/cam/src/physics/cam/phys_gmean.F90
M models/atm/cam/src/physics/cam/ghg_data.F90
M models/atm/cam/src/physics/cam/chem_surfvals.F90
M models/atm/cam/src/physics/cam/pkg_cld_sediment.F90
M models/atm/cam/src/physics/cam/co2_cycle.F90
M models/atm/cam/src/physics/cam/tidal_diag.F90
M models/atm/cam/src/physics/waccm/nlte_fomichev.F90
M models/atm/cam/src/physics/waccm/gw_drag.F90
M models/atm/cam/src/physics/waccm/gravity_waves_sources.F90
M models/atm/cam/src/chemistry/utils/mo_constants.F90
M models/atm/cam/src/chemistry/utils/solar_data.F90
M models/atm/cam/src/chemistry/utils/tracer_data.F90
M models/atm/cam/src/chemistry/utils/mo_msis_ubc.F90
M models/atm/cam/src/chemistry/bulk_aero/progseasalts_intr.F90
M models/atm/cam/src/chemistry/bulk_aero/dust_intr.F90
M models/atm/cam/src/chemistry/mozart/chlorine_loading_data.F90
M models/atm/cam/src/chemistry/mozart/mo_aero_settling.F90
M models/atm/cam/src/chemistry/mozart/mo_airglow.F90
M models/atm/cam/src/chemistry/mozart/mo_jlong.F90
M models/atm/cam/src/chemistry/mozart/mo_lightning.F90
M models/atm/cam/src/chemistry/mozart/lin_strat_chem.F90
M models/atm/cam/src/chemistry/mozart/mo_jshort.F90
M models/atm/cam/src/chemistry/mozart/mo_sulf.F90
M models/atm/cam/src/chemistry/mozart/set_cp.F90
M models/atm/cam/src/chemistry/mozart/mo_aurora.F90
M models/atm/cam/src/chemistry/mozart/mo_apex.F90
M models/atm/cam/src/chemistry/mozart/mo_snoe.F90
M models/atm/cam/src/chemistry/mozart/mo_sphers.F90
M models/atm/cam/src/chemistry/mozart/mo_sad.F90
M models/atm/cam/src/chemistry/mozart/mo_setinv.F90
M models/atm/cam/src/chemistry/mozart/chemistry.F90
M models/atm/cam/src/chemistry/mozart/mo_drydep.F90
M models/atm/cam/src/chemistry/mozart/iondrag.F90
M models/atm/cam/src/chemistry/mozart/mo_photo.F90
M models/atm/cam/src/chemistry/mozart/mo_sethet.F90
M models/atm/cam/src/chemistry/mozart/efield.F90
M models/atm/cam/src/chemistry/mozart/mz_aerosols_intr.F90
M models/atm/cam/src/chemistry/mozart/mo_usrrxt.F90
M models/atm/cam/src/chemistry/mozart/mo_jeuv.F90
M models/atm/cam/src/chemistry/mozart/exbdrift.F90
M models/atm/cam/src/chemistry/mozart/mo_solarproton.F90
M models/atm/cam/src/chemistry/mozart/mo_synoz.F90
M models/atm/cam/src/chemistry/mozart/mo_params.F90
M models/atm/cam/src/dynamics/sld/dyn_grid.F90
M models/atm/cam/src/dynamics/sld/linemsdyn.F90
M models/atm/cam/src/dynamics/sld/initcom.F90
M models/atm/cam/src/dynamics/sld/settau.F90
M models/atm/cam/src/dynamics/sld/scanslt.F90
M models/atm/cam/src/dynamics/sld/spetru.F90
M models/atm/cam/src/dynamics/sld/dyn.F90
M models/atm/cam/src/dynamics/sld/quad.F90
M models/atm/cam/src/dynamics/sld/grcalc.F90
M models/atm/cam/src/dynamics/eul/dyn_grid.F90
M models/atm/cam/src/dynamics/eul/initcom.F90
M models/atm/cam/src/dynamics/eul/scanslt.F90
M models/atm/cam/src/dynamics/eul/inidat.F90
M models/atm/cam/src/dynamics/eul/spetru.F90
M models/atm/cam/src/dynamics/eul/iop.F90
M models/atm/cam/src/dynamics/eul/dyn.F90
M models/atm/cam/src/dynamics/eul/quad.F90
M models/atm/cam/src/dynamics/eul/grcalc.F90
M models/atm/cam/src/dynamics/homme/external/physical_constants.F90
M models/atm/cam/src/dynamics/homme/dyn_grid.F90
M models/atm/cam/src/dynamics/fv/dyn_grid.F90
M models/atm/cam/src/dynamics/fv/initcom.F90
M models/atm/cam/src/dynamics/fv/dyn_comp.F90
M models/atm/cam/src/dynamics/fv/inital.F90
M models/atm/cam/src/dynamics/fv/restart_dynamics.F90
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
Since docn gives roundoff differences to dom all the baseline tests
were expected to fail
When the code was tested with dom all tests past as expected.
bluefire:
004 bl132 TBL.sh e48c4dh ghgrmp+1850_cam4 9s ..............................FAIL! rc= 7
017 bl331 TBL.sh f4c4dh co2rmp+1850_cam4 9s ...............................FAIL! rc= 7
033 bl336 TBL.sh wm4c4h outfrq3s+waccm_1850_cam4 9s .......................FAIL! rc= 7
050 bl374 TBL.sh fs1.9c4dh outfrq3s+1850-2005_cam4_super_fast_llnl 9s .....FAIL! rc= 7
054 bl387 TBL.sh f1.9c4dm outfrq3s+1850-2005_cam4 9s ......................FAIL! rc= 7
057 bl388 TBL.sh f1.9c4m outfrq24h+1850_cam4 2d ...........................FAIL! rc= 7
060 bl389 TBL.sh wm1.9c4m outfrq24h+waccm_1850_cam4 2d ....................FAIL! rc= 7
068 bl992 TBL_ccsm.sh f19_g16 E 2d ........................................FAIL! rc= 7
edinburgh/lf95:
034 bl314 TBL.sh wg10dm outfrq3s 9s .......................................FAIL! rc= 7
037 bl315 TBL.sh f10c3dm outfrq3s 9s ......................................FAIL! rc= 7
edinburgh/pgi:
034 bl314 TBL.sh wg10dm outfrq3s 9s .......................................FAIL! rc= 7
037 bl315 TBL.sh f10c3dm outfrq3s 9s ......................................FAIL! rc= 7
039 bl316 TBL.sh f10pdm aqpgro+aquaplanet_cam5 3s .........................SKIPPED*
bl316 is a new test for the aquaplanet5 use case and didn't exist in the base line.
CAM tag used for the baseline comparison tests if different than previous
tag:cam4_9_03
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
roundoff changes.
If bitwise differences were observed, how did you show they were no worse
than roundoff?
tested docn sstfield against original dom sst. Some values were different in the last bit.
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
Tag name: cam4_9_03
Originator(s): mvr, eaton, jet
Date: 14 May 2010
One-line Summary: made prescribed aerosol mode the default; mod to cam.cpl7.template to read users CCSM namelist
from CASEROOT/user_nl_cam; scam bugfix; back out workaround for esmf bug; new jan1 ncdata for 0.23x0.31
Purpose of changes:
- the user namelist was previously read from $CASEROOT/SourceMods/src.cam/user_nl
- reverting back to the cam4_1_14 version of time_manager.F90 to undo workaround for esmf bug
- cam's handling of prescribed aerosols and their depositions will allow surface models to remove this functionality
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
- added settings for aerosol deposition
List any changes to the defaults for the boundary datasets:
- new jan1 ncdata file for 0.23x0.31
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/build-namelist
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
M models/atm/cam/bld/namelist_files/use_cases/1850-2005_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/2005-2100_cam4_rcp26.xml
M models/atm/cam/bld/namelist_files/use_cases/2005-2100_cam4_rcp45.xml
M models/atm/cam/bld/namelist_files/use_cases/2005_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/2005-2100_cam4_rcp85.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1850-2005_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/2000_cam4_super_fast_llnl.xml
M models/atm/cam/bld/namelist_files/use_cases/1850_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/1850-2005_cam4_super_fast_llnl.xml
M models/atm/cam/bld/namelist_files/use_cases/2005-2100_cam4_rcp85v2.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1974-2005_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1850_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/1850_cam4_super_fast_llnl.xml
M models/atm/cam/bld/namelist_files/use_cases/1850_cam4_2xco2.xml
M models/atm/cam/src/physics/cam/phys_control.F90
- mods need to make prescribed aerosols the default
M models/atm/cam/bld/cam.cpl7.template
- new reading user namelist from $CASEROOT/user_nl_cam
M models/atm/cam/src/utils/time_manager.F90
- back out workaround for esmf bug
M models/atm/cam/src/utils/cam_dom/sst_data.F90
- scam bug fix
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
004 bl132 TBL.sh e48c4dh ghgrmp+1850_cam4 9s ..............................FAIL! rc= 7 at Fri May 14 07:05:21 MDT 2010
017 bl331 TBL.sh f4c4dh co2rmp+1850_cam4 9s ...............................FAIL! rc= 7 at Fri May 14 07:05:29 MDT 2010
033 bl336 TBL.sh wm4c4h outfrq3s+waccm_1850_cam4 9s .......................FAIL! rc= 7 at Fri May 14 07:05:55 MDT 2010
050 bl374 TBL.sh fs1.9c4dh outfrq3s+1850-2005_cam4_super_fast_llnl 9s .....FAIL! rc= 7 at Fri May 14 07:06:10 MDT 2010
054 bl387 TBL.sh f1.9c4dm outfrq3s+1850-2005_cam4 9s ......................FAIL! rc= 7 at Fri May 14 07:06:17 MDT 2010
057 bl388 TBL.sh f1.9c4m outfrq24h+1850_cam4 2d ...........................FAIL! rc= 7 at Fri May 14 07:06:23 MDT 2010
060 bl389 TBL.sh wm1.9c4m outfrq24h+waccm_1850_cam4 2d ....................FAIL! rc= 7 at Fri May 14 07:07:48 MDT 2010
068 bl992 TBL_ccsm.sh f19_g16 E 2d ........................................FAIL! rc= 7 at Fri May 14 07:07:54 MDT 2010
- any baseline test where prescribed aerosols were used were expected to fail
edinburgh/lf95:
034 bl314 TBL.sh wg10dm outfrq3s 9s .......................................FAIL! rc= 7 at Thu May 13 17:31:21 MDT 2010
037 bl315 TBL.sh f10c3dm outfrq3s 9s ......................................FAIL! rc= 7 at Thu May 13 17:33:41 MDT 2010
- any baseline test where prescribed aerosols were used were expected to fail
edinburgh/pgi:
034 bl314 TBL.sh wg10dm outfrq3s 9s .......................................FAIL! rc= 7 at Thu May 13 16:15:27 MDT 2010
037 bl315 TBL.sh f10c3dm outfrq3s 9s ......................................FAIL! rc= 7 at Thu May 13 16:15:27 MDT 2010
- any baseline test where prescribed aerosols were used were expected to fail
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: runs using prescribed aerosols are different by larger than roundoff but presumably
same climate - we are relying on dbaily and dlawrence signing off on validation runs with this change
===============================================================
Tag name: cam4_9_02
Originator(s): mvr, eaton, mirin
Date: 05 May 2010
One-line Summary: enhanced testing of cam4 configurations; updated externals of
surface components to match those of ccsm4_0_beta52; cleaned up test scripts to
reflect decommissioning of calgary; addition of fields to master field list;
fix of potential bug in fv dycore
Purpose of changes:
-the cgd linux cluster, calgary was decommissioned on 5/5/10...this was a primary
test platform for cam - testing will now take place on its replacement, edinburgh
-master field list needed some additional fields being requested for the "moar"
production runs
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
- fsurdat settings updated for T85 and 0.23x0.31
- fpftdyn and faerdep settings added for rcp=4.5,2.6
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
D models/atm/cam/test/system/config_files/fmo2dh
- removed obsolete test file
D models/atm/cam/test/system/config_files/scm64bfbiop_35
D models/atm/cam/test/system/config_files/e64bfbiop_35
- test files renamed (cam3_5 -> cam4)
D models/atm/cam/test/system/tests_posttag_calgary
D models/atm/cam/test/system/tests_pretag_calgary
- test files renamed to reflect decommissioning of calgary
List all subroutines added and what they do:
A + models/atm/cam/test/system/tests_posttag_edinburgh
A + models/atm/cam/test/system/tests_pretag_edinburgh
- files containing test lists renamed to reflect decommissioning of calgary
A models/atm/cam/test/system/config_files/wm1.9c4m
A models/atm/cam/test/system/config_files/f0.9c4m
A models/atm/cam/test/system/config_files/fm1.9dh
A models/atm/cam/test/system/nl_files/outfrq1m
- new test files for enhanced testing of cam4 configurations, trop_mozart chemistry
A models/atm/cam/test/system/config_files/scm64bfbiop_4
A models/atm/cam/test/system/config_files/e64bfbiop_4
- test files renamed (cam3_5 -> cam4)
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/input_tests_master
M models/atm/cam/test/system/tests_pretag_bluefire
M models/atm/cam/test/system/tests_posttag_jaguar_cb
M models/atm/cam/test/system/tests_posttag_jaguar
M models/atm/cam/test/system/tests_posttag_bluefire
- additional tests for cam4 physics, trop_mozart chemistry
M models/atm/cam/test/system/TCB.sh
- fixed the error trapping for calls to generate_cice_decomp.pl
M models/atm/cam/test/system/test_driver.sh
M models/atm/cam/test/system/CAM_runcmnd.sh
- stripped out settings in test scripts for calgary
M models/atm/cam/bld/configure
- default number of vertical levels is now set to 26 only for non-waccm chemistry runs
using cam3 or cam4 physics
M models/atm/cam/bld/run-cray.csh
- run template script for xt machines will now use default version of compilers
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- update to default inputdata files to coincide with updated clm tag
M models/atm/cam/bld/namelist_files/use_cases/waccm_1850-2005_cam4.xml
- added temporary clim_pi SST datasets to this use_case
M models/atm/cam/src/physics/cam/cam_diagnostics.F90
. add T200, T500, Z100, Z200 to master fieldlist
. add interpolation and outfld calls for new fields
. add hist_fld_active conditionals to avoid diagnostic calculations and/or
interpolations for fields that don't appear in any history file.
. remove duplicate routine polysvp
M models/atm/cam/src/dynamics/fv/dp_coupling.F90
. add some omp directives
M models/atm/cam/src/dynamics/fv/cd_core.F90
. fix dimension in work array (This is fixing a potential bug, not a
currently occurring bug.)
M models/atm/cam/src/dynamics/fv/sw_core.F90
. put declaration for u2, v2 inside ifdef FILTER_MASS_FLUXES conditional
(this option isn't being enabled in cam. This fix is just saving memory
by not declaring unused arrays)
M SVN_EXTERNAL_DIRECTORIES
M .
- updated externals to match settings for ccsm4_0_beta52 (drv, clm, cice, docn, csm_share, scripts)
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
009 bl134 TBL.sh e48adh adia 9s ...........................................FAIL! rc= 7 at Tue May 4 17:57:40 MDT 2010
024 bl334 TBL.sh f4adh fv2d_8tsk 9s .......................................FAIL! rc= 7 at Tue May 4 18:37:16 MDT 2010
028 bl335 TBL.sh f4idm idphys 9s ..........................................FAIL! rc= 7 at Tue May 4 18:39:03 MDT 2010
043 bl373 TBL.sh fm1.9dh outfrq3s 9s ......................................FAIL! rc= 5 at Tue May 4 19:41:52 MDT 2010
058 bl389 TBL.sh wm1.9c4m outfrq24h+waccm_1850_cam4 2d ....................FAIL! rc= 5 at Tue May 4 21:12:43 MDT 2010
002 bl330 TBL.sh f4c5pdh aqpgro 3s ........................................FAIL! rc= 5 at Wed May 5 11:13:44 MDT 2010
- adiabatic and ideal physics baseline tests failed because they're now running at 30-levels which require
a different initial file
- bl373, bl389 and bl330 failed because they introduce a new test configuration that was not recognized by the baseline code
edinburgh/lf95:
017 bl115 TBL.sh e8idm idphys 9s ..........................................FAIL! rc= 7 at Tue May 4 17:32:52 MDT 2010
- ideal physics baseline test failed because it's now running at 30-levels which requires a different initial file
edinburgh/pgi:
017 bl115 TBL.sh e8idm idphys 9s ..........................................FAIL! rc= 7 at Tue May 4 16:58:15 MDT 2010
- ideal physics baseline test failed because it's now running at 30-levels which requires a different initial file
CAM tag used for the baseline comparison tests if different than previous
tag: bl351, bl352, bl385, bl386, bl388, bl389 were run against cam4_9_00 to ensure nothing changed
answers for these configurations with previous tag
Summarize any changes to answers: b4b except for adiabatic and ideal physics configurations which now run at 30 levels
===============================================================
===============================================================
Tag name: cam4_9_01
Originator(s): Francis Vitt
Date: 3 May 2010
One-line Summary: Updates to the 1850-2005_cam4_super_fast_llnl build-namelist use case
and radiation output bug fixes.
Purpose of changes:
Changes to the 1850-2005_cam4_super_fast_llnl use case are needed for the
track1 with super_fast_llnl chemistry 20th century transient simulation.
Fix bugs:
- the SWCF radiation diagnostic needs to be in the dosw if block
- output of undefined fields when MG microphysics is not used
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Brian Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
U models/atm/cam/bld/namelist_files/use_cases/1850-2005_cam4_super_fast_llnl.xml
- mods for the 20th century transient simulation with super_fast_llnl chemistry
. use 1-degree emission data sets
. changed to be consistent with the 1850-2005_cam4 use case
. fincl1 output fields list was adjusted
U models/atm/cam/src/physics/cam/radiation.F90
- move the outfld call for SWCF diagnostic to the dosw if block
U models/atm/cam/src/physics/cam/radiation_data.F90
- modified to not output physics buffer fields which are not defined when
'MG' is not used
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: all PASS except
045 bl374 TBL.sh fs1.9c4dh outfrq3s+1850-2005_cam4_super_fast_llnl 9s .....FAIL! rc= 7 at Mon May 3 13:02:43 MDT 2010
- this is expected to fail due to changes in the 1850-2005_cam4_super_fast_llnl use case
edinburgh/lf95: all PASS
edinburgh/pgi: all PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam4_9_00
Originator(s): mvr, sungsu, hannay, eaton, others
Date: 29 Apr 2010
One-line Summary: updated the physics with code from the cam5 development branch;
some changes to the dynamics, including the enabling of fourth-order divergence damping
Purpose of changes:
Devlopment of the cam5 physics has been occuring on a code branch for some
time. The merge of that code base with that of the cam trunk is now complete,
including backwards compatibility with cam4 and cam3 physics. For more details
on the changes, see the ChangeLog file for the branch camdev_cam3_6_23.
There were some changes to the dynamics included with this merge including:
- Switch to enable fourth-order divergence damping except at the model top layers where regular
second-order (Laplacian) damping is applied (damps both divergence and vorticity).
- FV no longer writes/reads "DELP" to/from IC file.
- Added another nested loop to the FV dynamical core. allow multiple vertical
remappings within a physics time-step. added new namelist variable: NSPLTVRM (for
split vertical remapping) as an addtion to the NSPLIT and NSPLTRAC dynamics
splitting namelist variables. Default is 1.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
- changes in support of new cam5 physics (addition of flag to specify pbl scheme, etc)
Describe any changes made to the namelist:
- changes in support of new cam5 physics (settings for turbulent mountain stress,
convective water in radiation, condensate-to-rain autoconversion coefficients, etc)
List any changes to the defaults for the boundary datasets:
- files now specified for running with 30 vertical levels
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated: none
D models/atm/cam/src/physics/cam/cldwat2m.F90
- file split into two, one for microphysics and one for macrophysics
List all subroutines added and what they do:
A + models/atm/cam/bld/namelist_files/use_cases/1850_cam5_super_fast_llnl.xml
- new use_case using cam5 physics
A + models/atm/cam/src/physics/cam/cldwat2m_macro.F90
A + models/atm/cam/src/physics/cam/cldwat2m_micro.F90
- new physics files created from cldwat2m.F90
A + models/atm/cam/src/physics/cam/uwshcu.F90
- new physics for university of wash shallow convection scheme
List all existing files that have been modified, and describe the changes:
M models/atm/cam
- externals property updated to reflect changes in SVN_EXTERNAL_DIRECTORIES from cam4_1_08
M + models/atm/cam/test/system/nl_files/off1.9x2.5
- update of test file to use input file with 30 vertical levels
M models/atm/cam/test/system/nl_files/scm_b4b_o1
M models/atm/cam/test/system/nl_files/scm_prep
- update of scam tests to ensure 3d model and scam runs use identical inputs
M models/atm/cam/test/system/CAM_runcmnd.sh
- just a change in white space
M models/atm/cam/bld/configure
M models/atm/cam/bld/config_files/definition.xml
M models/atm/cam/bld/build-namelist
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
M models/atm/cam/bld/namelist_files/namelist_definition.xml
M models/atm/cam/src/control/runtime_opts.F90
M models/atm/cam/src/control/fv_control_mod.F90
M models/atm/cam/src/control/cam_comp.F90
M models/atm/cam/src/physics/cam/molec_diff.F90
M models/atm/cam/src/physics/cam/param_cldoptics.F90
M models/atm/cam/src/physics/cam/trb_mtn_stress.F90
M models/atm/cam/src/physics/cam/radiation.F90
M models/atm/cam/src/physics/cam/tphysbc.F90
M models/atm/cam/src/physics/cam/aer_rad_props.F90
M models/atm/cam/src/physics/cam/phys_prop.F90
M models/atm/cam/src/physics/cam/phys_control.F90
M models/atm/cam/src/physics/cam/check_energy.F90
M models/atm/cam/src/physics/cam/cam_diagnostics.F90
M models/atm/cam/src/physics/cam/stratiform.F90
M models/atm/cam/src/physics/cam/constituents.F90
M models/atm/cam/src/physics/cam/modal_aer_opt.F90
M models/atm/cam/src/physics/cam/physics_types.F90
M models/atm/cam/src/physics/cam/diffusion_solver.F90
M models/atm/cam/src/physics/cam/ndrop.F90
M models/atm/cam/src/physics/cam/physpkg.F90
M models/atm/cam/src/physics/cam/rad_constituents.F90
M models/atm/cam/src/physics/cam/convect_shallow.F90
M models/atm/cam/src/physics/cam/cloud_fraction.F90
M models/atm/cam/src/physics/cam/convect_deep.F90
M models/atm/cam/src/physics/cam/wv_saturation.F90
M models/atm/cam/src/physics/cam/gw_drag.F90
M models/atm/cam/src/physics/cam/phys_buffer.F90
M models/atm/cam/src/physics/cam/zm_conv_intr.F90
M models/atm/cam/src/physics/cam/tphysac.F90
M models/atm/cam/src/physics/cam/zm_conv.F90
M models/atm/cam/src/physics/cam/eddy_diff.F90
M models/atm/cam/src/physics/cam/cldwat.F90
M models/atm/cam/src/physics/cam/vertical_diffusion.F90
M models/atm/cam/src/physics/rrtmg/radiation.F90
M models/atm/cam/src/physics/rrtmg/cloud_diagnostics.F90
M models/atm/cam/src/physics/rrtmg/cloud_rad_props.F90
- changes in support of new cam5 physics
M models/atm/cam/src/dynamics/sld/spegrd.F90
M models/atm/cam/src/dynamics/sld/linemsdyn.F90
M models/atm/cam/src/dynamics/sld/scanslt.F90
M models/atm/cam/src/dynamics/sld/dyn_comp.F90
M models/atm/cam/src/dynamics/sld/grmult.F90
M models/atm/cam/src/dynamics/sld/tfilt_massfix.F90
M models/atm/cam/src/dynamics/eul/lagyin.F90
M models/atm/cam/src/dynamics/eul/sphdep.F90
M models/atm/cam/src/dynamics/eul/dyn_comp.F90
M models/atm/cam/src/dynamics/eul/cubydr.F90
M models/atm/cam/src/dynamics/fv/initcom.F90
M models/atm/cam/src/dynamics/fv/dp_coupling.F90
M models/atm/cam/src/dynamics/fv/inidat.F90
M models/atm/cam/src/dynamics/fv/dyn_comp.F90
M models/atm/cam/src/dynamics/fv/inital.F90
M models/atm/cam/src/dynamics/fv/cd_core.F90
M models/atm/cam/src/dynamics/fv/sw_core.F90
M models/atm/cam/src/dynamics/fv/dynamics_vars.F90
M models/atm/cam/src/dynamics/fv/diag_dynvar_ic.F90
M models/atm/cam/src/dynamics/fv/uv3s_update.F90
M models/atm/cam/src/dynamics/fv/te_map.F90
M models/atm/cam/src/dynamics/fv/stepon.F90
M models/atm/cam/src/dynamics/fv/restart_dynamics.F90
- changes made in support of dynamics changes (see above)
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Tue Apr 27 21:40:53 MDT 2010
019 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...............................FAIL! rc= 7 at Tue Apr 27 22:00:13 MDT 2010
034 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ......................................FAIL! rc= 7 at Tue Apr 27 22:37:02 MDT 2010
038 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .....................................FAIL! rc= 7 at Tue Apr 27 22:56:00 MDT 2010
042 bl375 TBL.sh f1.9c5m outfrq24h+1850-2005_cam5 2d ......................FAIL! rc= 7 at Tue Apr 27 23:20:40 MDT 2010
045 bl374 TBL.sh fs1.9c4dh outfrq3s+1850-2005_cam4_super_fast_llnl 9s .....FAIL! rc= 7 at Tue Apr 27 23:40:52 MDT 2010
049 bl387 TBL.sh f1.9c4dm outfrq3s+1850-2005_cam4 9s ......................FAIL! rc= 7 at Tue Apr 27 23:57:19 MDT 2010
057 bl992 TBL_ccsm.sh f19_g16 E 2d ........................................FAIL! rc= 7 at Wed Apr 28 00:34:35 MDT 2010
- any baseline test using cam5 physics was expected to fail (bl137, bl332, bl337, bl371, bl375)
- bl374, bl387, and bl992 failed due to the fact that delta pressure (delp) is no longer being read from the
initial conditions file
**additional tests using cam4 physics
2deg 1850 and 1850-2005
1deg 1850 and 1850-2005
2deg waccm 1850 and 1850-2005
- 2-month tests vs baseline were b4b, when accounting for the fact that delta pressure (delp) is no longer being
read from the initial conditions file
edinburgh/lf95:
004 bl111 TBL.sh e8t5mdm ttrac 9s .........................................FAIL! rc= 7 at Tue Apr 27 21:45:29 MDT 2010
008 bl112 TBL.sh e8dm ghgrmp 9s ...........................................FAIL! rc= 7 at Tue Apr 27 21:56:21 MDT 2010
010 bl113 TBL.sh e8pdm aqpgro 3s ..........................................FAIL! rc= 7 at Tue Apr 27 22:04:06 MDT 2010
014 bl114 TBL.sh e8dm co2rmp 9s ...........................................FAIL! rc= 7 at Tue Apr 27 22:10:06 MDT 2010
021 bl311 TBL.sh f10t5mdm ttrac 9s ........................................FAIL! rc= 7 at Tue Apr 27 22:36:18 MDT 2010
029 bl312 TBL.sh f10dm ghgrmp 9s ..........................................FAIL! rc= 7 at Tue Apr 27 23:00:53 MDT 2010
031 bl313 TBL.sh f10pdm aqpgro 3s .........................................FAIL! rc= 7 at Tue Apr 27 23:09:37 MDT 2010
034 bl314 TBL.sh wg10dm outfrq3s 9s .......................................FAIL! rc= 7 at Tue Apr 27 23:34:04 MDT 2010
037 bl315 TBL.sh f10c3dm outfrq3s 9s ......................................FAIL! rc= 7 at Tue Apr 27 23:39:46 MDT 2010
- any baseline test using cam5 physics was expected to fail (bl111, bl112, bl113, bl114, bl311, bl312, bl313, bl314)
- bl315 failed due to 1) round-off differences introduced in calculation of
condensate-to-rain autoconversion coefficient (c0) and 2) the fact that delta pressure (delp)
is no longer being read from the initial conditions file
edinburgh/pgi:
004 bl111 TBL.sh e8t5mdm ttrac 9s .........................................FAIL! rc= 7 at Tue Apr 27 17:08:20 MDT 2010
008 bl112 TBL.sh e8dm ghgrmp 9s ...........................................FAIL! rc= 7 at Tue Apr 27 17:13:58 MDT 2010
010 bl113 TBL.sh e8pdm aqpgro 3s ..........................................FAIL! rc= 7 at Tue Apr 27 17:18:45 MDT 2010
014 bl114 TBL.sh e8dm co2rmp 9s ...........................................FAIL! rc= 7 at Tue Apr 27 17:20:24 MDT 2010
021 bl311 TBL.sh f10t5mdm ttrac 9s ........................................FAIL! rc= 7 at Tue Apr 27 17:30:19 MDT 2010
029 bl312 TBL.sh f10dm ghgrmp 9s ..........................................FAIL! rc= 7 at Tue Apr 27 17:38:10 MDT 2010
031 bl313 TBL.sh f10pdm aqpgro 3s .........................................FAIL! rc= 7 at Tue Apr 27 17:42:53 MDT 2010
034 bl314 TBL.sh wg10dm outfrq3s 9s .......................................FAIL! rc= 7 at Tue Apr 27 17:49:13 MDT 2010
037 bl315 TBL.sh f10c3dm outfrq3s 9s ......................................FAIL! rc= 7 at Tue Apr 27 17:53:02 MDT 2010
- any baseline test using cam5 physics was expected to fail (bl111, bl112, bl113, bl114, bl311, bl312, bl313, bl314)
- bl315 failed due to 1) round-off differences introduced in calculation of
condensate-to-rain autoconversion coefficient (c0) and 2) the fact that delta pressure (delp)
is no longer being read from the initial conditions file
jaguar:
004 bl157 TBL.sh e64m ghgrmp 9s ...........................................FAIL! rc= 7 at Wed Apr 28 11:15:15 EDT 2010
006 bl158 TBL.sh e64pm aqpgro 3s ..........................................FAIL! rc= 7 at Wed Apr 28 11:15:17 EDT 2010
009 bl159 TBL.sh e64am adia 9s ............................................FAIL! rc= 7 at Wed Apr 28 11:15:21 EDT 2010
012 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Wed Apr 28 11:15:29 EDT 2010
015 bl372 TBL.sh f1.9m fvvp_lb2 9s ........................................FAIL! rc= 7 at Wed Apr 28 11:15:56 EDT 2010
018 sm375 TSM.sh f1.9c5m outfrq24h+1850-2005_cam5 2d ......................FAIL! rc= 8 at Tue Apr 27 22:47:39 EDT 2010
019 er375 TER.sh f1.9c5m outfrq24h+1850-2005_cam5 1+1d ....................FAIL! rc= 5 at Tue Apr 27 22:47:39 EDT 2010
020 bl375 TBL.sh f1.9c5m outfrq24h+1850-2005_cam5 2d ......................FAIL! rc= 4 at Wed Apr 28 11:16:57 EDT 2010
023 bl378 TBL.sh fm1.9m outfrq3s 9s .......................................FAIL! rc= 7 at Wed Apr 28 11:17:33 EDT 2010
024 sm379 TSM.sh wm1.9m outfrq3s+waccm_1850_cam4 9s .......................FAIL! rc= 8 at Tue Apr 27 22:54:29 EDT 2010
025 er379 TER.sh wm1.9m outfrq3s+waccm_1850_cam4 4+5s .....................FAIL! rc= 5 at Tue Apr 27 22:54:29 EDT 2010
026 bl379 TBL.sh wm1.9m outfrq3s+waccm_1850_cam4 9s .......................FAIL! rc= 4 at Wed Apr 28 11:17:49 EDT 2010
030 bl383 TBL.sh fma1.9m outfrq3s+1850_cam4_trop_bam 9s ...................FAIL! rc= 7 at Wed Apr 28 11:18:13 EDT 2010
031 sm384 TSM.sh fmgpa1.9dm outfrq3s+1850_cam4 9s .........................FAIL! rc= 8 at Tue Apr 27 23:00:19 EDT 2010
032 er384 TER.sh fmgpa1.9dm outfrq3s+1850_cam4 4+5s .......................FAIL! rc= 5 at Tue Apr 27 23:00:20 EDT 2010
033 br384 TBR.sh fmgpa1.9dm outfrq3s+1850_cam4 6+3s .......................FAIL! rc= 5 at Tue Apr 27 23:00:31 EDT 2010
034 bl384 TBL.sh fmgpa1.9dm outfrq3s+1850_cam4 9s .........................FAIL! rc= 4 at Wed Apr 28 11:18:28 EDT 2010
037 bl386 TBL.sh f1.9c4m outfrq24h+1850_cam4 2m ...........................FAIL! rc= 7 at Wed Apr 28 11:22:08 EDT 2010
- '375' '379' and '384' test failures are pre-existing
- any baseline test using cam5 physics was expected to fail (bl157, bl158, bl137, bl372, bl378, bl383)
- bl159 failed due to change in default cam initial conditions file being used
- bl386 failed due to either 1) round-off differences introduced in calculation of
condensate-to-rain autoconversion coefficient (c0) or 2) the fact that delta pressure (delp)
is no longer being read from the initial conditions file, or both (this was not investigated)
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers:
cam5: answers change due to new climate
cam4: the change to no longer read delp from the initial file may produce answers beyond round-off; the change
to the setting of the condensate-to-rain autoconversion coefficient (c0) from a scalar to a computation
introduces a round-off level difference on some compilers (for example, edinburgh w/pgi and w/lahey)
If bitwise differences were observed, how did you show they were no worse
than roundoff?
- cam5 answere are b4b with end of the development branch - see the ChangeLog file for branch camdev_cam3_6_23 for
validation details
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- see the ChangeLog file for branch camdev_cam3_6_23 for validation details
===============================================================
===============================================================
Tag name: cam4_1_15
Originator(s): Francis Vitt
Date: 26 Apr 2010
One-line Summary: WACCM and CAM-Chem changes
Purpose of changes:
- correction to WACCM Fomichev code for time-dependent surface co2
- bug fixes for dry deposition calculations over ocean and ice when CLM dry deposition is used
- give user the ability to turn on/off aerosol deposition fluxes to surface components
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Brian Eaton
List all subroutines eliminated:
D models/atm/cam/src/chemistry/utils/ocnice_aero.F90
caused errors in dry dep over ocean and ice
List all subroutines added and what they do:
A models/atm/cam/bld/namelist_files/use_cases/2000_cam4_super_fast_llnl.xml
new use case added
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/configure
impose cam5 physics restriction on all MAM chemistry packages
M models/atm/cam/bld/build-namelist
- remove the ISOP emis from super_fast_llnl namelist when CLM MEGAN is used
- include trop_mam7 in with the 'atm_dep_flux' default namelist setting
- get default prescribed_aero settings unless prescribed_aero_file is specified
- set co2vmr for waccm unless scenario_ghg is 'RAMPED'
- set srf_emis_type='CYCLICAL' and srf_emis_ymd=19970101 unless srf_emis_type is specified
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- corrected spelling of 'category'
M models/atm/cam/bld/namelist_files/use_cases/waccm_1850-2005_cam4.xml
- cldfrc_rhminl set to 0.90
- specify scenario_ghg='RAMPED' and bndtvghg input file
- use IC file from 1850 control simulation
- use new solar_parms file which spans period from 1845-2008
- added TROPP_FD diagnostic to h0 history
M models/atm/cam/bld/namelist_files/use_cases/1850-2005_cam4_super_fast_llnl.xml
- use correct solar data input file
- cldfrc_rhminl set to 0.9225
- added 1-degree volc vert emis files
- history fields added to fincl1
M models/atm/cam/bld/namelist_files/use_cases/waccm_1850_cam4.xml
- cldfrc_rhminl set to 0.90
- strat_aero_feedback=.true. added for volcanoes
M models/atm/cam/bld/namelist_files/use_cases/1850_cam4_super_fast_llnl.xml
- cleanup -- rearrange
M models/atm/cam/src/control/camsrfexch_types.F90
- initializing aero dep fluxes to shr_const_spval seems to have no effect -- this is removed
M models/atm/cam/src/utils/time_manager.F90
- work around for ESMF lib problem when time is negative
M models/atm/cam/src/cpl_mct/atm_comp_mct.F90
M models/atm/cam/src/cpl_esmf/atm_comp_esmf.F90
- set driver atm_aero flag to atm_dep_flux cam namelist
M models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
M models/atm/cam/src/chemistry/mozart/mo_chemini.F90
M models/atm/cam/src/chemistry/mozart/mo_drydep.F90
- corrections to dry dep over non-land regions
M models/atm/cam/src/physics/waccm/nlte_lw.F90
M models/atm/cam/src/physics/waccm/nlte_fomichev.F90
- set matrice amat and bmat every timestep since co2vmr at the surface can be time-dependent
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: all PASS except
031 bl336 TBL.sh wm4c4h outfrq3s+waccm_1850_cam4 9s .......................FAIL! rc= 7 at Mon Apr 26 14:57:40 MDT 2010
045 bl374 TBL.sh fs1.9c4dh outfrq3s+1850-2005_cam4_super_fast_llnl 9s .....FAIL! rc= 7 at Mon Apr 26 16:10:41 MDT 2010
- these are expected failures due to changes in the waccm_1850_cam4 and 1850-2005_cam4_super_fast_llnl use cases
edinburgh/lf95: all PASS
edinburgh/pgi: all PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam4_1_14
Originator(s): mvr
Date: Wed Apr 21 2010
One-line Summary: updated externals of surface components to
those of ccsm4_0_beta48
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets:
- update CLM datasets to be consistent with the clm3_7_10 tag
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- update CLM datasets to be consistent with the clm3_7_10 tag
M SVN_EXTERNAL_DIRECTORIES
M .
- updated external definitions
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: all PASS except
004 bl132 TBL.sh e48c4dh ghgrmp+1850_cam4 9s ..............................FAIL! rc= 7 at Wed Apr 21 10:01:48 MDT 2010
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 7 at Wed Apr 21 10:11:11 MDT 2010
015 bl331 TBL.sh f4c4dh co2rmp+1850_cam4 9s ...............................FAIL! rc= 7 at Wed Apr 21 10:15:58 MDT 2010
019 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...............................FAIL! rc= 7 at Wed Apr 21 10:24:06 MDT 2010
031 bl336 TBL.sh wm4c4h outfrq3s+waccm_1850_cam4 9s .......................FAIL! rc= 7 at Wed Apr 21 10:38:12 MDT 2010
034 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ......................................FAIL! rc= 7 at Wed Apr 21 10:45:25 MDT 2010
038 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .....................................FAIL! rc= 7 at Wed Apr 21 10:57:47 MDT 2010
042 bl375 TBL.sh f1.9c5m outfrq24h+1850-2005_cam5 2d ......................FAIL! rc= 7 at Wed Apr 21 11:13:11 MDT 2010
045 bl374 TBL.sh fs1.9c4dh outfrq3s+1850-2005_cam4_super_fast_llnl 9s .....FAIL! rc= 7 at Wed Apr 21 11:13:21 MDT 2010
049 bl387 TBL.sh f1.9c4dm outfrq3s+1850-2005_cam4 9s ......................FAIL! rc= 7 at Wed Apr 21 11:13:31 MDT 2010
057 bl992 TBL_ccsm.sh f19_g16 E 2d ........................................FAIL! rc= 7 at Wed Apr 21 11:13:38 MDT 2010
- baselines (other than aquaplanet, ideal physics, adiabatic) were expected to fail due to change in clm and cice
edinburgh/lf95: all PASS except
004 bl111 TBL.sh e8t5mdm ttrac 9s .........................................FAIL! rc= 7 at Tue Apr 20 16:45:42 MDT 2010
008 bl112 TBL.sh e8dm ghgrmp 9s ...........................................FAIL! rc= 7 at Tue Apr 20 16:45:43 MDT 2010
014 bl114 TBL.sh e8dm co2rmp 9s ...........................................FAIL! rc= 7 at Tue Apr 20 16:45:46 MDT 2010
021 bl311 TBL.sh f10t5mdm ttrac 9s ........................................FAIL! rc= 7 at Tue Apr 20 16:45:47 MDT 2010
029 bl312 TBL.sh f10dm ghgrmp 9s ..........................................FAIL! rc= 7 at Tue Apr 20 16:45:48 MDT 2010
034 bl314 TBL.sh wg10dm outfrq3s 9s .......................................FAIL! rc= 7 at Tue Apr 20 17:11:02 MDT 2010
037 bl315 TBL.sh f10c3dm outfrq3s 9s ......................................FAIL! rc= 7 at Tue Apr 20 17:16:49 MDT 2010
- baselines (other than aquaplanet, ideal physics, adiabatic) were expected to fail due to change in clm and cice
edinburgh/pgi: all PASS except
004 bl111 TBL.sh e8t5mdm ttrac 9s .........................................FAIL! rc= 7 at Tue Apr 20 13:02:10 MDT 2010
008 bl112 TBL.sh e8dm ghgrmp 9s ...........................................FAIL! rc= 7 at Tue Apr 20 13:08:08 MDT 2010
014 bl114 TBL.sh e8dm co2rmp 9s ...........................................FAIL! rc= 7 at Tue Apr 20 13:15:38 MDT 2010
021 bl311 TBL.sh f10t5mdm ttrac 9s ........................................FAIL! rc= 7 at Tue Apr 20 13:29:08 MDT 2010
029 bl312 TBL.sh f10dm ghgrmp 9s ..........................................FAIL! rc= 7 at Tue Apr 20 13:37:28 MDT 2010
034 bl314 TBL.sh wg10dm outfrq3s 9s .......................................FAIL! rc= 7 at Tue Apr 20 13:49:42 MDT 2010
037 bl315 TBL.sh f10c3dm outfrq3s 9s ......................................FAIL! rc= 7 at Tue Apr 20 13:53:19 MDT 2010
- baselines (other than aquaplanet, ideal physics, adiabatic) were expected to fail due to change in clm and cice
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: larger than roundoff but same climate - see clm ChangeLog for details
===============================================================
Tag name: cam4_1_13
Originator(s): mvr
Date: Tue Apr 20 2010
One-line Summary: corrected solar input file for use_cases with super_fast chem;
fix for pio bug introduced with previous tag; enhancement/cleanup of test scripts;
updated run template scripts and added one for cray xt's
test scripts/files were modified to:
- enhance testing of eul dycore at T31
- explicitly test cam5 physics rather than relying on default
- explicitly test cam4 physics (replacing cam3_5_1)
- add mpi-only tests to bluefire
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
- new solar input files for super_fast chemistry
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
D models/atm/cam/test/system/config_files/e32c4pdm
D models/atm/cam/test/system/config_files/f4idh
D models/atm/cam/test/system/config_files/e64h
D models/atm/cam/test/system/config_files/e32adh
D models/atm/cam/test/system/config_files/e32idh
D models/atm/cam/test/system/config_files/f4dh
D models/atm/cam/test/system/config_files/f1.9c4dh
D models/atm/cam/test/system/config_files/f1.9c351h
D models/atm/cam/test/system/config_files/f1.9c351m
D models/atm/cam/test/system/config_files/e32c4dh
D models/atm/cam/test/system/tests_posttag_lightning
- removed test files that were no longer being used
List all subroutines added and what they do:
A models/atm/cam/test/system/config_files/e48c4pdm
A models/atm/cam/test/system/config_files/e48c4dh
A models/atm/cam/test/system/config_files/f4idm
A models/atm/cam/test/system/config_files/e48m
A models/atm/cam/test/system/config_files/f4c5dm
A models/atm/cam/test/system/config_files/s48dh
A models/atm/cam/test/system/config_files/f4dm
A models/atm/cam/test/system/config_files/s48adh
A models/atm/cam/test/system/config_files/s48idh
A models/atm/cam/test/system/config_files/f1.9c4h
A models/atm/cam/test/system/config_files/f1.9c4m
A models/atm/cam/test/system/config_files/f1.9c4dm
A models/atm/cam/test/system/config_files/e48c5h
A models/atm/cam/test/system/config_files/e48adh
A models/atm/cam/test/system/config_files/f1.9c5m
A models/atm/cam/test/system/config_files/e48idh
A models/atm/cam/test/system/config_files/s48pdh
- new files to be used by test scripts
A models/atm/cam/test/system/CAM_decomp.sh
- new test utility to build fv 2d decomp string if required for test (needed to
enable broader testing of mpi-only configurations)
A models/atm/cam/bld/run-cray.csh
- new run template script for cray platforms
A models/atm/cam/bld/namelist_files/use_cases/1850-2005_cam5.xml
A models/atm/cam/bld/namelist_files/use_cases/1850_cam5.xml
- new use_cases brought over from cam5 branch to enable testing
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/tests_pretag_bluefire
M models/atm/cam/test/system/tests_posttag_jaguar_cb
M models/atm/cam/test/system/tests_posttag_jaguar
M models/atm/cam/test/system/tests_posttag_calgary
M models/atm/cam/test/system/tests_pretag_calgary
- modified to reflect changes in test names
M models/atm/cam/test/system/TCB.sh
- hardwired cice_maxblocks=4 to enable testing a wide variety of configurations
while limiting the number of model builds
M models/atm/cam/test/system/config_files/wm4c4h
M models/atm/cam/test/system/config_files/wm4dh
M models/atm/cam/test/system/config_files/wm1.9m
M models/atm/cam/test/system/config_files/wg10dm
- explicitly set dycore to fv - was relying on the default
M models/atm/cam/test/system/TSM.sh
M models/atm/cam/test/system/TBR.sh
M models/atm/cam/test/system/TER.sh
- now calls CAM_decomp.sh to build fv 2d decomp string for namelist if needed
M models/atm/cam/test/system/TPF.sh
- change to how test results are written to log file
M models/atm/cam/test/system/test_driver.sh
- edinburgh now tests on 8tsks; jaguar now tests on 64 processors and default version of
pgi compilers; modified format of test results
M models/atm/cam/test/system/input_tests_master
- modified with new test definitions
M models/atm/cam/test/system/CAM_runcmnd.sh
- improved how hybrid runs are set up for testing; improved the changing of tasks/threads
on restart tests
M models/atm/cam/bld/run-pc.csh
M models/atm/cam/bld/run-ibm.csh
- updated run template scripts to work with recent changes to model
MM models/atm/cam/bld/namelist_files/use_cases/1850-2005_cam4_super_fast_llnl.xml
M models/atm/cam/bld/namelist_files/use_cases/1850_cam4_super_fast_llnl.xml
- fix for use_cases using super_fast chemistry to use correct solar input file
M models/atm/cam/src/utils/cam_dom/sst_data.F90
- fix for pio bug introduced with previous tag
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: all PASS except
004 bl132 TBL.sh e48c4dh ghgrmp+1850_cam4 9s ..............................FAIL! rc= 5 at Tue Apr 20 09:01:21 MDT 2010
006 bl133 TBL.sh e48c4pdm aqpgro 3s .......................................FAIL! rc= 5 at Tue Apr 20 09:01:21 MDT 2010
009 bl134 TBL.sh e48adh adia 9s ...........................................FAIL! rc= 5 at Tue Apr 20 09:01:22 MDT 2010
011 bl137 TBL.sh e48c5h outfrq24h 2d ......................................FAIL! rc= 5 at Tue Apr 20 09:01:23 MDT 2010
019 bl332 TBL.sh f4c5dm ghgrmp+1850_cam5 9s ...............................FAIL! rc= 5 at Tue Apr 20 09:01:24 MDT 2010
028 bl335 TBL.sh f4idm idphys 9s ..........................................FAIL! rc= 5 at Tue Apr 20 09:01:25 MDT 2010
042 bl375 TBL.sh f1.9c5m outfrq24h+1850-2005_cam5 2d ......................FAIL! rc= 5 at Tue Apr 20 09:55:11 MDT 2010
045 bl374 TBL.sh fs1.9c4dh outfrq3s+1850-2005_cam4_super_fast_llnl 9s .....FAIL! rc= 7 at Tue Apr 20 10:14:54 MDT 2010
049 bl387 TBL.sh f1.9c4dm outfrq3s+1850-2005_cam4 9s ......................FAIL! rc= 5 at Tue Apr 20 10:27:22 MDT 2010
- bl374 failed due to new solar input file for super_fast chemistry
- all other baselines failures were due to them being new tests where the configuration was not recognizable in
the baseline test scripts
edinburgh/lf95: all PASS except
021 bl311 TBL.sh f10t5mdm ttrac 9s ........................................FAIL! rc= 5 at Mon Apr 19 17:58:14 MDT 2010
029 bl312 TBL.sh f10dm ghgrmp 9s ..........................................FAIL! rc= 5 at Mon Apr 19 18:16:56 MDT 2010
031 bl313 TBL.sh f10pdm aqpgro 3s .........................................FAIL! rc= 5 at Mon Apr 19 18:24:09 MDT 2010
034 bl314 TBL.sh wg10dm outfrq3s 9s .......................................FAIL! rc= 5 at Mon Apr 19 18:43:13 MDT 2010
037 bl315 TBL.sh f10c3dm outfrq3s 9s ......................................FAIL! rc= 5 at Mon Apr 19 18:48:00 MDT 2010
- baselines at 10x15 failed due to baseline test scripts not having new functionality to build fv 2d decomp namelist setting
edinburgh/pgi: all PASS except
021 bl311 TBL.sh f10t5mdm ttrac 9s ........................................FAIL! rc= 5 at Mon Apr 19 17:28:51 MDT 2010
029 bl312 TBL.sh f10dm ghgrmp 9s ..........................................FAIL! rc= 5 at Mon Apr 19 17:36:04 MDT 2010
031 bl313 TBL.sh f10pdm aqpgro 3s .........................................FAIL! rc= 5 at Mon Apr 19 17:41:24 MDT 2010
034 bl314 TBL.sh wg10dm outfrq3s 9s .......................................FAIL! rc= 5 at Mon Apr 19 17:47:14 MDT 2010
037 bl315 TBL.sh f10c3dm outfrq3s 9s ......................................FAIL! rc= 5 at Mon Apr 19 17:50:43 MDT 2010
- baselines at 10x15 failed due to baseline test scripts not having new functionality to build fv 2d decomp namelist setting
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB, except for super_fast chemistry
===============================================================
Tag name: cam4_1_12
Originator(s): Jim Edwards
Date: 04-14-2010
One-line Summary: chem port to pio, homme restart fix
Purpose of changes: Removing depreciated wrap_nf code and
replacing it with pio.
Bugs fixed (include bugzilla ID): homme restart was not bfb when
task count changed after restart
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: self
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
models/atm/cam/src/utils/cam_dom/ocn_spmd.F90
models/atm/cam/src/utils/cam_dom/sst_data.F90
models/atm/cam/src/utils/cam_dom/ocn_comp.F90
models/atm/cam/src/chemistry/utils/solar_data.F90
models/atm/cam/src/chemistry/mozart/chlorine_loading_data.F90
models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
models/atm/cam/src/chemistry/mozart/mo_photoin.F90
models/atm/cam/src/chemistry/mozart/spedata.F90
models/atm/cam/src/chemistry/mozart/mo_seto2.F90
models/atm/cam/src/chemistry/mozart/mo_jshort.F90
models/atm/cam/src/chemistry/mozart/mo_chm_diags.F90
models/atm/cam/src/chemistry/mozart/mo_extfrc.F90
models/atm/cam/src/chemistry/mozart/mo_snoe.F90
models/atm/cam/src/chemistry/mozart/euvac.F90
models/atm/cam/src/chemistry/mozart/mo_drydep.F90
models/atm/cam/src/chemistry/mozart/iondrag.F90
models/atm/cam/src/chemistry/mozart/mo_photo.F90
models/atm/cam/src/chemistry/mozart/mo_tuv_inti.F90
models/atm/cam/src/chemistry/mozart/mo_srf_emissions.F90
models/atm/cam/src/dynamics/fv/metdata.F90
replaced serial netcdf io functions with pio equivalents.
models/atm/cam/src/dynamics/homme/external/global_norms_mod.F90
models/atm/cam/src/dynamics/homme/external/prim_driver_mod.F90
models/atm/cam/src/dynamics/homme/external/viscosity_mod.F90
models/atm/cam/src/dynamics/homme/inidat.F90
replaced an internal homme global average function with the
task count independent cam version in repro_sum_mod
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: all pass except
052 bl731 TBL.sh h16x4dm aqua 9s
due to roundoff level change in global sum calculation
edinburgh/lf95: all pass
calgary/pgi or jaguar/pgi: all pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam4_1_11
Originator(s): fvitt, aconley
Date: 12 Apr 2010
One-line Summary: Correction to the solar spectral scaling in CAMRT used by WACCM.
Purpose of changes:
Corrected solar band spectral scaling in CAMRT used in WACCM
and super_fast_llnl simulations. Added history fields for
off-line radiation calculations. Band-level scaling is now
computed relative to a reference spectrum of 4 solar cycles
rather than a reference from the CAMRT code.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
Namelist variable "solar_htng_spctrl_scl" added. This is a switch that can be used to
turn on/off the solar band spectral heating. This is set to TRUE by default for waccm_mozart
or when radiation package is RRTMG, otherwise it is set to FALSE
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do:
A models/atm/cam/src/physics/rrtmg/rad_solar_var.F90
Manages the solar spectral scaling specific to RRTMG radiation package.
This is the same as the previous version, just moved to the rrtmg directory.
Band level scaling in RRTMG continues to be relative to the base Kurucz
spectrum assumed by RRTMG.
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/build-namelist
- solar_cnst explicitly set to -9999. when solar_data_file is specified
- namelist var solar_htng_spctrl_scl to TRUE for waccm_mozart or when rad package is rrtmg
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- change in default solar_data_file for waccm_mozart
- solar_photons_files default setting removed
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- solar_htng_spctrl_scl namelist variable added
- rad_data_avgflag namelist variable added
- rad_data_output and rad_data_histfile_num namelist variables moved from rad_cnst_nl to rad_data_nl
M models/atm/cam/bld/namelist_files/use_cases/waccm_1850-2005_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1850_cam4.xml
- change in solar_data_file input for these waccm use cases
M models/atm/cam/src/control/runtime_opts.F90
- rad_data_readnl call added
M models/atm/cam/src/physics/cam/radiation.F90
- init_rad_data moved from physpkg
- output_rad_data call moved from tphysbc
- FSDTOA diagnostic field added to history
M models/atm/cam/src/physics/cam/tphysbc.F90
- output_rad_data call moved to radiation.F90
M models/atm/cam/src/physics/cam/radsw.F90
- calculate the column value of the fsdtoa diagnostic field
- correction to the solar band scaling
M models/atm/cam/src/physics/cam/radconstants.F90
- wavemin_true data added
- correction to the get_sw_spectral_boundaries subroutine
M models/atm/cam/src/physics/cam/radiation_data.F90
- output of the rad constituents moved from rad_constituents
- solar zeneth weighted surface fields added
- code cleanup
M models/atm/cam/src/physics/cam/physpkg.F90
- init_rad_data moved to radiation.F90
M models/atm/cam/src/physics/cam/rad_constituents.F90
- output of the rad constituents moved to radiation_data.F90
- rad_data_output and rad-data_histfile_num namelist vars moved to radiation_data
- rad_cnst_get_clim_info subroutine was extended to provide more information
M models/atm/cam/src/physics/cam/rad_solar_var.F90
- reference spectrum used in the solar band scaling is now based on a
reference spectrum provided in the solar input dataset rather than a spectrum
provided by CAMRT
- set band scaling factors to 1.0 for TSI solar heating
M models/atm/cam/src/physics/rrtmg/radiation.F90
- init_rad_data moved from physpkg
- output_rad_data call moved from tphysbc
M models/atm/cam/src/chemistry/utils/mo_util.F90
- constant "0." changed to "0._r8"
M models/atm/cam/src/chemistry/utils/solar_data.F90
- solar_htng_spctrl_scl namelist variable added. This is a switch that can be used to
turn on/off the solar band spectral heating.
- endrun called if both solar_const and solar_data_file are both specified
- read both tsi and ssi if the input dataset contains both
- read in a reference spectrum (ref_ssi) and reference TSI (ref_tsi)
- correction to the units of the solar photon fluxes used by photolysis
M models/atm/cam/src/chemistry/mozart/mo_jlong.F90
- comments added to clearly specify units of the input solar fluxes
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
031 bl336 TBL.sh wm4c4h outfrq3s+waccm_1850_cam4 9s ...............FAIL! rc= 7
This is an expected failure due to changes in the band-level solar spectral scaling in
CAMRT for waccm_mozart.
045 bl374 TBL.sh fs1.9c4dh outfrq3s+1850-2005_cam4_super_fast_llnl 9s FAIL! rc= 7
This is an expected failure due to change to use of TSI in CAMRT heating rather than
band-level solar spectral scaling for super_fast_llnl chemistry configuration.
edinburgh/lf95:
td.1285.status:034 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 7
This is an expected failure due to change to use of TSI in CAMRT heating rather than
band-level solar spectral scaling for waccm_ghg chemistry configuration.
edinburgh/pgi:
td.1282.status:034 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 7
This is an expected failure due to change to use of TSI in CAMRT heating rather than
band-level solar spectral scaling for waccm_ghg chemistry configuration.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam4_1_10
Originator(s): eaton
Date: Fri Apr 9 09:47:41 MDT 2010
One-line Summary: update for carbon cycle runs
Purpose of changes:
. Updates to configure and build-namelist to facilitate "out of the box"
capability for carbon cycle runs. When the configure option -co2_cycle
is set, build-namelist will by default set the flags to turn on the
co2_cycle code in CAM (adds 4 tracers) and will set the flag to read the
fossil fuel emissions dataset. Note that the code that reads the
emission dataset doesn't do any spatial interpolation. We currently only
have the required dataset at fv 0.9x1.25 resolution.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets:
. add default fossil fuel emissions dataset:
inputdata/atm/cam/ggas/co2flux_fossil_1751-2006-monthly_0.9x1.25_c20100204.nc
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/configure
. remove the check that only allows the -co2_cycle option to be specified
with -ccsm_seq. This option can be run in standalone mode to test the
emission and flux datasets.
models/atm/cam/bld/build-namelist
. add section to set defaults for co2_flag, co2_readflux_fuel, and
co2flux_fuel_file when co2_cycle is true.
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. add default for co2flux_fuel_file
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS.
edinburgh/lf95: All PASS.
edinburgh/pgi: All PASS.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam4_1_09
Originator(s): eaton
Date: Wed Apr 7 16:51:52 MDT 2010
One-line Summary: add rcp26 and rcp45 use cases; change default fields on
initial file; misc cleanup
Purpose of changes:
. Add use cases for rcp2.6 and rcp4.5 runs.
. Change default value of inithist_all to .false.. This affects the
writing of extra fields to the initial file that are used in CAPT
experiments.
. Change some defaults in the FV communications per recommendation of Pat
Worley.
. Misc cleanup.
Bugs fixed (include bugzilla ID):
. Fix build-namelist deal correctly with the case when both -ignore_ic_year
and -ignore_ic_date are specified. You now get the -ignore_ic_date
behavior from this setting. This fix was needed for ncdata. It was
implemented correctly for finidat (used by cam standalone).
Describe any changes made to build system: none
Describe any changes made to the namelist:
. Change default value of inithist_all to .false..
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do:
models/atm/cam/bld/namelist_files/use_cases/2005-2100_cam4_rcp26.xml
models/atm/cam/bld/namelist_files/use_cases/2005-2100_cam4_rcp45.xml
. update the GHG, aerosol, and ozone datasets
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/cam.cpl7.template
. change \$BLDROOT/Makefile to \$CASETOOLS/Makefile. (requested by Tony)
models/atm/cam/bld/configure
. add check_fc test
. don't try to run esmf test in spmd mode
models/atm/cam/bld/Makefile.in
. remove old platforms: nec-sx6, cray-x1, earth_simulator
models/atm/cam/bld/namelist_files/namelist_definition.xml
. change documentation for inithist_all default to .false.
. change "catagory" to "category".
. remove bndtvsf6
models/atm/cam/bld/perl5lib/Build/NamelistDefinition.pm
. change "catagory" to "category".
models/atm/cam/src/control/filenames.F90
. remove bndtvsf6
models/atm/cam/src/control/ncdio_atm.F90
. change "warning" to "info" in logfile message.
models/atm/cam/src/control/runtime_opts.F90
. change documentation for inithist_all default to .false.
. remove bndtvsf6
models/atm/cam/src/dynamics/fv/diag_dynvar_ic.F90
models/atm/cam/src/dynamics/fv/dyn_comp.F90
. only add DELP to IC file when inithist_all is true.
models/atm/cam/src/dynamics/fv/inidat.F90
. don't issue warning if DELP isn't found on IC file
models/atm/cam/src/dynamics/fv/spmd_dyn.F90
. change default of modc_onetwo from 1 to 2.
. change default of modc_tracers from 0 to 3.
models/atm/cam/src/physics/cam/cam_diagnostics.F90
. change default for inithist_all to .false.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS.
edinburgh/lf95: All PASS.
edinburgh/pgi: All PASS.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam4_1_08
Originator(s): Francis Vitt
Date: 19 March 2010
One-line Summary: WACCM features added and bug fixes
Purpose of changes:
Added new features and bug fixes:
- 'INTERP_MISSING_MONTHS' capability added to WACCM upper boundary
- new WACCM QBO capability added
- fix a bug in the turbulent mountain stress (TMS) module
- fix for QBO problem introduced in cam4_1_05 (uzm was not set)
- fix a bug in CLY family diagnostic
- switch added to age of air tracers to read from IC file
- clean up of waccm use cases
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Brian Eaton
List all subroutines eliminated:
D models/atm/cam/bld/namelist_files/use_cases/waccm_1953_ramped_qbo.xml
D models/atm/cam/bld/namelist_files/use_cases/waccm_1995_climo.xml
D models/atm/cam/bld/namelist_files/use_cases/waccm_1995_smin.xml
D models/atm/cam/bld/namelist_files/use_cases/waccm_1850_cntrl.xml
D models/atm/cam/bld/namelist_files/use_cases/waccm_1995_smax.xml
D models/atm/cam/bld/namelist_files/use_cases/waccm_1950_smin.xml
D models/atm/cam/bld/namelist_files/use_cases/waccm_1950_smax.xml
D models/atm/cam/bld/namelist_files/use_cases/waccm_1950_ramped.xml
cleaned out deprecated waccm use cases
D models/atm/cam/src/chemistry/mozart/mo_drydep_tables.F90
this was duplicated in the drv/shr code and is no longer needed
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
U models/atm/cam/SVN_EXTERNAL_DIRECTORIES
update in the chemistry pre-processor
U models/atm/cam/bld/build-namelist
change default waccm_ghg namelist settings
U models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
tgcm upper boundary file name change
U models/atm/cam/bld/namelist_files/namelist_definition.xml
age of air tracer namelist option added
U models/atm/cam/bld/namelist_files/use_cases/waccm_1974-2005_cam4.xml
U models/atm/cam/bld/namelist_files/use_cases/waccm_refb1.xml
U models/atm/cam/bld/namelist_files/use_cases/waccm_1850_cam4.xml
U models/atm/cam/bld/namelist_files/use_cases/waccm_1850-2005_cam4.xml
waccm use cases corrections
U models/atm/cam/src/physics/waccm/qbo.F90
U models/atm/cam/src/control/wrap_nf.F90
functionality added to QBO code to operate over long time spans
U models/atm/cam/src/physics/cam/molec_diff.F90
U models/atm/cam/src/physics/cam/trb_mtn_stress.F90
TMS bug fix
U models/atm/cam/src/chemistry/mozart/chemistry.F90
U models/atm/cam/src/physics/cam/vertical_diffusion.F90
U models/atm/cam/src/physics/waccm/nlte_lw.F90
U models/atm/cam/src/physics/cam/advnce.F90
U models/atm/cam/src/physics/cam/upper_bc.F90
U models/atm/cam/src/chemistry/mozart/mo_tgcm_ubc.F90
ability to interpolate over long time spans added to upper boundary
U models/atm/cam/src/physics/cam/aoa_tracers.F90
namelist option added
U models/atm/cam/src/chemistry/utils/tracer_data.F90
U models/atm/cam/src/chemistry/mozart/mo_chm_diags.F90
bug fix in CLY diagnostic
U models/atm/cam/src/dynamics/fv/ctem.F90
U models/atm/cam/src/dynamics/fv/dp_coupling.F90
changes needed for QBO bug fix introduced in cam4_1_05
U models/atm/cam/src/chemistry/mozart/mo_drydep.F90
- bug fixes and code clean up
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
019 bl332 TBL.sh f4dh ghgrmp 9s ...................................FAIL! rc= 7 at Fri Mar 19 12:11:59 MDT 2010
031 bl336 TBL.sh wm4c4h outfrq3s+waccm_1850_cam4 9s ...............FAIL! rc= 7 at Fri Mar 19 12:35:46 MDT 2010
034 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ..............................FAIL! rc= 7 at Fri Mar 19 12:48:13 MDT 2010
038 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .............................FAIL! rc= 7 at Fri Mar 19 13:06:04 MDT 2010
042 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7 at Fri Mar 19 13:28:04 MDT 2010
045 bl374 TBL.sh fs1.9c4dh outfrq3s+1850-2005_cam4_super_fast_llnl 9s FAIL! rc= 7 at Fri Mar 19 13:48:08 MDT 2010
These are expected failures. The waccm baseline fails due to change in upper boundary code and dry deposition.
The other chemistry configurations fail due to change in dry deposition.
edinburgh/lf95:
004 bl111 TBL.sh e8t5mdm ttrac 9s .................................FAIL! rc= 7 at Fri Mar 19 11:26:34 MDT 2010
008 bl112 TBL.sh e8dm ghgrmp 9s ...................................FAIL! rc= 7 at Fri Mar 19 11:32:21 MDT 2010
010 bl113 TBL.sh e8pdm aqpgro 3s ..................................FAIL! rc= 7 at Fri Mar 19 11:38:23 MDT 2010
014 bl114 TBL.sh e8dm co2rmp 9s ...................................FAIL! rc= 7 at Fri Mar 19 11:40:10 MDT 2010
021 bl311 TBL.sh f10t5mdm ttrac 9s ................................FAIL! rc= 7 at Fri Mar 19 11:50:58 MDT 2010
029 bl312 TBL.sh f10dm ghgrmp 9s ..................................FAIL! rc= 7 at Fri Mar 19 11:59:49 MDT 2010
031 bl313 TBL.sh f10pdm aqpgro 3s .................................FAIL! rc= 7 at Fri Mar 19 12:04:56 MDT 2010
040 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7 at Fri Mar 19 12:18:26 MDT 2010
These are expected failures. The chemistry configurations fail the baseline tests due to change in dry deposition.
edinburgh/pgi:
004 bl111 TBL.sh e8t5mdm ttrac 9s .................................FAIL! rc= 7 at Fri Mar 19 11:26:34 MDT 2010
008 bl112 TBL.sh e8dm ghgrmp 9s ...................................FAIL! rc= 7 at Fri Mar 19 11:32:21 MDT 2010
010 bl113 TBL.sh e8pdm aqpgro 3s ..................................FAIL! rc= 7 at Fri Mar 19 11:38:23 MDT 2010
014 bl114 TBL.sh e8dm co2rmp 9s ...................................FAIL! rc= 7 at Fri Mar 19 11:40:10 MDT 2010
021 bl311 TBL.sh f10t5mdm ttrac 9s ................................FAIL! rc= 7 at Fri Mar 19 11:50:58 MDT 2010
029 bl312 TBL.sh f10dm ghgrmp 9s ..................................FAIL! rc= 7 at Fri Mar 19 11:59:49 MDT 2010
031 bl313 TBL.sh f10pdm aqpgro 3s .................................FAIL! rc= 7 at Fri Mar 19 12:04:56 MDT 2010
040 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7 at Fri Mar 19 12:18:26 MDT 2010
These are expected failures. The chemistry configurations fail the baseline tests due to change in dry deposition.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam4_1_07
Originator(s): eaton
Date: Sat Mar 13 11:33:42 MST 2010
One-line Summary: add dust tuning param to namelist; misc configure/build-namelist mods
Purpose of changes:
. Make the dust emission tuning parameter namelist accessable.
. Modify the dust emission tuning parameter for trop_bam only.
. Remove any remaining references to track1 from the CAM regression
tests.
. The CLM deposition velocity calc is a runtime setting. Remove the switch
from configure and require it to be set via build-namelist.
. Add error trapping in build-namelist to protect against the user setting
tms_orocnst without setting do_tms.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
. Add dust_emis_fact to new namelist group aerosol_nl. This is the dust
emissions tuning factor.
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/input_tests_master
. replace track1 by cam4 (use case names have previously been updated)
M models/atm/cam/bld/configure
. remove -lnd_drydep switch
. change -lnd_vocsrc to -clm_vocsrc, and only process when land model is
CLM.
. made verbose mode output more consistent.
M models/atm/cam/bld/config_files/definition.xml
. remove lnd_drydep
. change lnd_vocsrc to clm_vocsrc
M models/atm/cam/bld/build-namelist
. add dust_emis_fact to namelist when dust is prognostic -- default value
depends on the value of do_tms
. change lnd_vocsrc to clm_vocsrc
. remove dependence of drydep_method on the lnd_drydep configure parameter
which was removed. The default for drydep_method is set in
namelist_defaults_cam.xml. To use deposition velocities from CLM the
user specifies drydep_method='xactive_lnd' in the namelist.
. add check that do_tms has been set if tms_oroconst has been set.
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. add default values for dust_emis_fact -- allow them to depend on attribute tms
. add default for drydep_method
M models/atm/cam/bld/namelist_files/namelist_definition.xml
. add dust_emis_fact to namelist group aerosol_nl
. move soil_erod to namelist group aerosol_nl
M models/atm/cam/src/control/filenames.F90
. remove soil_erod from module data
M models/atm/cam/src/physics/cam/vertical_diffusion.F90
. add log output for do_tms and tms_orocnst
M models/atm/cam/src/chemistry/bulk_aero/progseasalts_intr.F90
. fixed log output to only be written from masterproc
M models/atm/cam/src/chemistry/bulk_aero/dust_intr.F90
. declare namelist variables as module data
. set namelist variables by passing them through the arg list of
dust_initialize.
. add soil_erod to module data
. set namelist variables by passing them through the arg list of
dust_initialize.
. in dust_emis_intr set the variable "fact" from the namelist setting
dust_emis_fact. Remove conditional logic for setting fact since this
will now be done by build-namelist.
M models/atm/cam/src/chemistry/bulk_aero/aerosol_intr.F90
. declare namelist variables as module data
. add aerosol_readnl to read namelist group aerosol_nl
. add soil_erod to module data and to namelist group aerosol_nl
. add soil_erod to calling args for dust_initialize
M models/atm/cam/src/chemistry/mozart/chemistry.F90
. call aerosol_readnl from the chem_readnl method
. remove soil_erod from chem_inparm namelist group
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS
edinburgh/lf95: All PASS
edinburgh/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB except trop_bam
===============================================================
===============================================================
Tag name: cam4_1_06
Originator(s): eaton
Date: Tue Mar 9 08:08:52 MST 2010
One-line Summary: finish "track1" to "cam4" conversion; update externals
Purpose of changes:
. Finish cleanup of use cases that contain "track1" in name. All
references to "track1" have been replaced by "cam4".
. comment out the 'dpcoup dqreq' and 'dpcoup cant adjust' log messages.
They're making the output log file unusable.
. remove old utilities for SOM and reduced grid
. update externals
. update the ccsm build template to recognize the USE_MPISERIAL environment
variable set by the ccsm scripts
Bugs fixed (include bugzilla ID):
. change attribute "cell_method" to "cell_methods" to be compliant with CF
conventions.
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
models/atm/cam/bld/namelist_files/use_cases/1850_track1_2xco2.xml
models/atm/cam/bld/namelist_files/use_cases/1850_track1_super_fast_llnl.xml
. moved to "cam4" versions
models/atm/cam/bld/namelist_files/use_cases/1850-2005_track1.xml
models/atm/cam/bld/namelist_files/use_cases/1850_track1.xml
. deleted (cam4 versions already exist)
models/atm/cam/tools/definemld/*
models/atm/cam/tools/defineqflux/*
. old SOM utilities
models/atm/cam/tools/mkrgrid/*
models/atm/cam/tools/mkrgridnew/*
models/atm/cam/tools/mkrgridsst/*
. old reduced grid utilities
models/ice/csim4/*
. move to models/atm/cam/src/utils/cam_csim4/*
models/ocn/dom/*
. move to models/atm/cam/src/utils/cam_dom/*
List all subroutines added and what they do:
models/atm/cam/bld/namelist_files/use_cases/waccm_1974-2005_cam4.xml
. new waccm use case
models/atm/cam/bld/namelist_files/use_cases/1850_cam4_2xco2.xml
models/atm/cam/bld/namelist_files/use_cases/1850_cam4_super_fast_llnl.xml
. moved from "track1" versions
models/atm/cam/src/utils/cam_csim4/*
. moved from models/ice/csim4/*
models/atm/cam/src/utils/cam_dom/*
. moved from models/ocn/dom/*
List all existing files that have been modified, and describe the changes:
SVN_EXTERNAL_DIRECTORIES
. drvseq3_1_18 --> drvseq3_1_19
MCT2_7_0_100106 --> MCT2_7_0_100228
pio1_0_12 --> pio1_0_15
scripts4_100225 --> scripts4_100306
models/atm/cam/bld/cam.cpl7.template
. add logic to recognize USE_MPISERIAL env var and set -nospmd arg to
configure when it's value is 'TRUE'.
models/atm/cam/bld/configure
. change the filepaths for dom and csim4
. remove unused arg esmf_libvers
models/atm/cam/src/control/cam_history.F90
. change attribute "cell_method" to "cell_methods" to be compliant with CF
conventions.
models/atm/cam/src/control/runtime_opts.F90
. Put log output for divdampn inside conditional logic so it's only written
to the logfile for the spectral dycores.
models/atm/cam/src/dynamics/fv/dp_coupling.F90
. comment out the 'dpcoup dqreq' and 'dpcoup cant adjust' log messages.
They're making the output log file unusable.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS
edinburgh/lf95: All PASS
edinburgh/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam4_1_05
Originator(s): Francis Vitt
Date: 8 Mar 2010
One-line Summary: Added features and misc code cleanup
Purpose of changes:
Features added:
- ability to use CLM generated MEGAN VOC emissions
- ability to use CLM generated dry deposition velocities
- ability to read in prescribed aerosol deposition fluxes
to be passed to surface models via the coupler
- ability to use the WACCM age-of-air diagnostic tracers
in any configuration
- ability to use the WACCM TEM circulation diagnostics in
non-WACCM configurations that use the finite-volume dycore
- ability to output data fields need for offline radiation
calculations
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Brian Eaton
List all subroutines eliminated:
D models/atm/cam/src/physics/waccm/tracers.F90
D models/atm/cam/src/physics/waccm/ctem.F90
these were moved so that they can be made available outside of waccm
List all subroutines added and what they do:
A models/atm/cam/src/physics/cam/aoa_tracers.F90
waccm age-of-air tracers moved make available outside of waccm
A models/atm/cam/src/physics/cam/radiation_data.F90
module added to output field for offline radiation calculations
A models/atm/cam/src/chemistry/utils/ocnice_aero.F90
utility module added to compute surface aerodynamic quatities for
dry deposition
A models/atm/cam/src/chemistry/utils/aerodep_flx.F90
used to read in precribed aerosol deposition fluxes
A models/atm/cam/src/dynamics/fv/ctem.F90
waccm TEM diagnostics moved make available outside of waccm
List all existing files that have been modified, and describe the changes:
U models/atm/cam/bld/config_files/definition.xml
U models/atm/cam/bld/configure
configure options added:
-[no]age_of_air_trcs
-lnd_vocsrc
-lnd_drydep
U models/atm/cam/bld/build-namelist
changed needed for WACCM TEM diagnostics, age-of-air, MEGAN VOCs and dry depostion
U models/atm/cam/bld/namelist_files/namelist_definition.xml
namelist variables added:
do_circulation_diags
rad_data_output
rad_data_histfile_num
aoa_tracers_flag
aerdep_flx_*
U models/atm/cam/bld/namelist_files/use_cases/1850-2005_cam4.xml
U models/atm/cam/bld/namelist_files/use_cases/2005-2100_cam4_rcp85.xml
U models/atm/cam/bld/namelist_files/use_cases/1850_cam4.xml
added TEM diagnostics
U models/atm/cam/src/control/runtime_opts.F90
cleanup - the presrcibed_* utility data modules now read there own namelist
U models/atm/cam/src/cpl_esmf/atm_comp_esmf.F90
U models/atm/cam/src/cpl_mct/atm_comp_mct.F90
U models/atm/cam/src/control/camsrfexch_types.F90
mods to receive the CLM generated VOC emissions and dry dep velocities
U models/atm/cam/src/control/cam_restart.F90
U models/atm/cam/src/control/cam_comp.F90
U models/atm/cam/src/dynamics/fv/dp_coupling.F90
U models/atm/cam/src/dynamics/sld/restart_dynamics.F90
U models/atm/cam/src/dynamics/eul/restart_dynamics.F90
U models/atm/cam/src/dynamics/fv/restart_dynamics.F90
U models/atm/cam/src/dynamics/homme/restart_dynamics.F90
U models/atm/cam/src/dynamics/fv/dyn_comp.F90
U models/atm/cam/src/dynamics/fv/inital.F90
mods for TEM diagnostics initialization
U models/atm/cam/src/physics/cam/tropopause.F90
U models/atm/cam/src/physics/cam/radiation.F90
U models/atm/cam/src/physics/cam/radsw.F90
U models/atm/cam/src/physics/cam/radconstants.F90
U models/atm/cam/src/physics/cam/radlw.F90
cleanup
fixed threading bug in calculation of SOLIN diagnostic
U models/atm/cam/src/physics/cam/rad_constituents.F90
U models/atm/cam/src/physics/cam/tphysbc.F90
mods to output radiation data for offline calculations
U models/atm/cam/src/physics/cam/physpkg.F90
changes in initialization
U models/atm/cam/src/physics/cam/advnce.F90
advances aoa_tracers and aerodep_flx modules added
U models/atm/cam/src/physics/cam/tphysac.F90
aoa_tracers tendancies invoked
U models/atm/cam/src/chemistry/utils/prescribed_ghg.F90
U models/atm/cam/src/chemistry/utils/prescribed_aero.F90
U models/atm/cam/src/chemistry/utils/prescribed_ozone.F90
U models/atm/cam/src/chemistry/utils/prescribed_volcaero.F90
these read their own namelists to set options
U models/atm/cam/src/chemistry/utils/tracer_data.F90
path90 compilier bug fix
U models/atm/cam/src/chemistry/mozart/mo_drydep.F90
U models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
U models/atm/cam/src/chemistry/mozart/mo_chemini.F90
U models/atm/cam/src/chemistry/mozart/chemistry.F90
changes to make use of the CLM generated dry deposition velocities
U models/atm/cam/src/chemistry/mozart/mz_aerosols_intr.F90
code cleanup
U models/atm/cam/src/chemistry/mozart/mo_srf_emissions.F90
changes to make use of the CLM generated MEGAN VOC emissions
U models/atm/cam/src/dynamics/sld/inidat.F90
U models/atm/cam/src/dynamics/eul/inidat.F90
U models/atm/cam/src/dynamics/homme/inidat.F90
U models/atm/cam/src/dynamics/fv/inidat.F90
U models/atm/cam/src/physics/cam/initindx.F90
changes for age-of-air tracers
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS
edinburgh/lf95: All PASS
edinburgh/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam4_1_04
Originator(s): eaton
Date: Mon Mar 1 14:23:48 MST 2010
One-line Summary: update externals; update rcp8.5 use case
Purpose of changes:
. update externals to match ccsm4_0_beta45
. update the solar and ghg files in the rcp8.5 use case
. Move the pilgrim directory location so that CCSM tags don't need to
explicitly declare pilgrim to be an external. It will now automatically
come with the declared CAM tag. Since pilgrim is maintained as part of
CAM, this change enforces the consistency between pilgrim and CAM tags
that is already implicitly assumed by the development process.
. misc cleanup
Bugs fixed (include bugzilla ID):
. There was a bug in share code that caused incorrect interpolation of
annual cycle datasets when the model start date was 101. This code was
only being used by CICE in prescribed mode for the interpolation of ice
fraction. CAM inherited the fix by updating the share external.
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets:
. CLM datasets updated with the new external.
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
models/utils/pilgrim/*
. pilgrim directory moved to models/atm/cam/src/utils
List all subroutines added and what they do:
models/atm/cam/src/utils/pilgrim/*
. pilgrim directory moved from models/utils/
List all existing files that have been modified, and describe the changes:
SVN_EXTERNAL_DIRECTORIES
. drvseq3_1_16 --> drvseq3_1_18
clm3_6_63 --> clm3_7_05
cice4_0_20100105 --> cice4_0_20100222
docn8_100103 --> docn8_100124
share3_091217 --> share3_100228
timing_090929 --> timing_091021
scripts4_100112b --> scripts4_100225
models/atm/cam/bld/cam.cpl7.template
. remove the atm_in and atm_in.tmp file once they're no longer needed
(after the cam.buildnml.csh script is complete).
models/atm/cam/bld/configure
. update filepath for pilgrim:
models/utils/pilgrim --> models/atm/cam/src/utils/pilgrim
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. update CLM datasets to be consistent with the clm3_7_05 tag
models/atm/cam/bld/namelist_files/use_cases/2005-2100_cam4_rcp85v2.xml
models/atm/cam/bld/namelist_files/use_cases/2005-2100_cam4_rcp85.xml
. update solar and ghg files
models/atm/cam/src/control/cam_comp.F90
. remove superfluous declaration and setting of dtime in cam_final.
models/atm/cam/src/cpl_esmf/atm_comp_esmf.F90
models/atm/cam/src/cpl_mct/atm_comp_mct.F90
. fix the comment in atm_run_mct about looping over the radiation interval.
Should say looping over atm coupling interval.
models/atm/cam/src/physics/cam/radsw.F90
. remove unused JPE_VMATH ifdefs
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
004 bl132 TBL.sh e32c4dh ghgrmp+1850_cam4 9s ......................FAIL! rc= 7 at Fri Feb 26 19:16:35 MST 2010
011 bl151 TBL.sh e64h outfrq24h 2d ................................FAIL! rc= 7 at Fri Feb 26 19:36:47 MST 2010
015 bl331 TBL.sh f4c4dh co2rmp+1850_cam4 9s .......................FAIL! rc= 7 at Fri Feb 26 19:43:58 MST 2010
019 bl332 TBL.sh f4dh ghgrmp 9s ...................................FAIL! rc= 7 at Fri Feb 26 19:59:01 MST 2010
031 bl336 TBL.sh wm4c4h outfrq3s+waccm_1850_cam4 9s ...............FAIL! rc= 7 at Fri Feb 26 20:24:58 MST 2010
034 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ..............................FAIL! rc= 7 at Fri Feb 26 20:36:28 MST 2010
038 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .............................FAIL! rc= 7 at Mon Mar 1 10:36:21 MST 2010
042 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7 at Mon Mar 1 10:52:18 MST 2010
057 bl992 TBL_ccsm.sh f19_g16 E 2d ................................FAIL! rc= 5 at Fri Feb 26 22:28:50 MST 2010
bl132, bl151, bl331, bl332, bl336, bl337 fail due to updating the fsurdat files
at resolutions other than fv 1 and 2 deg.
bl151, bl337, bl371, bl375 fail due to fixing the bug in share code that
was giving incorrect results for runs with start dates of 101.
bl992 fails due to a pre-existing failure of the build in the previous
tag.
edinburgh/lf95:
004 bl111 TBL.sh e8t5mdm ttrac 9s .................................FAIL! rc= 7 at Fri Feb 26 18:56:30 MST 2010
008 bl112 TBL.sh e8dm ghgrmp 9s ...................................FAIL! rc= 7 at Fri Feb 26 18:59:52 MST 2010
014 bl114 TBL.sh e8dm co2rmp 9s ...................................FAIL! rc= 7 at Fri Feb 26 19:03:41 MST 2010
040 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7 at Mon Mar 1 11:31:37 MST 2010
edinburgh/pgi:
004 bl111 TBL.sh e8t5mdm ttrac 9s .................................FAIL! rc= 7 at Fri Feb 26 19:07:59 MST 2010
008 bl112 TBL.sh e8dm ghgrmp 9s ...................................FAIL! rc= 7 at Fri Feb 26 19:21:32 MST 2010
014 bl114 TBL.sh e8dm co2rmp 9s ...................................FAIL! rc= 7 at Fri Feb 26 19:39:55 MST 2010
040 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7 at Mon Mar 1 09:52:21 MST 2010
bl111, bl112, bl114 fail due to updating the fsurdat files
at resolutions other than fv 1 and 2 deg.
bl317 fails due to fixing the bug in share code that was giving incorrect
results for runs with start dates of 101.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB except configs using new CLM datasets
(updated CLM tag) and runs starting on 101 (bug fix in share code).
===============================================================
===============================================================
Tag name: cam4_1_03
Originator(s): Jim Edwards, Mark Taylor, Pat Worley
Date: 2/23/2010
One-line Summary: Homme dycore update, pio netcdf4 support
Purpose of changes: Homme dycore development, unstructured grid support
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
cam_trunk/SVN_EXTERNAL_DIRECTORIES
Updated pio external
cam_trunk/models/atm/cam/SVN_EXTERNAL_DIRECTORIES
Updated cprnc to the same as used by ccsm
cam_trunk/models/atm/cam/bld/build-namelist
cam_trunk/models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
Namelist cleanup for the homme dycore
cam_trunk/models/atm/cam/src/chemistry/mozart/mo_airplane.F90
cam_trunk/models/atm/cam/src/chemistry/mozart/mo_drydep.F90
Replaced wrap_nf with pio.
cam_trunk/models/atm/cam/src/chemistry/utils/tracer_data.F90
cam_trunk/models/atm/cam/src/physics/cam/modal_aer_opt.F90
Replaced pio_noclobber with pio_nowrite, this change was required for
some netcdf4 versions
cam_trunk/models/atm/cam/src/control/cam_history.F90
Removed dead code
cam_trunk/models/atm/cam/src/dynamics/homme/dp_coupling.F90
cam_trunk/models/atm/cam/src/dynamics/homme/inital.F90
cam_trunk/models/atm/cam/src/dynamics/homme/external/control_mod.F90
cam_trunk/models/atm/cam/src/dynamics/homme/external/global_norms_mod.F90
cam_trunk/models/atm/cam/src/dynamics/homme/external/namelist_mod.F90
cam_trunk/models/atm/cam/src/dynamics/homme/external/prim_advance_mod.F90
cam_trunk/models/atm/cam/src/dynamics/homme/external/prim_driver_mod.F90
cam_trunk/models/atm/cam/src/dynamics/homme/external/prim_state_mod.F90
cam_trunk/models/atm/cam/src/dynamics/homme/external/viscosity_mod.F90
cam_trunk/models/atm/cam/src/dynamics/homme/restart_dynamics.F90
cam_trunk/models/atm/cam/src/dynamics/homme/stepon.F90
Ongoing development changes for the homme dycore
Enable subcycling & foward-in-time dynamics by default
Update to time-split forcing for forward-in-time dynamics
Removed unnecessary data from restart files
cam_trunk/models/atm/cam/src/physics/cam/phys_gmean.F90
cam_trunk/models/atm/cam/src/physics/cam/phys_grid.F90
Memory reductions in loadbalance options for unstructured grids
cam_trunk/models/atm/cam/src/physics/cam/radconstants.F90
move psf from module global to subroutine local
cam_trunk/models/atm/cam/src/utils/cam_pio_utils.F90
Added support for two new pio file types
netcdf4c (hdf5 compressed) and netcdf4p (hdf5 parallel)
these are somewhat mutually exclusive types: a file opened
to write compressed data must be serial.
PIO will detect and open files to read with the correct
option for that file type regardless of option requested.
cam_trunk/models/atm/cam/test/system/test_driver.sh
Update bluefire pnetcdf path
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: all pass except ...
052 bl731 TBL.sh h16x4dm aqua 9s ..................................FAIL!
053 sm991 TSM_ccsm.sh f19_f19 F 2d ................................FAIL!
054 er991 TER_ccsm.sh f19_f19 F 1+1d ..............................FAIL!
055 sm992 TSM_ccsm.sh f19_g16 E 2d ................................FAIL!
056 er992 TER_ccsm.sh f19_g16 E 1+1d ..............................FAIL!
057 bl992 TBL_ccsm.sh f19_g16 E 2d ................................FAIL!
bl731 fails due to update of homme dycore.
- 991 and 992 tests failed to build due to update of drv external
without corresponding required updates to other external
components - carried forward from 4_1_02
edinburgh/lf95: all pass
calgary/pgi: all pass
CAM tag used for the baseline comparison tests if different than previous
tag: testing on bluefire required a mod to cam4_1_02 to allow that pio tag to build
with pnetcdf-1.1.1svn. The pnetcdf version check in pio had to be commented out.
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
Tag name: cam4_1_02
Originator(s): mvr
Date: Fri Feb 19 2010
One-line Summary: enhancement of test suite on bluefire; added use_cases for
rcp runs; qarch setting returned to Makefile and xlf compiler patch added;
mods for the atm_aero infodata flag; build-namelist bug fix for dynamic PFT dataset
notes:
1 rcp use_cases will need update to solar data file when it is ready
2 test suite on bluefire enhanced to include more configurations including better
coverage of cam4 (track1) runs - more enhancements to follow
3 atm_aero flag indicates whether CAM will provide carbon and dust deposition fluxes
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: none
Describe any changes made to the namelist: none
- improved error messages; addition of cam_chempkg; bug fix for dynamic PFT dataset
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
D models/atm/cam/test/system/config_files/f1.9c351dh
D models/atm/cam/test/system/config_files/f1.9dm
D models/atm/cam/test/system/config_files/e32dh
D models/atm/cam/test/system/config_files/f4c351pdh
D models/atm/cam/test/system/config_files/fm1.9dh
D models/atm/cam/test/system/config_files/wm4h
D models/atm/cam/test/system/config_files/e32pdh
- removed obsolete configuration options files no longer needed in test suite
D models/atm/cam/bld/namelist_files/use_cases/waccm_1850-2005_track1.xml
D models/atm/cam/bld/namelist_files/use_cases/1850-2005_track1_super_fast_llnl.xml
D models/atm/cam/bld/namelist_files/use_cases/waccm_1850_track1.xml
- changed name of use_cases from 'track1' to cam4
D models/atm/cam/tools/newcprnc
D models/atm/cam/tools/cprnc
- removed all versions of netcdf comparison tool - will be added as external in subsequent tag
List all subroutines added and what they do: none
A models/atm/cam/test/system/config_files/e32c4pdm
A models/atm/cam/test/system/config_files/f1.9c5dm
A models/atm/cam/test/system/config_files/fs1.9c4dh
A models/atm/cam/test/system/config_files/f4c5dh
A models/atm/cam/test/system/config_files/wm4c4h
A models/atm/cam/test/system/config_files/f1.9c4dh
A models/atm/cam/test/system/config_files/f4c4dh
A models/atm/cam/test/system/config_files/e32c4dh
A models/atm/cam/test/system/config_files/f4c4pdh
- new configuration options files needed for enhancements to test suite
A + models/atm/cam/bld/namelist_files/use_cases/waccm_1850-2005_cam4.xml
A + models/atm/cam/bld/namelist_files/use_cases/1850-2005_cam4_super_fast_llnl.xml
A + models/atm/cam/bld/namelist_files/use_cases/waccm_1850_cam4.xml
- changed name of use_cases from 'track1' to cam4
A models/atm/cam/bld/namelist_files/use_cases/2005-2100_cam4_rcp85.xml
A models/atm/cam/bld/namelist_files/use_cases/2005-2100_cam4_rcp85v2.xml
- new use_cases required for rcp runs - will require mod for new solar data file
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/tests_pretag_bluefire
- changes made to set of required pretag tests on bluefire
M models/atm/cam/test/system/TBL.sh
M models/atm/cam/test/system/TBR.sh
M models/atm/cam/test/system/TER.sh
M models/atm/cam/test/system/TEQ_ccsm.sh
M models/atm/cam/test/system/TER_ccsm.sh
M models/atm/cam/test/system/TBL_ccsm.sh
M models/atm/cam/test/system/TEQ.sh
- b4b checks now include any initial files generated
M models/atm/cam/test/system/nl_files/aqpgro
- namelist options test file now included inithist='ENDOFRUN'
M models/atm/cam/test/system/input_tests_master
- master list of test definitions changes to reflect enhancements made to test suite
M models/atm/cam/bld/Makefile.in
- qarch setting replaced for bluefire compilations (was removed in cam4_1_00) and compiler
patch was added to FFLAGS and CFLAGS
M models/atm/cam/bld/build-namelist
- improved error messages; addition of cam_chempkg; bug fix for dynamic PFT dataset
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- info added for cam_chempkg
M models/atm/cam/bld/namelist_files/use_cases/1850-2005_cam4.xml
- fix to xml syntax for several recently added fields; added pre-industrial sst datasets
M models/atm/cam/bld/namelist_files/use_cases/2005_cam4.xml
M models/atm/cam/bld/namelist_files/use_cases/1850_cam4.xml
- fix to xml syntax for several recently added fields
M models/atm/cam/src/control/cam_history.F90
- mods to ensure initial files get the default output variables when empty_htapes=.true.
M models/atm/cam/src/cpl_mct/atm_comp_mct.F90
M models/atm/cam/src/cpl_esmf/atm_comp_esmf.F90
- use atm_aero flag to provide carbon and dust deposition fluxes when desired
M models/atm/cam/src/physics/cam/phys_control.F90
- addition of cam_chempkg to phys_ctl_nl namelist
M SVN_EXTERNAL_DIRECTORIES
M .
- update of drv tag needed for use of the atm_aero info flag
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: all PASS except
004 bl132 TBL.sh e32c4dh ghgrmp+1850_cam4 9s ......................FAIL! rc= 7 at Fri Feb 19 10:14:28 MST 2010
006 bl133 TBL.sh e32c4pdm aqpgro 3s ...............................FAIL! rc= 7 at Fri Feb 19 10:14:28 MST 2010
011 bl151 TBL.sh e64h outfrq24h 2d ................................FAIL! rc= 7 at Fri Feb 19 10:14:37 MST 2010
015 bl331 TBL.sh f4c4dh co2rmp+1850_cam4 9s .......................FAIL! rc= 7 at Fri Feb 19 10:14:39 MST 2010
021 bl333 TBL.sh f4c4pdh aqpgro 3s ................................FAIL! rc= 7 at Fri Feb 19 10:14:44 MST 2010
031 bl336 TBL.sh wm4c4h outfrq3s+waccm_1850_cam4 9s ...............FAIL! rc= 7 at Fri Feb 19 10:14:49 MST 2010
034 bl337 TBL.sh f4c5dh fv2d_8tsk 9s ..............................FAIL! rc= 7 at Fri Feb 19 10:14:50 MST 2010
038 bl371 TBL.sh f1.9c5dm fvvp_lb2 9s .............................FAIL! rc= 7 at Fri Feb 19 10:14:50 MST 2010
042 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7 at Fri Feb 19 10:15:02 MST 2010
045 bl374 TBL.sh fs1.9c4dh outfrq3s+1850-2005_cam4_super_fast_llnl 9s FAIL! rc= 7 at Fri Feb 19 10:15:03 MST 2010
049 bl387 TBL.sh f1.9c4dh outfrq3s+1850-2005_cam4 9s ..............FAIL! rc= 7 at Fri Feb 19 10:15:04 MST 2010
053 sm991 TSM_ccsm.sh f19_f19 F 2d ................................FAIL! rc= 4 at Fri Feb 19 10:15:31 MST 2010
054 er991 TER_ccsm.sh f19_f19 F 1+1d ..............................FAIL! rc= 6 at Fri Feb 19 10:15:31 MST 2010
055 sm992 TSM_ccsm.sh f19_g16 E 2d ................................FAIL! rc= 4 at Fri Feb 19 10:15:47 MST 2010
056 er992 TER_ccsm.sh f19_g16 E 1+1d ..............................FAIL! rc= 6 at Fri Feb 19 10:15:47 MST 2010
057 bl992 TBL_ccsm.sh f19_g16 E 2d ................................FAIL! rc= 4 at Fri Feb 19 10:15:47 MST 2010
- bl132, bl133, bl331, bl333, bl336, bl337, bl371, bl374, bl387 failed due to change in test definition
- bl151, bl336, bl375 failed due to replacement of qarch setting in Makefile and inclusion of compiler patch
- 991 and 992 tests failed to build due to update of drv external without corresponding required updates to
other external components
edinburgh/lf95: all PASS except
010 bl113 TBL.sh e8pdm aqpgro 3s ..................................FAIL! rc= 7 at Thu Feb 18 15:41:54 MST 2010
031 bl313 TBL.sh f10pdm aqpgro 3s .................................FAIL! rc= 7 at Thu Feb 18 16:47:54 MST 2010
- these failed because initial files are now generated and included in b4b checks
edinburgh/pgi: all PASS except
010 bl113 TBL.sh e8pdm aqpgro 3s ..................................FAIL! rc= 7 at Thu Feb 18 15:21:07 MST 2010
031 bl313 TBL.sh f10pdm aqpgro 3s .................................FAIL! rc= 7 at Thu Feb 18 15:38:12 MST 2010
- these failed because initial files are now generated and included in b4b checks
CAM tag used for the baseline comparison tests if different than previous
tag: additional testing done with cam and drv changes placed into ccsm4_0_beta43 - CME.f19_f19.F.bluefire PASSED
Summarize any changes to answers: BFB, but pergro test was performed on bluefire for configuration where things
were not b4b due to replacement of qarch setting and addition of compiler patch (fv 1.9x2.5 hybrid - passed)
===============================================================
Tag name: cam4_1_01
Originator(s): mvr
Date: Wed Feb 10 2010
One-line Summary: adjustment to dust mods of previous tag; general cleanup of test scripts
Purpose of changes: dust mods of previous tag were applied for all configurations, but
were not intended when using cam5 physics (indicated when MODAL_AERO is defined)
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
D models/atm/cam/test/system/config_files/h5x8dm
D models/atm/cam/test/system/config_files/h5x8adm
D models/atm/cam/test/system/config_files/h5x8idm
- removed configuration options test files for unused homme grid
D models/atm/cam/test/system/posttag_cron_lightning.sh
D models/atm/cam/bld/run-lightning.csh
- removed test and template scripts for obsolete machine lightning
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/config_files/scm64bfbiop_35
M models/atm/cam/test/system/config_files/e64bfbiop_35
M models/atm/cam/test/system/TBR.sh
M models/atm/cam/test/system/test_driver.sh
M models/atm/cam/test/system/CAM_runcmnd.sh
- cleanup of test scripts, including removal of code for obsolete machines lightning and dublin
M models/atm/cam/test/system/input_tests_master
- corrected test definitions where usecase 1870_prog_aero was replaced with 1850_cam4_trop_bam
M models/atm/cam/src/chemistry/bulk_aero/dust_intr.F90
- adj to dust changes of previous tag, where mods will not have an affect when MODAL_AERO
is defined
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: all PASS except
004 bl132 TBL.sh e32dh ghgrmp 9s ..................................FAIL! rc= 7 at Wed Feb 10 09:04:46 MST 2010
011 bl151 TBL.sh e64h outfrq24h 2d ................................FAIL! rc= 7 at Wed Feb 10 09:04:54 MST 2010
015 bl331 TBL.sh f4dh co2rmp 9s ...................................FAIL! rc= 7 at Wed Feb 10 09:05:02 MST 2010
019 bl332 TBL.sh f4dh ghgrmp 9s ...................................FAIL! rc= 7 at Wed Feb 10 09:05:06 MST 2010
033 bl337 TBL.sh f4dh fv2d_8tsk 9s ................................FAIL! rc= 7 at Wed Feb 10 09:05:11 MST 2010
037 bl371 TBL.sh f1.9dm fvvp_lb2 9s ...............................FAIL! rc= 7 at Wed Feb 10 09:05:27 MST 2010
041 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7 at Wed Feb 10 09:05:38 MST 2010
- all were expected, as they are cam5 configurations where dust mod of cam4_1_00 was undone
edinburgh/lf95: all PASS except
021 bl311 TBL.sh f10t5mdm ttrac 9s ................................FAIL! rc= 7 at Tue Feb 9 19:05:35 MST 2010
029 bl312 TBL.sh f10dm ghgrmp 9s ..................................FAIL! rc= 7 at Tue Feb 9 19:42:09 MST 2010
040 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7 at Tue Feb 9 20:56:52 MST 2010
- all were expected, as they are cam5 configurations where dust mod of cam4_1_00 was undone
edinburgh/pgi: all PASS except
021 bl311 TBL.sh f10t5mdm ttrac 9s ................................FAIL! rc= 7 at Tue Feb 9 18:20:27 MST 2010
029 bl312 TBL.sh f10dm ghgrmp 9s ..................................FAIL! rc= 7 at Tue Feb 9 18:29:45 MST 2010
040 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7 at Tue Feb 9 18:49:06 MST 2010
- all were expected, as they are cam5 configurations where dust mod of cam4_1_00 was undone
CAM tag used for the baseline comparison tests if different than previous
tag: additional testing was done vs cam3_6_79 to verify the baseline tests listed above were b4b
Summarize any changes to answers: BFB
===============================================================
Tag name: cam4_1_00
Originator(s): eaton
Date: Wed Feb 3 10:35:05 MST 2010
One-line Summary: update names to reflect CAM4 release; update trop_bam dust; SST datasets.
Purpose of changes:
. The decision was made to publicly release what we have been calling
track1 or cam3_5_1 during the development process as CAM-4.0. This
update is mainly to make name changes that are consistent with that
decision.
The default physics on the trunk is still in an intermediate development
state between cam4 and the new track5 physics. For that reason the trunk
tag name has been advanced beyond cam4_0_00. We expect to merge the
track5 development branch to the trunk in the near future.
In the public release of CAM-4.0 the cam4 physics will obviously be the
default. But to get cam4 physics from a trunk tag you must give the
argument "-phys cam4" to configure. For backwards compatibility with
existing scripts configure will continue to recognize the value cam3_5_1
as an alias for cam4.
. Update SST datasets for 1850 climatology and add 1850-2008 AMIP.
. Changes for dust in trop_bam
- Update soil_erod files for 1, 1/2, and 1/4 degree FV
- New use case: 1850_cam4_trop_bam
- tuning mod in dust_emis_intr
. Add workaround to Makefile for problems with track5 physics on bluefire.
This workaround causes a roundoff level change in other configurations.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets:
. updated soil_erod files for 1, 1/2, and 1/4 degree FV
. updated the "clim_pi" SST datasets which are used when sim_year=1850.
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
models/atm/cam/bld/namelist_files/use_cases/1990_prog_aero.xml
. deleted
models/atm/cam/bld/namelist_files/use_cases/1850-2005_cam351.xml
models/atm/cam/bld/namelist_files/use_cases/1850_cam351.xml
models/atm/cam/bld/namelist_files/use_cases/1870_prog_aero.xml
models/atm/cam/bld/namelist_files/use_cases/2005_cam351.xml
. old names
List all subroutines added and what they do:
models/atm/cam/bld/namelist_files/use_cases/1850-2005_cam4.xml
models/atm/cam/bld/namelist_files/use_cases/1850_cam4.xml
models/atm/cam/bld/namelist_files/use_cases/2005_cam4.xml
. new names
models/atm/cam/bld/namelist_files/use_cases/1850_cam4_trop_bam.xml
. renamed from 1870_prog_aero.xml
. update 1850 GHG values
. update prescribed ozone and oxidants datasets
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/build-namelist
models/atm/cam/bld/config_files/definition.xml
models/atm/cam/bld/configure
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. change valid values of phys to be cam3, cam4, cam5, ideal, adiabatic.
cam5 is default on the trunk. For backwards compatibility accept
cam3_5_1 as an alias for cam4.
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. add SST datasets for 1850 climatology and 1850-2008 AMIP.
. update soil_erod files for 1, 1/2, and 1/4 degree FV
models/atm/cam/bld/Makefile.in
. remove the -qarch=auto flag for AIX builds. This is a temporary
workaround to address current problems with the track5 code on bluefire.
models/atm/cam/bld/namelist_files/namelist_definition.xml
. documentation update
models/atm/cam/bld/build-namelist
models/atm/cam/bld/namelist_files/namelist_definition.xml
models/atm/cam/src/physics/cam/phys_control.F90
. add namelist variable to be set by build-namelist that communicates the
name of the physics package.
. add logical query function, physpkg_is(name), to allow control logic to
be based on the name of the physics package.
models/atm/cam/src/chemistry/bulk_aero/dust_intr.F90
. modify a conditional on soil_erod_tmp at line 858 (tuning mod)
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
004 bl132 TBL.sh e32dh ghgrmp 9s ..................................FAIL! rc= 7 at Tue Feb 2 19:23:17 MST 2010
011 bl151 TBL.sh e64h outfrq24h 2d ................................FAIL! rc= 7 at Mon Feb 1 13:08:47 MST 2010
015 bl331 TBL.sh f4dh co2rmp 9s ...................................FAIL! rc= 7 at Tue Feb 2 19:46:49 MST 2010
019 bl332 TBL.sh f4dh ghgrmp 9s ...................................FAIL! rc= 7 at Tue Feb 2 19:54:22 MST 2010
030 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7 at Mon Feb 1 13:40:03 MST 2010
033 bl337 TBL.sh f4dh fv2d_8tsk 9s ................................FAIL! rc= 7 at Tue Feb 2 20:11:27 MST 2010
037 bl371 TBL.sh f1.9dm fvvp_lb2 9s ...............................FAIL! rc= 7 at Tue Feb 2 20:24:18 MST 2010
041 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7 at Mon Feb 1 14:13:58 MST 2010
044 bl374 TBL.sh fm1.9dh outfrq3s 9s ..............................FAIL! rc= 7 at Tue Feb 2 20:58:25 MST 2010
048 bl387 TBL.sh f1.9c351dh outfrq3s+1850-2005_track1 9s ..........FAIL! rc= 4 at Tue Feb 2 21:02:27 MST 2010
bl151, bl336, and bl375 fail due to the removal of the -qarch=auto flag.
That change only introduced roundoff diffs (validated w/ a pergro test)
into configs running without debug flags.
bl132, bl151, bl331, bl332, bl337, bl371, bl375, bl374 are failing due to
the tuning mod in dust_intr.F90. This is larger than roundoff.
bl387 fails due to updating the "clim_pi" SST dataset.
edinburgh/lf95:
021 bl311 TBL.sh f10t5mdm ttrac 9s ................................FAIL! rc= 7 at Tue Feb 2 20:08:59 MST 2010
029 bl312 TBL.sh f10dm ghgrmp 9s ..................................FAIL! rc= 7 at Tue Feb 2 20:35:01 MST 2010
040 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7 at Tue Feb 2 21:25:21 MST 2010
edinburgh/pgi:
021 bl311 TBL.sh f10t5mdm ttrac 9s ................................FAIL! rc= 7 at Tue Feb 2 19:21:44 MST 2010
029 bl312 TBL.sh f10dm ghgrmp 9s ..................................FAIL! rc= 7 at Tue Feb 2 19:27:41 MST 2010
040 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7 at Tue Feb 2 19:39:15 MST 2010
bl311, bl312, and bl317 fail with both lf95 and pgi due to the tuning mod
in dust_intr.F90. This is larger than roundoff.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB except:
. non-debug runs on bluefire all have roundoff diffs due to qarch flag
. trop_mozart and trop_mam3 are both affected by the tuning mod in
dust_intr.F90
. any run using new SST datasets (*clim_pi* versions)
===============================================================
===============================================================
Tag name: cam3_6_79
Originator(s): mvr, hannay
Date: Fri Jan 22 2010
One-line Summary: New SSTs data for 1850 runs; Make the SST dataset "sim_year" dependant.
Purpose of changes:
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets:
- new sst datasets for 1850 configurations
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/build-namelist
- Make the SST dataset "sim_year" dependant.
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- New SSTs data for 1850 runs; Make the SST dataset "sim_year" dependant.
M models/atm/cam/bld/namelist_files/use_cases/1850-2005_track1.xml
M models/atm/cam/bld/namelist_files/use_cases/1870_prog_aero.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1850-2005_track1.xml
M models/atm/cam/bld/namelist_files/use_cases/1850_track1.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1850_cntrl.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1850_track1.xml
M models/atm/cam/bld/namelist_files/use_cases/1850_track1_super_fast_llnl.xml
M models/atm/cam/bld/namelist_files/use_cases/1850_track1_2xco2.xml
- documentation for use cases changed to reflect sst forcings are based on sim_year
M models/atm/cam/test/system/test_driver.sh
- increased wall time for running test suite on jaguar
M models/atm/cam/test/system/tests_posttag_jaguar
M models/atm/cam/test/system/tests_pretag_bluefire
M models/atm/cam/test/system/input_tests_master
- new test id's for equivalence tests to avoid confusion; added test cb372
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: all PASS except
048 bl387 TBL.sh f1.9c351dh outfrq3s+1850-2005_track1 9s ..........FAIL! rc= 7 at Thu Jan 21 20:05:20 MST 2010
- bl387 was expected to fail since it's using a new sst dataset
edinburgh/lf95: all PASS
edinburgh/pgi: all PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB except configs that use 1850 sst datasets
===============================================================
Tag name: cam3_6_78
Originator(s): pworley, fvitt, eaton
Date: Mon Jan 18 15:23:36 MST 2010
One-line Summary: Merge mods/fixes from the release branch; update externals
Purpose of changes:
. Add workaround for problem with BGP in the repro_sum_mod module. (from
pworley)
. Cleanup of some old and unused files.
. fixes to waccm and super fast chem use cases (from fvitt)
. update externals
pio1_0_6 --> pio1_0_7
scripts4_100108b --> scripts4_100112b
Bugs fixed (include bugzilla ID):
. the problem with the SCAM runs (link failure) in the last tag is fixed by
updating to pio1_0_7
. workaround apparent compiler bug that was affecting the SOLIN diagnostic
field in some configurations. (from fvitt)
. subroutine calc_col_mean should use pdeldry rather than pdel to
convert the GHG mixing ratios to masses. Only affects CAM-RT radiation
scheme.
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
Copyright
README
. Old CAM3 files
models/atm/cam/bld/script_tests/namelist.pl
models/atm/cam/bld/perl5lib/t/*
. delete unused test code
models/atm/cam/bld/camdom.cpl7.template
. delete since camdom no longer supported from ccsm scripts
models/atm/cam/bld/README
. delete old README
models/atm/cam/src/physics/cam3_6/*
. delete unsupported physics option
models/atm/cam/test/unit/*
. delete unused test code
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
SVN_EXTERNAL_DIRECTORIES
. update pio and scripts externals
models/atm/cam/bld/namelist_files/use_cases/1850-2005_track1_super_fast_llnl.xml
models/atm/cam/bld/namelist_files/use_cases/1850_track1_super_fast_llnl.xml
. mods from fvitt
models/atm/cam/bld/namelist_files/use_cases/waccm_1850-2005_track1.xml
models/atm/cam/bld/namelist_files/use_cases/waccm_1850_cntrl.xml
models/atm/cam/bld/namelist_files/use_cases/waccm_1850_track1.xml
. new 4x5 IC file
models/atm/cam/src/physics/cam/radsw.F90
. replace an intrinsic sum by an explicit do loop. This is a workaround
for a problem where the diagnostic field SOLIN would contain bad values
in some configurations (cam3_5_1 plus trop_mozart). (from fvitt)
models/atm/cam/src/physics/cam/aer_rad_props.F90
models/atm/cam/src/physics/cam/radiation.F90
. fix lines longer than 132 chars
models/atm/cam/src/utils/repro_sum_mod.F90
. Add a workaround, inside a BGP ifdef, to use i4 with the scalable
reproducible algorithm. All other platforms use i8.
models/atm/cam/src/physics/cam/radiation.F90
. Change subroutine calc_col_mean to use pdeldry rather than pdel to
convert the GHG mixing ratios to masses.
models/atm/cam/src/physics/rrtmg/radiation.F90
. Change subroutine calc_col_mean to use pdeldry rather than pdel to
convert the GHG mixing ratios to masses. This subroutine is only used
for a diagnostic calculation in the RRTMG version.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: all PASS except
021 bl333 TBL.sh f4c351pdh aqpgro 3s ..............................FAIL! rc= 7 at Mon Jan 18 12:36:15 MST 2010
030 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7 at Mon Jan 18 12:47:31 MST 2010
048 bl387 TBL.sh f1.9c351dh outfrq3s+1850-2005_track1 9s ..........FAIL! rc= 7 at Mon Jan 18 13:52:43 MST 2010
051 bl731 TBL.sh h16x4dm aqua 9s ..................................FAIL! rc= 7 at Mon Jan 18 13:57:35 MST 2010
056 bl992 TBL_ccsm.sh f19_g16 E 2d ................................FAIL! rc= 7 at Mon Jan 18 14:19:13 MST 2010
edinburgh/lf95:
034 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 7 at Mon Jan 18 13:48:38 MST 2010
037 bl315 TBL.sh f10c3dm outfrq3s 9s ..............................FAIL! rc= 7 at Mon Jan 18 13:53:14 MST 2010
edinburgh/pgi:
034 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 7 at Mon Jan 18 12:21:25 MST 2010
037 bl315 TBL.sh f10c3dm outfrq3s 9s ..............................FAIL! rc= 7 at Mon Jan 18 12:23:34 MST 2010
bl314 and bl336 fail due to the workaround in radsw for the SOLIN
calculation. It's only the diagnostic field SOLIN that contains roundoff
level diffs.
All baselines above fail because they
use the CAM-RT radiation code which
uses the calc_col_mean routine to compute a value of both O2 and CO2.
Changing the weights used in the calculation results in a roundoff level
change if the constituent has a constant mmr to begin with, and a larger
than roundoff change in any constituent that is not already constant in the
column. That causes larger than roundoff diffs in WACCM runs that have
prognostic CO2 and O2.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB except configs that use CAM-RT. That
config changes by rounoff if O2 and CO2 are constant, and by more than
roundoff if O2 or CO2 have 3D variation.
===============================================================
===============================================================
Tag name: cam3_6_77
Originator(s): mataylo, jedwards, eaton
Date: Mon Jan 11 18:33:04 MST 2010
One-line Summary: fix namespace conflict w/ new pio lib; update externals
Purpose of changes:
. Fix CAM trunk to work with new pio tag by changing the name of the
pio_utils module to cam_pio_utils.
**Note** The SCAM mode is not linking correctly with the new PIO lib.
This will be fixed shortly, but this tag was made anyway due to timelines
of the release branch for CCSM4.0.
. update PIO and CLM externals
Bugs fixed (include bugzilla ID):
. bug fix for the totalsize calculation in pbuf_init_restart (from mataylo)
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
models/atm/cam/src/utils/pio_utils.F90
. this name conflicts with one in the new PIO lib
models/atm/cam/bld/perl5lib/XML/man3/XML::Lite.3
models/atm/cam/bld/perl5lib/XML/man3/XML::Lite::Element.3
. "::" in filenames causes problems when trying to check out code from the
svn repo on a Windows system.
List all subroutines added and what they do:
models/atm/cam/src/utils/cam_pio_utils.F90
. renamed from pio_utils.F90
. also change module name from pio_utils to cam_pio_utils
models/atm/cam/bld/perl5lib/XML/man3/XML_Lite.3
models/atm/cam/bld/perl5lib/XML/man3/XML_Lite_Element.3
. renamed from XML::Lite.3 and XML::Lite::Element.3
List all existing files that have been modified, and describe the changes:
SVN_EXTERNAL_DIRECTORIES
. clm3_6_58 --> clm3_6_63
pio60_prod --> pio1_0_6
models/atm/cam/src/chemistry/utils/mo_solar_parms.F90
models/atm/cam/src/chemistry/utils/tracer_data.F90
models/atm/cam/src/control/cam_comp.F90
models/atm/cam/src/control/cam_history.F90
models/atm/cam/src/control/cam_restart.F90
models/atm/cam/src/control/ncdio_atm.F90
models/atm/cam/src/control/startup_initialconds.F90
models/atm/cam/src/cpl_esmf/atm_comp_esmf.F90
models/atm/cam/src/cpl_mct/atm_comp_mct.F90
models/atm/cam/src/dynamics/eul/restart_dynamics.F90
models/atm/cam/src/dynamics/fv/restart_dynamics.F90
models/atm/cam/src/dynamics/homme/inidat.F90
models/atm/cam/src/dynamics/homme/restart_dynamics.F90
models/atm/cam/src/dynamics/sld/restart_dynamics.F90
models/atm/cam/src/physics/cam/modal_aer_opt.F90
models/atm/cam/src/physics/cam/phys_prop.F90
models/atm/cam/src/physics/cam/radae.F90
models/atm/cam/src/physics/cam/restart_physics.F90
models/atm/cam/src/physics/cam/tropopause.F90
. change pio_utils to cam_pio_utils
models/atm/cam/src/physics/cam/phys_buffer.F90
. change pio_utils to cam_pio_utils
. bug fix for the totalsize calculation in pbuf_init_restart (from mataylo)
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except
057 sc001 TSC.sh e64bfbiop scm_prep scm64bfbiop scm_b4b_o1 7s .....FAIL! rc= 4 at Mon Jan 11 15:14:53 MST 2010
edinburgh/lf95: All PASS except
042 sc003 TSM.sh scmarmiop scmarm 7s ..............................FAIL! rc= 4 at Mon Jan 11 15:00:07 MST 2010
edinburgh/pgi: All PASS except
042 sc003 TSM.sh scmarmiop scmarm 7s ..............................FAIL! rc= 4 at Mon Jan 11 13:20:45 MST 2010
These failures are all build time link failures.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam3_6_76
Originator(s): fvitt, mataylo, eaton
Date: Sat Jan 9 13:25:13 MST 2010
One-line Summary: use case mods; update externals
Purpose of changes:
. add use cases to supply additional sub-monthly output for AR5 runs
. modify waccm use cases to use a different 4x5 initial file, and add
workaround for problem in waccm_1850_track1 (from fvitt)
. update externals to be consistent with ccsm4_0_beta39
. some cleanup work in namelist code
Bugs fixed (include bugzilla ID):
. fix for HOMME subcycling bug (from mataylo)
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets:
. waccm use cases using a new 4x5 initial file
. new initial (jan 01) and topo files for HOMME ne30np4 grid
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do:
models/atm/cam/bld/namelist_files/use_cases/1850_cam351.xml
. This is the 1850_track1 use case with daily fields requested for AR5 runs
added ('TREFHTMN','TREFHTMX','TREFHT','PRECC','PRECL','PSL').
models/atm/cam/bld/namelist_files/use_cases/1850-2005_cam351.xml
. This is the 1850-2005_track1 use case with daily fields requested for AR5 runs
models/atm/cam/bld/namelist_files/use_cases/2005_cam351.xml
. Use case for fixed 2005 forcings.
List all existing files that have been modified, and describe the changes:
SVN_EXTERNAL_DIRECTORIES
drvseq3_1_10 --> drvseq3_1_11
cice4_0_20091202 --> cice4_0_20100105
docn8_091208 --> docn8_100103
scripts4_091208 --> scripts4_100108b
MCT2_6_0_090926 --> MCT2_7_0_100106
models/atm/cam/bld/build-namelist
. refactor code that makes sure consistent land/ocn fractions are used by
surface components.
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. remove bndtvs and focndomain settings for homme grids
. add ncdata (jan 01) and bnd_topo files for ne30np4 HOMME grid
models/atm/cam/bld/namelist_files/namelist_definition.xml
. remove old SOM variables
models/atm/cam/bld/namelist_files/use_cases/waccm_1850-2005_track1.xml
models/atm/cam/bld/namelist_files/use_cases/waccm_1850_cntrl.xml
models/atm/cam/bld/namelist_files/use_cases/waccm_1850_track1.xml
. update 4x5 ncdata to waccm3548_4x_t43.cam2.i.1870-01-01-00000.nc
models/atm/cam/bld/namelist_files/use_cases/waccm_1850_track1.xml
. set ext_frc_ymd=18500102 to work around a problem that occurs when
setting it to 18500101.
models/atm/cam/src/dynamics/homme/dyn_comp.F90
. fix for subcycling bug (from Mark Taylor)
models/atm/cam/test/system/input_tests_master
. change SOM tests from f19_g15 to f19_g16
models/atm/cam/test/system/config_files/h16x4dm
models/atm/cam/test/system/config_files/h5x8adm
. remove machine specific settings for -pnc_inc and -pnc_lib
models/atm/cam/test/system/test_driver.sh
. Add {INC,LIB}_PNETCDF variables for bluefire to point to the required
version of PNetCDF.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except
056 bl992 TBL_ccsm.sh f19_g16 E 2d ................................FAIL! rc= 7 at Thu Jan 7 12:21:24 MST 2010
This baseline failure is expected due to changing the test definition from
g15 to g16.
edinburgh/lf95: All PASS.
edinburgh/pgi: All PASS.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam3_6_75
Originator(s): tcraig, cacraig, mvertens, eaton
Date: Tue Jan 5 09:45:39 MST 2010
One-line Summary: Support for DOCN; mods for running with CCSM scripts; misc fixes and cleanup.
Purpose of changes:
. Changes to allow running the DOCN data model in place of CAM-DOM.
CAM-DOM is currently still the default when running in a standalone CAM
mode (DOCN doesn't yet support the aqua-planet SSTs). But only DOCN is
supported for running F compsets with the CCSM scripts.
. Mod to allow putting namelist variables in the file
$CASEROOT/SourceMods/src.cam/user_nl when running with CCSM scripts.
This is an easier way to modify namelist settings than via the
CAM_NAMELIST_OPTS attribute in the compset definition.
. Add ESMF interfaces and support for running with an external ESMF library
via the CCSM scripts.
. Changes to support running single column mode from the CCSM scripts.
. Code cleanup -- added implicit none, private, and save statements to
modules that were missing them.
. Fix bugs that were causing problems on BG/P.
Bugs fixed (include bugzilla ID):
. bug 1073 -- change the first argument of mpi_get_processor_name from char
array to string.
. bug 1074 -- Add error return arg to mpi_abort.
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
models/atm/cam/bld/namelist_files/use_cases/1870-2000_control.xml
models/atm/cam/bld/namelist_files/use_cases/1870_bam.xml
models/atm/cam/bld/namelist_files/use_cases/1870_control.xml
models/atm/cam/bld/namelist_files/use_cases/1990_control.xml
. delete old use cases
List all subroutines added and what they do:
models/atm/cam/src/cpl_esmf/atm_comp_esmf.F90
models/atm/cam/src/cpl_esmf/atm_comp_mct.F90
. esmf interface layer
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/build-namelist
. Modify so that build-namelist doesn't try to find defaults for bndtvs when
ocn has been set to none. This was causing a problem trying to run HOMME
from the CCSM scripts.
. Add code to write the files that docn needs.
models/atm/cam/bld/cam.cpl7.template
. Add support for "PTS_MODE" == single_column
. change use of ATM_NY -- now it's resolved at build time rather than
configure time.
. mod to allow the user to put a file with namelist settings in the source
mods directory in a file named user_nl, i.e.,
$CASEROOT/SourceMods/src.cam/user_nl. This file should have the
format of a file that is supplied to build-namelist via the -infile
argument.
models/atm/cam/bld/configure
. Remove the -esmf_libvers option. The cpp macro ESMF_v2 has been replaced
by USE_ESMF_LIB for consistency with CCSM. We assume that whenever an
external ESMF lib is used, that its version is 3.1 or later. If an
external ESMF lib is not used, the default WRF implementation of the ESMF
time manager utilities has interfaces from ESMF version 2.
Modify configure to recognize the USE_ESMF_LIB environment variable which
is set by the CCSM scripts. This must be used to determine whether or
not and external lib is being used since ESMF_LIBDIR is always set by the
CCSM scripts whether it will be used or not.
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. Remove the years from the ic_ymd attributes for the finidat files. This
way they will be used with the default 101 start date without the user
having to specify -ignore_ic_year to build-namelist. This is the way
CAM's ncdata files work as well.
models/atm/cam/src/dynamics/fv/inidat.F90
models/atm/cam/src/physics/cam/buffer.F90
models/atm/cam/src/physics/cam/cam_diagnostics.F90
models/atm/cam/src/physics/cam/cloud_diagnostics.F90
models/atm/cam/src/physics/cam/cmparray_mod.F90
models/atm/cam/src/physics/cam/convect_ke_intr.F90
models/atm/cam/src/physics/cam/geopotential.F90
models/atm/cam/src/physics/cam/hirsbtpar.f90
models/atm/cam/src/physics/cam/iop_surf.F90
models/atm/cam/src/physics/cam/param_cldoptics.F90
models/atm/cam/src/physics/cam/pkg_cld_sediment.F90
models/atm/cam/src/physics/cam/pkg_cldoptics.F90
models/atm/cam/src/physics/cam/ppgrid.F90
models/atm/cam/src/physics/cam/restart_physics.F90
models/atm/cam/src/physics/cam/wv_saturation.F90
models/atm/cam/src/physics/rrtmg/cloud_diagnostics.F90
. code cleanup -- mostly adding missing implicit none, private, and save
statements to modules
. eliminate unused parameter pvermx
models/atm/cam/src/utils/abortutils.F90
. Add error return arg to mpi_abort. This fixes bug 1074.
models/atm/cam/src/utils/spmd_utils.F90
. Change the first argument of mpi_get_processor_name from char array to
string. This fixes bug 1073.
models/atm/cam/src/utils/time_manager.F90
. initialize_clock: arg ref_date - change intent from in to inout
. set_date_from_time_float: Distinquish between the esmf-v2 (the wrf_esmf
code uses these interfaces) and the esmf-v3 or later interfaces by the
cpp macro USE_ESMF_LIB. The ESMF_v2 macro is deprecated.
models/atm/cam/test/system/input_tests_master
. The 1870_bam use case was only being used in the *384 tests on jaguar.
Replace that use case by 1850_track1.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS.
dublin/lf95: All PASS.
dublin/pgi: All PASS.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam3_6_74
Originator(s): Francis Vitt
Date: 23 Dec 2009
One-line Summary: Update chemistry code with track5 branch code
Purpose of changes:
- Update trunk with latest chemistry code which is in the track5 branch.
- Eliminate the discrepancies in chem code between trunk and track5 branch.
- Fix the super_fast_llnl build-namelist use cases.
The changes merged in from track5 branch:
- added capability to specify 'INTERP_MISSING_MONTHS' to the surface and elevated emissions codes
- modifications needed for running super_fast_llnl chemistry with prognostic modal aerosols
- radiation packages camrt and rrtmg both have the ability to use spectral solar variance
- chemistry photolysis reactions and radiation heating use the same input dataset
so in chemistry configurations rad heating will have spectral solar variance
- woods and neckel schemes of using F10.7 data to model the solar spectrum is removed
- additional chemistry code changes:
. use tropopause level determined by physics/cam/tropopause.F90
. unified "users" reaction labeling scheme for waccm and cam-chem
. aircraft sources optional
. minimum mixing ratio of all chemical tracers greater than zero
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
D models/atm/cam/src/physics/cam/solvar_interface.F90
D models/atm/cam/src/chemistry/mozart/solar_photons.F90
D models/atm/cam/src/chemistry/mozart/mo_tropopause.F90
D models/atm/cam/src/chemistry/mozart/solvar_interface.F90
D models/atm/cam/src/chemistry/mozart/woods.F90
D models/atm/cam/src/chemistry/mozart/llnl_set_chem_trop.F90
D models/atm/cam/src/chemistry/mozart/solvar_woods.F90
D models/atm/cam/src/chemistry/mozart/solvar_data.F90
D models/atm/cam/src/chemistry/mozart/neckel.F90
List all subroutines added and what they do:
A models/atm/cam/bld/config_files/defaults_super_fast_llnl_mam3.xml
- configure defaults file for new super_fast_llnl_mam3 chem option
A models/atm/cam/bld/namelist_files/use_cases/1850-2005_track1_super_fast_llnl.xml
- track1 with chemistry use case
A models/atm/cam/src/physics/cam/rad_solar_var.F90
- needed for the spectral solar variance
A models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_lu_solve.F90
A models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_setrxt.F90
A models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/m_rxt_id.F90
A models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_phtadj.F90
A models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_nln_matrix.F90
A models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_adjrxt.F90
A models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/m_het_id.F90
A models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_lu_factor.F90
A models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_indprd.F90
A models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_sim_dat.F90
A models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/m_spc_id.F90
A models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_imp_sol.F90
A models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/chem_mods.F90
A models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_prod_loss.F90
A models/atm/cam/src/chemistry/pp_super_fast_llnl_mam3/mo_lin_matrix.F90
- preprocessor generated files for new super_fast_llnl_mam3 chemistry with package
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/configure
M models/atm/cam/bld/config_files/definition.xml
- chem option "super_fast_llnl_mam3" added
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- update MAM inputs
- optics file for super_fast_llnl_mam3 configuration
- default solar input file
- prescribed constituents updates
- tropopause, woods, neckel inputs removed
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- tropopause, woods, neckel namelist vars removed
- added capability to specify 'INTERP_MISSING_MONTHS' to the
surface and elevated emissions
M models/atm/cam/bld/namelist_files/use_cases/1870_prog_aero.xml
M models/atm/cam/bld/namelist_files/use_cases/1990_prog_aero.xml
M models/atm/cam/bld/namelist_files/use_cases/1870_control.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1850-2005_track1.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1850_cntrl.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1850_track1.xml
M models/atm/cam/bld/namelist_files/use_cases/1850_track1_super_fast_llnl.xml
M models/atm/cam/bld/build-namelist
- changes to build-namelist and use cases to be consistent with namelist changes
M models/atm/cam/bld/perl5lib/Build/ChemPreprocess.pm
- update to super_fast_llnl input file
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
- update chem preprocessor and rrtmg
M models/atm/cam/src/control/wrap_nf.F90
- added routine to get number of vars in a netcdf file
M models/atm/cam/src/physics/cam/tropopause.F90
- added hybrid-stobie method
M models/atm/cam/src/chemistry/utils/solar_data.F90
M models/atm/cam/src/physics/cam/physpkg.F90
M models/atm/cam/src/physics/cam/radsw.F90
M models/atm/cam/src/physics/cam/radconstants.F90
M models/atm/cam/src/physics/cam/radiation.F90
M models/atm/cam/src/physics/rrtmg/radconstants.F90
M models/atm/cam/src/physics/rrtmg/radiation.F90
M models/atm/cam/src/physics/rrtmg/radsw.F90
- radiation packages camrt and rrtmg both have the ability to use spectral solar variance
- shortened the routine names in radconstants.F90 to be less than 32 characters (F90 standard limit)
M models/atm/cam/src/chemistry/pp_trop_mam7/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_trop_mam7/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_trop_mozart/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_trop_mam3/chem_mods.F90
- unified usrrxt tagging scheme
- ext frc from file added for time-dependent aircraft emissions
M models/atm/cam/src/physics/cam/modal_aer_opt.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_data.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_initialize_data.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_newnuc.F90
M models/atm/cam/src/chemistry/bulk_aero/progseasalts_intr.F90
M models/atm/cam/src/chemistry/bulk_aero/drydep_mod.F90
M models/atm/cam/src/chemistry/bulk_aero/dust_sediment_mod.F90
M models/atm/cam/src/chemistry/bulk_aero/wetdep.F90
M models/atm/cam/src/chemistry/bulk_aero/dust_intr.F90
M models/atm/cam/src/chemistry/bulk_aero/aerosol_intr.F90
M models/atm/cam/src/chemistry/bulk_aero/sulchem.F90
M models/atm/cam/src/chemistry/mozart/mz_aerosols_intr.F90
- MAM track5 branch mods
M models/atm/cam/src/chemistry/mozart/mo_photo.F90
- check for spectral solar data input
M models/atm/cam/src/chemistry/mozart/mo_jlong.F90
M models/atm/cam/src/chemistry/mozart/mo_jshort.F90
- removed woods and neckel etf
M models/atm/cam/src/chemistry/mozart/mo_airplane.F90
M models/atm/cam/src/chemistry/mozart/mo_setext.F90
- check for aircraft input file
M models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
- tropopause level passed in
- call to one usrrxt interface
M models/atm/cam/src/chemistry/mozart/upper_bc.F90
- waccm check
M models/atm/cam/src/chemistry/mozart/mo_cph.F90
- change in usrrxt label scheme
M models/atm/cam/src/chemistry/mozart/chemistry.F90
M models/atm/cam/src/chemistry/mozart/mo_chemini.F90
- tropopause, woods, neckel namelist vars removed
- emis namelist var changes
M models/atm/cam/src/chemistry/utils/prescribed_ghg.F90
M models/atm/cam/src/chemistry/utils/prescribed_aero.F90
M models/atm/cam/src/chemistry/utils/tracer_data.F90
M models/atm/cam/src/chemistry/utils/prescribed_ozone.F90
M models/atm/cam/src/chemistry/utils/prescribed_volcaero.F90
M models/atm/cam/src/chemistry/mozart/mo_sulf.F90
M models/atm/cam/src/chemistry/mozart/tracer_cnst.F90
M models/atm/cam/src/chemistry/mozart/tracer_srcs.F90
M models/atm/cam/src/chemistry/mozart/linoz_data.F90
M models/atm/cam/src/chemistry/mozart/mo_setsoa.F90
M models/atm/cam/src/chemistry/mozart/mo_strato_sad.F90
M models/atm/cam/src/control/cam_history.F90
- misc cleanup
M models/atm/cam/src/chemistry/mozart/mo_usrrxt.F90
- unified tagging scheme
M models/atm/cam/src/chemistry/mozart/mo_extfrc.F90
M models/atm/cam/src/chemistry/mozart/mo_srf_emissions.F90
- added capability to specify 'INTERP_MISSING_MONTHS' to the
surface and elevated emissions codes
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
004 bl132 TBL.sh e32dh ghgrmp 9s ..................................FAIL! rc= 7
006 bl133 TBL.sh e32pdh aqpgro 3s .................................FAIL! rc= 7
011 bl151 TBL.sh e64h outfrq24h 2d ................................FAIL! rc= 7
015 bl331 TBL.sh f4dh co2rmp 9s ...................................FAIL! rc= 7
019 bl332 TBL.sh f4dh ghgrmp 9s ...................................FAIL! rc= 7
030 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7
033 bl337 TBL.sh f4dh fv2d_8tsk 9s ................................FAIL! rc= 7
037 bl371 TBL.sh f1.9dm fvvp_lb2 9s ...............................FAIL! rc= 7
041 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7
044 bl374 TBL.sh fm1.9dh outfrq3s 9s ..............................FAIL! rc= 7
056 bl992 TBL_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 7
- these are expected to due to changes in chemistry configurations
- the configurations that test cam3 and cam3_5_1 physics pass
calgary/lf95:
004 bl111 TBL.sh e8t5mdm ttrac 9s .................................FAIL! rc= 7
008 bl112 TBL.sh e8dm ghgrmp 9s ...................................FAIL! rc= 7
010 bl113 TBL.sh e8pdm aqpgro 3s ..................................FAIL! rc= 7
014 bl114 TBL.sh e8dm co2rmp 9s ...................................FAIL! rc= 7
021 bl311 TBL.sh f10t5mdm ttrac 9s ................................FAIL! rc= 7
029 bl312 TBL.sh f10dm ghgrmp 9s ..................................FAIL! rc= 7
031 bl313 TBL.sh f10pdm aqpgro 3s .................................FAIL! rc= 7
034 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 7
040 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7
- these are expected to due to changes in chemistry configurations
- the configurations that test cam3 and cam3_5_1 physics pass
calgary/pgi or jaguar/pgi:
004 bl111 TBL.sh e8t5mdm ttrac 9s .................................FAIL! rc= 7
008 bl112 TBL.sh e8dm ghgrmp 9s ...................................FAIL! rc= 7
010 bl113 TBL.sh e8pdm aqpgro 3s ..................................FAIL! rc= 7
014 bl114 TBL.sh e8dm co2rmp 9s ...................................FAIL! rc= 7
021 bl311 TBL.sh f10t5mdm ttrac 9s ................................FAIL! rc= 7
029 bl312 TBL.sh f10dm ghgrmp 9s ..................................FAIL! rc= 7
031 bl313 TBL.sh f10pdm aqpgro 3s .................................FAIL! rc= 7
034 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 7
040 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7
- these are expected to due to changes in chemistry configurations
- the configurations that test cam3 and cam3_5_1 physics pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_6_73
Originator(s): eaton
Date: Mon Dec 21 14:01:12 MST 2009
One-line Summary: update prescribed GHG data; update externals; misc fixes
Purpose of changes:
. update external references to be consistent with ccsm4_0_beta37
. Add new timer events for performance evaluation.
. Modify physics buffer error checking in pbuf_add to be consistent with
code in pbuf_get_fld_idx. The physics buffer names are assumed to be
case insensitive.
. Modify code in the parcel_dilute subroutine of zm_conv.F90 to avoid bug
in cases where the model top is below 40 mb.
Bugs fixed (include bugzilla ID):
. The values of atm_nx and atm_ny were not being set correctly in the
driver infobuffer for the HOMME dycore.
. The interpic_new Makefile had a bug in the logic for setting netcdf
default locations.
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets:
. add new IC file for T31 valid at Jan 01.
. update the prescribed GHG data for AR5 20th century runs.
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
SVN_EXTERNAL_DIRECTORIES
. update externals as follows:
drvseq3_0_40 --> drvseq3_1_08
clm3_6_54 --> clm3_6_58
cice4_0_20091104 --> cice4_0_20091202
docn7_090720 --> docn8_091208
share3_091026 --> share3_091208
MCT2_6_0_090317 --> MCT2_6_0_090926
scripts4_091102 --> scripts4_091208
models/atm/cam/bld/config_files/definition.xml
. allow -ocn to be set to 'docn'.
models/atm/cam/bld/configure
. modify the Filepaths file for the new driver directory structure.
. modify the Filepaths file for the new docn directory.
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. Add T31 IC file for Jan 01.
. update bndtvghg to use atm/cam/ggas/ghg_hist_1765-2005_c091218.nc
. update the 1.9x2.5 fsurdat and fpftdyn files. Last time I tried this I
models/atm/cam/bld/namelist_files/use_cases/1850-2005_track1.xml
. update bndtvghg to use atm/cam/ggas/ghg_hist_1765-2005_c091218.nc
models/atm/cam/src/dynamics/atm_comp_mct.F90
. set atm_nx and atm_ny values in infobuffer by using the dyn_grid method
get_horiz_grid_dim_d. Remove reference to pmgrid which should not appear
outside of the dynamics.
models/atm/cam/src/dynamics/homme/stepon.F90
. bugfix from Mark Taylor
models/atm/cam/src/physics/cam/ghg_data.F90
. fixed comment
models/atm/cam/src/physics/cam/phys_buffer.F90
. modify the pbuf_add error checking for a name that's already been defined
by making the name matching case insensitive. The code that returns the
buffer index for a specified name is doing case insensitive matching, so
pbuf_add needs to be consistent with that.
models/atm/cam/src/physics/cam/tphysac.F90
. add timer events for rayleigh friction and aerosol drydep
models/atm/cam/src/physics/cam/radiation.F90
models/atm/cam/src/physics/rrtmg/radiation.F90
. modify timer events for consistency between rrtmg and camrt code, and to
separate out the cost of the aerosol optics calc.
models/atm/cam/src/physics/cam/zm_conv.F90
. modify code in parcel_dilute to avoid bug in cases where the model top is
below 40 mb.
models/atm/cam/tools/interpic_new/Makefile
. fix bug in logic for setting defaults of netcdf lib and inc dirs.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
004 bl132 TBL.sh e32dh ghgrmp 9s ..................................FAIL! rc= 7 at Sat Dec 19 12:37:39 MST 2009
011 bl151 TBL.sh e64h outfrq24h 2d ................................FAIL! rc= 7 at Tue Dec 15 16:20:16 MST 2009
015 bl331 TBL.sh f4dh co2rmp 9s ...................................FAIL! rc= 7 at Tue Dec 15 16:30:08 MST 2009
019 bl332 TBL.sh f4dh ghgrmp 9s ...................................FAIL! rc= 7 at Tue Dec 15 16:36:56 MST 2009
030 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7 at Tue Dec 15 16:49:22 MST 2009
033 bl337 TBL.sh f4dh fv2d_8tsk 9s ................................FAIL! rc= 7 at Tue Dec 15 16:53:16 MST 2009
037 bl371 TBL.sh f1.9dm fvvp_lb2 9s ...............................FAIL! rc= 7 at Tue Dec 15 17:05:34 MST 2009
041 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7 at Tue Dec 15 17:20:21 MST 2009
044 bl374 TBL.sh fm1.9dh outfrq3s 9s ..............................FAIL! rc= 7 at Tue Dec 15 17:39:57 MST 2009
048 bl387 TBL.sh f1.9c351dh outfrq3s+1850-2005_track1 9s ..........FAIL! rc= 7 at Tue Dec 15 17:53:23 MST 2009
051 bl731 TBL.sh h16x4dm aqua 9s ..................................FAIL! rc= 7 at Tue Dec 15 17:58:06 MST 2009
056 bl992 TBL_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 7 at Tue Dec 15 18:36:55 MST 2009
dublin/lf95: All PASS except:
004 bl111 TBL.sh e8t5mdm ttrac 9s .................................FAIL! rc= 7 at Tue Dec 15 12:56:24 MST 2009
008 bl112 TBL.sh e8dm ghgrmp 9s ...................................FAIL! rc= 7 at Tue Dec 15 13:07:32 MST 2009
021 bl311 TBL.sh f10t5mdm ttrac 9s ................................FAIL! rc= 7 at Tue Dec 15 14:04:10 MST 2009
029 bl312 TBL.sh f10dm ghgrmp 9s ..................................FAIL! rc= 7 at Tue Dec 15 14:47:06 MST 2009
034 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 7 at Tue Dec 15 15:36:47 MST 2009
037 bl315 TBL.sh f10c3dm outfrq3s 9s ..............................FAIL! rc= 7 at Tue Dec 15 15:43:40 MST 2009
040 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7 at Tue Dec 15 16:00:22 MST 2009
dublin/pgi: All PASS except:
004 bl111 TBL.sh e8t5mdm ttrac 9s .................................FAIL! rc= 7 at Tue Dec 15 11:47:18 MST 2009
008 bl112 TBL.sh e8dm ghgrmp 9s ...................................FAIL! rc= 7 at Tue Dec 15 11:50:23 MST 2009
021 bl311 TBL.sh f10t5mdm ttrac 9s ................................FAIL! rc= 7 at Tue Dec 15 12:00:44 MST 2009
029 bl312 TBL.sh f10dm ghgrmp 9s ..................................FAIL! rc= 7 at Tue Dec 15 12:06:04 MST 2009
034 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 7 at Tue Dec 15 12:13:03 MST 2009
037 bl315 TBL.sh f10c3dm outfrq3s 9s ..............................FAIL! rc= 7 at Tue Dec 15 12:15:06 MST 2009
040 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7 at Tue Dec 15 12:16:32 MST 2009
Most of the baseline failures appear to be due to a roundoff level change
introduced into either CLM or CICE. For resolutions except FV 1.9x2.5 the
diffs don't show up at nstep=0, but are first present at nstep=3. For the 2
degree FV runs the diffs show up at nstep=0 but just in a few fields, and
the diffs again appear to be roundoff level changes. I suspect this is due
to updating the 2 degree CLM fsurdat file.
In addition to the differences just mentioned:
bl731 fails due to a homme bug fix.
bl112, bl132, bl312, bl332, bl387 fail due to updating the prescribed GHG
dataset.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: Answers change due to updating externals
and updating the prescribed GHG dataset. HOMME answers change due to a
bug fix.
===============================================================
===============================================================
Tag name: cam3_6_72
Originator(s): Francis Vitt
Date: 11 Dec 2009
One-line Summary: WACCM changes needed for climate runs.
Purpose of changes:
- compute "CFC11STAR" for radiation in WACCM configuration
- adjust inputs to the waccm_1850-2005_track1 use case
- misc cleanup
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
A models/atm/cam/src/chemistry/mozart/cfc11star.F90
- calculates "CFC11STAR" and addes it the the physics buffer for radiation
A models/atm/cam/bld/namelist_files/use_cases/waccm_1850-2005_track1.xml
- use case added for 1850-2005 transient waccm climate run
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- default solar_photons_file added
- defaults added for time dependent NO2 and BC aircraft emissions
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- "ippc_aircraft_emis" variable added which sets the ext_frc namelist
variables appropriately for the time dependent aircraft emissions
M models/atm/cam/bld/namelist_files/use_cases/waccm_1850_track1.xml
- adjustments to the waccm 1850 track1 use case
M models/atm/cam/bld/build-namelist
- specify "CFC11STAR" radiatively active by default for WACCM
- added "ipcc_aircraft_emis" namelist variable
- set xactive_drydep = .true. for waccm
- as default use the Lean solar data rather than Woods and Neckel data for waccm
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_waccm_mozart/chem_mods.F90
- NO2 external forcing added to mechanism for aircraft emissions
M models/atm/cam/src/chemistry/mozart/woods.F90
- initialize only if input file is specified
M models/atm/cam/src/chemistry/mozart/neckel.F90
- initialize only if input file is specified
M models/atm/cam/src/chemistry/mozart/chemistry.F90
- invocations of the cfc11star module
M models/atm/cam/src/chemistry/mozart/llnl_set_chem_trop.F90
- removed 'chem_trop_linoz' field from default history
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
030 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL
expected failure due to waccm changes in input data and cfc11star
calgary/lf95: all pass
calgary/pgi or jaguar/pgi: all pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_6_71
Originator(s): Jim Edwards
Date: 12/09/2009
One-line Summary: update homme dycore
Purpose of changes: Update HOMME dycore
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Remove homme requirement for lapack library
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/configure
Removed requirement for lapack
models/atm/cam/src/control/ncdio_atm.F90
Moved call to pio_setframe
models/atm/cam/src/dynamics/homme/
A host of new features and options added.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All pass except
051 bl731 TBL.sh h16x4dm aqua 9s ..................................FAIL
expected failure due to dycore update
dublin/lf95: All pass
calgary/pgi or jaguar/pgi: All pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_6_70
Originator(s): eaton
Date: Tue Dec 8 17:05:37 MST 2009
One-line Summary: fix memory leaks
Purpose of changes:
. fix memory leaks and EUL OMP bug (see below)
. misc configure/Makefile changes:
- enable building HOMME from CCSM scripts
- add flexibility to deal with new dycores in set_horiz_grid method
- remove Switch module from configure due to strange behavior on bluefire
Bugs fixed (include bugzilla ID):
. memory leaks (pointed out by Chuck Bardeen) due to missing deallocation
of allocated local variables in:
models/atm/cam/src/chemistry/utils/tracer_data.F90
models/atm/cam/src/dynamics/fv/dyn_comp.F90
models/atm/cam/src/physics/cam/polar_avg.F90
. fix OMP directive in eulerian dycore (missing private variable
declaration) (from Pat Worley)
Describe any changes made to build system: none
Describe any changes made to the namelist:
. add some HOMME initial files
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
models/atm/cam/doc/ChangeSum
models/atm/cam/doc/Copyright
models/atm/cam/doc/DatasetLog
models/atm/cam/doc/README
. old files relevent to earlier ccm/cam versions
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/Makefile.in
. remove --pca from lf95 debug flags. Documentation says it overrides the
effect of --chk a
. remove -DCAM from the CPPDEF macro. It is only used by HOMME, so is
added to the USER_CPPDEFS by configure.
models/atm/cam/bld/configure
. Remove check for smp=off when building for scam. Should run with threads
enabled even if they aren't kept busy.
. fix set_horiz_grid so that the fv grid recognition is specific to 'fv',
to allow extension to other dycores that start with fv, like fvcubed.
. add "-DCAM" to the cppdefs for the homme dycore. It is the only
component using this macro. Having configure set the macro means that it
will be available when running from the CCSM scripts as well as from CAM
standalone scripts.
. remove use of Switch module. We ran into some bizzare problems on
bluefire. The if/elsif conditional is just as easy to read.
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. update homme initial files to provide settings for non-aqua-planet runs.
models/atm/cam/src/chemistry/utils/tracer_data.F90
. remove unused allocatable arrays filename and filelist in subroutine
specify_fields.
models/atm/cam/src/dynamics/eul/sphdep.F90:
. add S_NVAL to omp private variables at line 413
models/atm/cam/src/dynamics/fv/dyn_comp.F90
. add missing deallocate for phisxy3 in dyn_run
models/atm/cam/src/physics/cam/polar_avg.F90
. deallocate lats, n_pole, s_pole in subroutines polar_average{2,3}d
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS
dublin/lf95: All PASS
dublin/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam3_6_69
Originator(s): Francis Vitt
Date: 24 Nov 2009
One-line Summary: Updates to super_fast_llnl/LINOZ chemistry
Purpose of changes:
Correct the LINOZ code and allow interactive "super_fast_llnl" chemistry
to be included in a "track1" coupled CCSM simulation..
The changes are summarized as:
* ISOP added to the super_fast_llnl chemistry mechanism
* CH4 is now prescribed
* The prescribed CH4 is radiatively active
* LINOZ chlorine loading is now read in from file
* Use case added for 1850 track 1 with super_fast_llnl chem
* FIXED solar data restart bug fix
* Corrections LINOZ code
* Added another tropopause method which is used in super_fast_llnl chemistry
* Added new super_fast_llnl mechanism file to preprocessor
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Brian Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
A models/atm/cam/bld/namelist_files/use_cases/1850_track1_super_fast_llnl.xml
- build namelist use case added for track1 with super_fast_llnl chemistry simulation
A models/atm/cam/src/chemistry/mozart/chlorine_loading_data.F90
- reads in time-dependent chlorine data for LINOZ
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- default input files added for the prescribed CH4 and chlorine loading used
in the super_fast_llnl chemistry
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- namelist variables added for the chlorine loading input data
M models/atm/cam/bld/build-namelist
- prognostic SO4 and prescribed CH4 is now radiatively active by default
for super_fast_llnl chemistry configuration
- surface emissions for ISOP added to the default super_fast_llnl
chemistry namelist
- vertical emmisions of SO2 and SO4 added to the default
super_fast_llnl chemistry namelist
- prescribed CH4 namelist variable settings added to the default
super_fast_llnl chemistry namelist
M models/atm/cam/bld/perl5lib/Build/ChemPreprocess.pm
- use the new super_fast_llnl chemistry preprocessor input file
M models/atm/cam/src/chemistry/utils/solar_data.F90
- FIXED solar data restart bug fix
M models/atm/cam/src/chemistry/utils/prescribed_ghg.F90
- minor code cleanup
M models/atm/cam/src/chemistry/utils/tracer_data.F90
- fixed bug in datapath specification
- out log info in masterproc only
M models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_lu_solve.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_setrxt.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/m_rxt_id.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_adjrxt.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_lu_factor.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_indprd.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_sim_dat.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/m_spc_id.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/chem_mods.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_prod_loss.F90
M models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_lin_matrix.F90
- preproccessor generated file for the new super_fast_llnl chemistry mechanism
M models/atm/cam/src/chemistry/mozart/lin_strat_chem.F90
- corrections to the LINOZ code
M models/atm/cam/src/chemistry/mozart/tracer_cnst.F90
M models/atm/cam/src/chemistry/mozart/tracer_srcs.F90
- fix bug in data_type specifier (INTERP_MISSING_MONTHS)
M models/atm/cam/src/chemistry/mozart/chemistry.F90
- namelist variables added for chlorine loading inputs
- subroutine calls added to invoke chlorine loading module
- default method for tropopause set to "Stobie-LINOZ hybrid"
M models/atm/cam/src/chemistry/mozart/llnl_set_chem_trop.F90
- "Stobie-LINOZ hybrid" method added to find tropopause
- diagnostic outputs added
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
- updated chemistry preprocessor input files
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All Pass
calgary/lf95: All Pass
calgary/pgi or jaguar/pgi: All Pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_6_68
Originator(s): eaton
Date: Thu Nov 19 14:56:35 MST 2009
One-line Summary: tuning for 1.9x2.5 track1 runs
Purpose of changes:
. The default value of rhminl was changed to 0.91, and a multiplier for the
ice fall velocity in pkg_cld_sediment.F90 was set to 0.5. Because this
multiplier depends both on resolution and the physics/chemistry packages
it was made settable via the namelist.
These mods only change answers for FV 1.9x2.5 (a.k.a. 2 deg) track1 runs.
The changes were made so that the tunings for that configuration match
the ones used in WACCM runs.
. The WACCM use case waccm_1850_track1 was updated to be consistent with
the prescribed aerosol changes made in the 1850_track1 use case in the
previous tag. This will change WACCM answers for runs using that use
case for all horizontal resolutions except 0.9x1.25.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist:
. add namelist group cldsed_nl which includes the variable
cldsed_ice_stokes_fac used to set the factor applied to the ice fall
velocity computed from Stokes terminal velocity.
List any changes to the defaults for the boundary datasets:
. defaults for prescribed aerosols changed in waccm_1850_track1 use case.
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/build-namelist
. add an add_default call for cldsed_ice_stokes_fac
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. change cldfrc_rhminl for 1.9x2.5 track1 to match the waccm_mozart value
of 0.91
. add defaults for cldsed_ice_stokes_fac. Default remains 1.0 for
backwards compatibility with older cam physics packages. Change default
to 0.5 for any waccm_phys run, and also for 2 degree track1 runs.
models/atm/cam/bld/namelist_files/namelist_definition.xml
. add cldsed_ice_stokes_fac
models/atm/cam/bld/namelist_files/use_cases/waccm_1850_track1.xml
. make the prescribed aerosols consistent with 1850_track1.xml
models/atm/cam/src/control/runtime_opts.F90
. add call to cld_sediment_readnl
models/atm/cam/src/physics/cam/pkg_cld_sediment.F90
. add cld_sediment_readnl to read namelist
. add namelist variable cldsed_ice_stokes_fac to set the factor to be
applied to the ice fall velocity via the Stokes calculation.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except
048 bl387 TBL.sh f1.9c351dh outfrq3s+1850-2005_track1 9s ..........FAIL! rc= 7 at Thu Nov 19 15:05:24 MST 2009
This is a 2 deg track1 baseline which is expected to fail.
**Note**
050 er731 TER.sh h16x4dm aqua 4+5s ................................PASS at Thu Nov 19 15:09:59 MST 2009
We've been seeing some erratic behavior with this test. I had both failures
and passes of it during testing for the previous tag. More examination of
this test is needed.
calgary/lf95: All PASS
calgary/pgi or jaguar/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB except 2 deg track1 and WACCM runs
using the waccm_1850_track1 use case.
===============================================================
===============================================================
Tag name: cam3_6_67
Originator(s): jedwards, eaton
Date: Wed Nov 18 11:44:51 MST 2009
One-line Summary: update prescribed aerosol and ozone datasets.
Purpose of changes:
. bug fixes and updated datasets -- see below.
Bugs fixed (include bugzilla ID):
. The attributes _FillValue and missing_value were being written to the
history file as 8-byte reals even when the corresponding data value were
written as 4-byte reals. These attributes must match the data type for
successful post-processing. This bug was introduced at cam3_6_39 with
the changes to pio.
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets:
. The prescribed aerosol and ozone datasets have been updated. The new
data will be the default in the 1850_track1 and 1850_track1_2xco2 use
cases for all resolutions except FV 0.9x1.25. This is for backwards
compatibility with the existing control run. The new data will be the
default for the 1850-2005_track1 use case.
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/namelist_files/use_cases/1850-2005_track1.xml
. update prescribed aerosol and ozone datasets
models/atm/cam/bld/namelist_files/use_cases/1850_track1.xml
models/atm/cam/bld/namelist_files/use_cases/1850_track1_2xco2.xml
. update prescribed aerosol and ozone datasets for all resolutions except
FV 0.9x1.25 (for backwards compatibility with control run).
models/atm/cam/src/control/cam_history.F90
. Fix the real kind of data written to the attributes _FillValue and
missing_value.
. Remove old logic for "fullgrid". The reduced grid option is not functional.
models/atm/cam/src/dynamics/homme/external/prim_driver_mod.F90
models/atm/cam/src/dynamics/homme/external/prim_state_mod.F90
. mods from Jim Edwards
models/atm/cam/test/system/TCB.sh
. mod to allow stub ice model
models/atm/cam/test/system/config_files/h16x4dm
models/atm/cam/test/system/config_files/h5x8adm
models/atm/cam/test/system/config_files/h5x8dm
models/atm/cam/test/system/config_files/h5x8idm
. use stub ice and lnd in all these standalone mode configs which
only supports adiabatic, ideal, or aqua-planet physics.
models/atm/cam/test/system/CAM_runcmnd.sh
models/atm/cam/test/system/test_driver.sh
. add edinburgh
models/atm/cam/test/system/tests_posttag_jaguar_cb
. add cb372
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except
004 bl387 TBL.sh f1.9c351dh outfrq3s+1850-2005_track1 9s ..........FAIL! rc= 7 at Tue Nov 17 15:09:41 MST 2009
006 er731 TER.sh h16x4dm aqua 4+5s ................................FAIL! rc= 10 at Tue Nov 17 15:10:36 MST 2009
bl387 fails because the new prescribed aerosol and ozone datasets changes
answers.
er731 failure is unexplained. cam3_6_66 is also exhibiting this failure
even though the test passed at the time of that commit on Nov 5.
dublin/lf95: All PASS
dublin/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB except configurations using the
updated prescribed aerosol and ozone datasets.
===============================================================
===============================================================
Tag name: cam3_6_66
Originator(s): mvertens, eaton
Date: Thu Nov 5 10:11:53 MST 2009
One-line Summary: add SNOWHICE to diagnostic output; update externals
Purpose of changes:
. Add SNOWHICE to the default monthly avg history output.
. Update externals to be mostly consistent with ccsm4_0_beta33
drvseq3_0_29 --> drvseq3_0_40
clm3_6_51 --> clm3_6_54
cice4_0_20090831 --> cice4_0_20091104
share3_090825 --> share3_091026
timing_081221 --> timing_090929
scripts4_090901a --> scripts4_091102
stubs1_1_01 --> stubs1_2_02
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none for CAM
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/build-namelist
. remove forganic (no longer used by CLM)
models/atm/cam/bld/configure
. updates for directory paths used by stub components
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. update defaults for CLM files
models/atm/cam/bld/namelist_files/namelist_definition.xml
. two driver pio variable name changes
models/atm/cam/src/control/camsrfexch_types.F90
models/atm/cam/src/cpl_mct/atm_comp_mct.F90
models/atm/cam/src/physics/cam/cam_diagnostics.F90
. add SNOWHICE to default monthly avg output
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except
056 bl992 TBL_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 7 at Thu Nov 5 11:54:53 MST 2009
This failure is due to updating the CICE external. CICE is run in
prognostic mode in the E compsets (SOM mode).
dublin/lf95: All PASS
dublin/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam3_6_65
Originator(s): eaton
Date: Mon Nov 2 14:16:30 MST 2009
One-line Summary: Add output for CCR diagnostics, and to verify forcings
for 20th century track1 runs.
Purpose of changes:
. Add diagnostic fields needed by the CCR package and for verifying that
the intended forcings have been used in the track1 20th century
simulations.
- The scalar GHG and solar const values were added to history output.
co2vmr was already there. The values n2ovmr, ch4vmr, f11vmr, f12vmr,
sol_tsi have been added to all history files.
- PRECT was added by default to monthly avg output.
- The aerosol optical depths (visible) of all species affecting the
climate have been added by default to the monthly avg output.
- The ability to output the column burdens of all aerosol and gas species
that affect the climate was added. The burdens of SO4 and O3 have been
added to the default monthly avg file (via fincl1) for the
1850-2005_track1 use case.
. Modify regression tests to improve coverage of the production 1850_track1
and 1850-2005_track1 use cases.
. Mod to configure needed for building for BGL and BGP from CCSM scripts.
. Mods to PGI settings in Makefile.in for consistency with CCSM scripts.
. Misc cleanup
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do:
models/atm/cam/test/system/config_files/f1.9c351h
models/atm/cam/test/system/config_files/f1.9c351m
. new files for track1 physics (c351 ==> cam3_5_1)
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/Makefile.in
. PGI settings:
- change optimization from -fast to -O2 (keep -Mvect=nosse -Kieee)
- apply the threading options to FFLAGS_NOOPT macro
- remove special compilation rule for seq_timemgr_mod.o for consistency
with CCSM build.
models/atm/cam/bld/build-namelist
. fix logic for when defaults are added for co2vmr. That value is needed
for the RAMP_CO2_ONLY option and for waccm_mozart which were previously not
adding defaults for co2vmr (hence the hardcoded value in chem_surfvals
was being used).
. remove redundant specification of default values for tracer_cnst_type and
tracer_cnst_ymd which are present in the namelist_defaults_cam.xml file.
models/atm/cam/bld/configure
. Only add the Fortran name mangling macro to the cppdefs list for a cam
standalone build. When build from the CCSM scripts they take care of
setting this macro.
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
models/atm/cam/bld/namelist_files/namelist_definition.xml
. add nu_q definition and defaults (for homme)
models/atm/cam/bld/namelist_files/use_cases/1850-2005_track1.xml
. add cb_ozone_c and cb_sulf_c to fincl1 list
models/atm/cam/src/chemistry/utils/solar_data.F90
. initialize sol_tsi to -1.0. Default value set by build-namelist.
models/atm/cam/src/control/cam_history.F90
. add n2ovmr, ch4vmr, f11vmr, f12vmr, sol_tsi
models/atm/cam/src/control/hycoef.F90
. add initializers for module data ps0, psr, and change them to parameters.
models/atm/cam/src/physics/cam/aer_rad_props.F90
. add an add_default call for the individual aerosol optical depth
(visible) fields.
models/atm/cam/src/physics/cam/cam_diagnostics.F90
. add PRECT to default monthly avg history file
models/atm/cam/src/physics/cam/chem_surfvals.F90
. change initialization values of co2vmr, n2ovmr, ch4vmr, f11vmr, f12vmr to
-1.0. The default values are set by build-namelist.
models/atm/cam/src/physics/cam/rad_constituents.F90
. add diagnostics for column burden of both gases and aerosols
models/atm/cam/src/physics/cam/radiation.F90
models/atm/cam/src/physics/rrtmg/radiation.F90
. replace rad_cnst_clim_aer_out by generic rad_cnst_out
models/atm/cam/src/utils/pio_utils.F90
. remove base=0 optional arg that's hardcoded in the init_pio_subsystem
call.
models/atm/cam/src/control/runtime_opts.F90
models/atm/cam/src/utils/time_manager.F90
. remove timemgr_preset from time_manager.F90 and remove the call to
timemgr_preset from subroutine preset in module runtime_opts. dtime is
the only thing being set and setting this value has been moved to
build-namelist.
models/atm/cam/test/system/CAM_runcmnd.sh
. add -d ${CAM_THREADS} to aprun command for hybrid mode.
models/atm/cam/test/system/input_tests_master
. define new tests:
- sm372 TSM.sh f1.9m fvvp_lb2 9s
- er372 TER.sh f1.9m fvvp_lb2 4+5s
- bl372 TBL.sh f1.9m fvvp_lb2 9s
- eq374 TEQ.sh f1.9m fvvp_lb2 f1.9m fv1d_lb2 9s
- eq375 TEQ.sh f1.9m fvvp_lb2 f1.9m fvvp_lb0 9s
. change definition:
- sm379 TSM.sh wm1.9m outfrq3s 9s
changed to sm379 TSM.sh wm1.9m outfrq3s+waccm_1850_track1 9s
- er379 TER.sh wm1.9m outfrq3s 4+5s
changed to er379 TER.sh wm1.9m outfrq3s+waccm_1850_track1 4+5s
- bl379 TBL.sh wm1.9m outfrq3s 9s
changed to bl379 TBL.sh wm1.9m outfrq3s+waccm_1850_track1 9s
- cb386 TCB.sh f1.9m
changed to cb386 TCB.sh f1.9c351m
- sm385 TSM.sh f1.9h outfrq24h+1870_control 2m
changed to sm385 TSM.sh f1.9c351h outfrq24h+1850_track1 2m
- sm386 TSM.sh f1.9m outfrq24h+1870_control 2m
changed to sm386 TSM.sh f1.9c351m outfrq24h+1850_track1 2m
- er385 TER.sh f1.9h outfrq24h+1870_control 1+1m
changed to er385 TER.sh f1.9c351h outfrq24h+1850_track1 1+1m
- er386 TER.sh f1.9m outfrq24h+1870_control 1+1m
changed to er386 TER.sh f1.9c351m outfrq24h+1850_track1 1+1m
- bl385 TBL.sh f1.9h outfrq24h+1870_control 2m
changed to bl385 TBL.sh f1.9c351h outfrq24h+1850_track1 2m
- bl386 TBL.sh f1.9m outfrq24h+1870_control 2m
changed to bl386 TBL.sh f1.9c351m outfrq24h+1850_track1 2m
- pf385 TPF.sh f1.9h outfrq24h+1870_control 2m
changed to pf385 TPF.sh f1.9c351h outfrq24h+1850_track1 2m
- pf386 TPF.sh f1.9m outfrq24h+1870_control 2m
changed to pf386 TPF.sh f1.9c351m outfrq24h+1850_track1 2m
models/atm/cam/test/system/tests_posttag_jaguar
. replace sm371 er371 bl371 eq372 eq373
by sm372 er372 bl372 eq374 eq375
models/atm/cam/test/system/tests_posttag_jaguar_cb
. remove cb371
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
015 bl331 TBL.sh f4dh co2rmp 9s ...................................FAIL! rc= 7 at Mon Nov 2 12:49:10 MST 2009
030 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7 at Mon Nov 2 13:00:02 MST 2009
dublin/lf95: All PASS except:
014 bl114 TBL.sh e8dm co2rmp 9s ...................................FAIL! rc= 7 at Sun Nov 1 12:18:36 MST 2009
dublin/pgi: All PASS except:
014 bl114 TBL.sh e8dm co2rmp 9s ...................................FAIL! rc= 7 at Sat Oct 31 16:46:56 MDT 2009
The above baseline failures are all due to removing the hardcoded GHG
values from the chem_surfvals module and requiring that these values be set
via the namelist. The failed tests were all making use of the hardcoded
value of co2vmr, and that value changed when the default came from the
namelist.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB except for configurations that
were using the hardcoded GHG values in chem_surfvals.
===============================================================
===============================================================
Tag name: cam3_6_64
Originator(s): Francis Vitt
Date: 27 Oct 2009
One-line Summary: Miscellaneous chemistry cleanup and bug fixes
Purpose of changes:
The main purpose is to make waccm physics features available to
tropospheric chemistry configurations. The changes are summarized as:
- added ability to configure waccm physics with any chemistry configuration
- fix needed to customize chemistry mechanism without dust and sea salt
- fix needed to get troposphere level with customized chemistry mech without
prognostic ozone present
- cleaned up physics/cam/chemistry.F90 -- removed deprecated chemistry code
- cleaned up aerosol_drydep_intr interface -- removed MODAL_AERO cpp def
- fix in waccm ctem module for cases where more processors are
operating on physics than on dynamics
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Brian Eaton
List all subroutines eliminated:
D models/atm/cam/bld/chem_preprocess.pl
Moved to the bld/perl5lib/Build/ChemPreprocess.pm.
List all subroutines added and what they do:
A models/atm/cam/bld/perl5lib/Build/ChemPreprocess.pm
Chemistry preprocessor perl code file was moved to the bld/perl5lib/Build
subdirectory and implemented as a perl module. A check for prognostic
sea salt and dust was added which was needed for customized chemistry
mechanisms.
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/configure
M models/atm/cam/bld/config_files/definition.xml
M models/atm/cam/bld/config_files/defaults_waccm_ghg.xml
M models/atm/cam/bld/config_files/defaults_waccm_mozart.xml
Option "-waccm_phys" was added to get the users the ability to
configure with waccm physics for any of the chemistry packages.
M models/atm/cam/bld/build-namelist
Set the appropriate gravity wave namelist when configured with -waccm_phys option
M models/atm/cam/src/control/runtime_opts.F90
Misc cleanup.
Replaced cpp flags WACCM_GHG, WACCM_MOZART with WACCM_PHYS.
M models/atm/cam/src/control/camsrfexch_types.F90
Nullify ram1 and fv pointers before they are allocated.
M models/atm/cam/src/cpl_mct/atm_comp_mct.F90
Removed DUST and PROGSSLT cpp flags .
M models/atm/cam/src/physics/cam/initindx.F90
M models/atm/cam/src/physics/cam/trb_mtn_stress.F90
M models/atm/cam/src/physics/cam/cldwat2m.F90
M models/atm/cam/src/physics/cam/physics_types.F90
M models/atm/cam/src/physics/cam/physpkg.F90
M models/atm/cam/src/physics/cam/esinti.F90
M models/atm/cam/src/physics/cam/pkg_cld_sediment.F90
M models/atm/cam/src/physics/cam/convect_deep.F90
M models/atm/cam/src/physics/cam/pkg_cldoptics.F90
M models/atm/cam/src/physics/cam/advnce.F90
M models/atm/cam/src/physics/cam/cldwat.F90
M models/atm/cam/src/physics/cam3_6/cldwat2m.F90
M models/atm/cam/src/dynamics/fv/dp_coupling.F90
Replaced cpp flags WACCM_GHG, WACCM_MOZART with WACCM_PHYS.
This makes waccm physics available to tropospheric chemistry
configurations.
M models/atm/cam/src/physics/cam/tphysbc.F90
M models/atm/cam/src/physics/cam/tphysac.F90
Made clybry_fam module available to TROCHEM configuration.
M models/atm/cam/src/physics/cam/cloud_fraction.F90
Cleanup -- removed to not-used waccm check.
M models/atm/cam/src/physics/cam/chemistry.F90
Cleanup -- removed to not-used chemistry code .
M models/atm/cam/src/physics/waccm/ctem.F90
Fix bug when more PEs are used in physics than in dynamics.
M models/atm/cam/src/physics/waccm/radheat.F90
Add a check for waccm heating.
Misc clean up.
M models/atm/cam/src/physics/waccm/gw_drag.F90
Jack Chen's changes:
1) change the source level of Beres scheme to the top of heating instead of 100 mb.
2) momentum/energy tendency below gravity wave source level to ensure the
total momentum/energy of the entire column is conserved.
M models/atm/cam/src/physics/waccm/qbo.F90
M models/atm/cam/src/chemistry/mozart/mo_setz.F90
Misc clean up.
M models/atm/cam/src/chemistry/bulk_aero/aerosol_intr.F90
Removed the MODAL_AERO cpp flag from the aerosol_drydep_intr subroutine
interface.
M models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
M models/atm/cam/src/chemistry/mozart/llnl_set_chem_trop.F90
Fixed bug where the tropopause level was not set when mechanism does
not have prognostic ozone.
M models/atm/cam/src/chemistry/mozart/mo_tropopause.F90
Removed the check for the need for tropopause level.
M models/atm/cam/src/chemistry/mozart/chemistry.F90
Made stratosphere aerosol feedback available to troposphere chemistry
configurations.
M models/atm/cam/src/chemistry/mozart/mo_waccm_hrates.F90
Made has_hrates variable public.
M models/atm/cam/src/chemistry/mozart/mo_tuv_inti.F90
Added "_r8" kind specifier to zero constant.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
030 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7
- Failed due to changes in waccm/gw_drag.F90.
calgary/lf95:
034 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 7
- Failed due to changes in waccm/gw_drag.F90 and physics for waccm_ghg.
calgary/pgi or jaguar/pgi:
034 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 7
- Failed due to changes in waccm/gw_drag.F90 and physics for waccm_ghg.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_6_63
Originator(s): Jim Edwards
Date: 10-19-2009
One-line Summary: update of tropopause interpolation for cs grids
Purpose of changes: code development and cleanup
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: self
List all subroutines eliminated:
control/landfrac_file.F90
no longer used
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
control/runtime_opts.F90
some namelist strings were declared too short for possible options
control/ncdio_atm.F90
control/interpolate_data.F90
control/startup_initialconds.F90
utils/pio_utils.F90
dynamics/eul/inidat.F90
dynamics/homme/inidat.F90
dynamics/fv/inidat.F90
code cleanup, added support for files with more than one timelevel
physics/cam/tropopause.F90
physics/cam/phys_prop.F90
physics/cam/modal_aer_opt.F90
physics/cam/radae.F90
update IO method to PIO, update tropopause interpolation to one compatible
with unstructured grids. This change causes some non bfb differences in
tropopause level diagnostic fields.
chemistry/utils/mo_solar_parms.F90
chemistry/utils/tracer_data.F90
Update IO method to PIO
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All pass except
037 bl371 TBL.sh f1.9dm fvvp_lb2 9s
041 bl375 TBL.sh f1.9h outfrq24h 2d
048 bl387 TBL.sh f1.9c351dh outfrq3s+1850-2005_track1 9s
dublin/lf95: All pass except
004 bl111 TBL.sh e8t5mdm ttrac 9s
008 bl112 TBL.sh e8dm ghgrmp 9s
014 bl114 TBL.sh e8dm co2rmp 9s
calgary/pgi or jaguar/pgi: All pass except
004 bl111 TBL.sh e8t5mdm ttrac 9s
008 bl112 TBL.sh e8dm ghgrmp 9s
014 bl114 TBL.sh e8dm co2rmp 9s
All failures due to differences in diagnostic tropopause height fields due to change in
the interpolation algorythm for climatological tropopause file.
These fields are diagnostic only and do not feed back to model prognostics.
Differences may be greater than roundoff but generally in the 6th digit or less.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_6_62
Originator(s): Jim Edwards / Pat Worley / Mark Taylor
Date: 8-13-2009
One-line Summary: Homme Dynamics update/ handshake logic update / pio interface update
Purpose of changes: Code development and improvement
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Self
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
Changed mpi handshake logic to avoid potential deadlock.
M 18837 models/utils/pilgrim/mod_comm.F90
M 18837 models/atm/cam/src/utils/spmd_utils.F90
Added a timestamp to the status message printed
M 18837 models/atm/cam/test/system/test_driver.sh
Moved initialization of surface fields sgh, sgh30,and landm from dynamics to physics
M 18837 models/atm/cam/src/physics/cam/comsrf.F90
M 18837 models/atm/cam/src/physics/cam/physpkg.F90
M 18837 models/atm/cam/src/dynamics/sld/inidat.F90
M 18837 models/atm/cam/src/dynamics/eul/inidat.F90
M 18837 models/atm/cam/src/dynamics/homme/inidat.F90
M 18837 models/atm/cam/src/dynamics/fv/inidat.F90
New pio interface no longer requires arrays to be flattened before passing
M 18837 models/atm/cam/src/physics/cam/phys_buffer.F90
M 18837 models/atm/cam/src/control/ncdio_atm.F90
M 18837 models/atm/cam/src/control/cam_history.F90
M 18837 models/atm/cam/src/physics/cam/restart_physics.F90
Message to stdout limited to masterproc
M 18837 models/atm/cam/src/chemistry/mozart/mo_jlong.F90
Homme dynamics updated with support for tracer/dynamics subcycling
M 18837 models/atm/cam/src/dynamics/homme/external/filter_mod.F90
M 18837 models/atm/cam/src/dynamics/homme/external/diffusion_mod.F90
M 18837 models/atm/cam/src/dynamics/homme/external/prim_advance_mod.F90
M 18837 models/atm/cam/src/dynamics/homme/external/derivative_mod.F90
M 18837 models/atm/cam/src/dynamics/homme/external/prim_state_mod.F90
M 18837 models/atm/cam/src/dynamics/homme/external/dimensions_mod.F90
M 18837 models/atm/cam/src/dynamics/homme/external/coordinate_systems_mod.F90
M 18837 models/atm/cam/src/dynamics/homme/external/global_norms_mod.F90
M 18837 models/atm/cam/src/dynamics/homme/external/control_mod.F90
M 18837 models/atm/cam/src/dynamics/homme/external/prim_driver_mod.F90
M 18837 models/atm/cam/src/dynamics/homme/external/interpolate_mod.F90
M 18837 models/atm/cam/src/dynamics/homme/external/element_mod.F90
M 18837 models/atm/cam/src/dynamics/homme/external/viscosity_mod.F90
M 18837 models/atm/cam/src/dynamics/homme/external/prim_advection_mod.F90
M 18837 models/atm/cam/src/dynamics/homme/external/edge_mod.F90
M 18837 models/atm/cam/src/dynamics/homme/external/namelist_mod.F90
M 18837 models/atm/cam/src/dynamics/homme/dp_coupling.F90
M 18837 models/atm/cam/src/dynamics/homme/stepon.F90
M 18837 models/atm/cam/src/dynamics/homme/dyn_comp.F90
M 18837 models/atm/cam/src/dynamics/homme/restart_dynamics.F90
Updated PIO
M 18837 SVN_EXTERNAL_DIRECTORIES
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: all pass except
051 bl731 TBL.sh h16x4dm aqua 9s ..................................FAIL
expected due to ongoing development of dycore.
dublin/lf95: all pass
calgary/pgi or jaguar/pgi: calgary - all pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
Tag name: cam3_6_61
Originator(s): pworley
Date: Wed Sep 30 2009
One-line Summary: spectral Eulerian dycore performance optimizations;
T341 support; new repro_sum option
Purpose of changes: Improve performance of spectral Eulerian dycore when
have exhausted MPI parallelism; introduce new, cheaper, reproducible sum
option; add support for T341 resolution
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
* added OUTER_OMP cpp token to revert to original implementation of
OpenMP in spectral Eulerian dycore
Describe any changes made to the namelist:
* added phys_twin_algorithm to cam_inparm to specify whether to use
twin algorithm in creating physics chunks (potentially overriding default)
* added repro_sum_use_ddpdd in cam_inparm to select alternative
reproducible sum algorithm
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
* Can improve performance of spectral Eulerian dycore significantly (~2x)
on systems with OpenMP parallelism when using large MPI process counts
* Can degrade performance of spectral Eulerian dycore slightly on systems
with OpenMP parallelism when using small MPI process counts. Adding
OUTER_OMP to cpp tokens when compiling will restore original performance
Code reviewed by: self
List all subroutines eliminated:
* usetlbuf in cam/src/control/usetlbuf.c
call to the routine was eliminated in cam3_4_04. Eliminating file now.
List all subroutines added and what they do:
* fc_gathervc in cam/src/utils/spmd_utils.F90
flow control gather for character strings
* repro_sum_ddpdd and DDPDD in cam/src/utils/repro_sum_mod.F90
He and Ding's implementation of the double-double reproducible sum
algorithm
* x86_fix_start and x86_fix_end in cam/src/utils/repro_sum_x86.c
routines to set and restore round-to-double flag in the control word
of a x86 fpu (required for double-double algorithm)
List all existing files that have been modified, and describe the changes:
1) The primary performance optimization has been moving OpenMP off of the loops
that are also decomposed with MPI and onto inner loops. Some of these loops
have also been restructured to improve the effectiveness of the OpenMP
parallelism. Legacy Cray vectorization and streaming directives were also
removed during this process.
src/advection/slt/bandij.F90
src/advection/slt/difcor.F90
src/advection/slt/kdpfnd.F90
src/advection/slt/omcalc.F90
src/advection/slt/pdelb0.F90
src/advection/slt/plevs0.F90
src/advection/slt/qmassa.F90
src/advection/slt/xqmass.F90
src/dynamics/eul/cubydr.F90
src/dynamics/eul/cubzdr.F90
src/dynamics/eul/dyndrv.F90
src/dynamics/eul/dp_coupling.F90
src/dynamics/eul/grmult.F90
src/dynamics/eul/herxin.F90
src/dynamics/eul/heryin.F90
src/dynamics/eul/herzin.F90
src/dynamics/eul/lagyin.F90
src/dynamics/eul/limdy.F90
src/dynamics/eul/limdz.F90
src/dynamics/eul/linemsdyn.F90
src/dynamics/eul/quad.F90
src/dynamics/eul/realloc4.F90
src/dynamics/eul/realloc7.F90
src/dynamics/eul/scan2.F90
src/dynamics/eul/scandyn.F90
src/dynamics/eul/scanslt.F90
src/dynamics/eul/spegrd.F90
src/dynamics/eul/sphdep.F90
src/dynamics/eul/stats.F90
src/dynamics/eul/tfilt_massfix.F90
src/dynamics/eul/trjmps.F90
src/dynamics/eul/tstep.F90
src/physics/cam/virtem.F90
2) Added support for T341 resolution
bld/config_files/horiz_grid.xml
src/dynamics/eul/dycore.F90
src/physics/cam/cldwat.F90
src/physics/cam/cloud_fraction.F90
src/physics/cam/hk_conv.F90
src/physics/cam/zm_conv.F90
3) Added a flow control version of gather used to determine names of nodes
during intialization
src/utils/spmd_utils.F90
4) Added DDPDD alternative reproducible sum algorithm.
src/utils/repro_sum_mod.F90
src/utils/repro_sum_x86.c
src/control/mpishorthand.F
5) Added namelist support for choice of algorithm to use in creating physics
chunks. The default is for lon/lat grids to look for north/south; day/night
"twins" when assigning columns to chunks, and not to do this for unstructured
grids. Setting phys_twin_algorithm=.true. will force the twin algorithm to be
used. Setting phys_twin_algorithm=.false. will disable the twin algorithm.
Also added namelist support for the DDPDD alternative reproducible
sum algorithm. Setting repro_sum_use_ddpdd=.true. will enable this option.
bld/namelist_files/namelist_definition.xml
src/control/runtime_opts.F90
6) MPI communication optimization
src/dynamics/eul/bndexch.F90
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: PASSED
calgary/lf95: PASSED
calgary/pgi: PASSED
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers:
NONE (bit for bit)
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
Tag name: cam3_6_60
Originator(s): rneale,mvr
Date: Mon Sep 28 2009
One-line Summary: additional output fields for ipcc analysis; fix to scam test;
code work-arounds to enable testing with intel compilers; adjustment to waccm
tuning parameter
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: eaton
List all subroutines eliminated:
D models/atm/cam/test/system/nl_files/scm_prep_35
- removing obsolete namelist options test file
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/config_files/scm64bfbiop_35
M models/atm/cam/test/system/config_files/e64bfbiop_35
M models/atm/cam/test/system/input_tests_master
- corrected config settings for testing scam with cam3_5_1 physics
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- change to tuning parameter (rhminl) default for waccm at 1.9x2.5
M models/atm/cam/src/physics/cam/radiation.F90
M models/atm/cam/src/physics/cam/cam_diagnostics.F90
M models/atm/cam/src/physics/cam/radlw.F90
M models/atm/cam/src/physics/cam/physpkg.F90
M models/atm/cam/src/physics/cam/convect_shallow.F90
M models/atm/cam/src/physics/cam/zm_conv_intr.F90
- added output of fields required for ipcc analysis
M models/atm/cam/src/dynamics/sld/inidat.F90
M models/atm/cam/src/dynamics/sld/restart_dynamics.F90
M models/atm/cam/src/advection/slt/phcs.F90
- code work-arounds to enable testing with intel compilers
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:none
calgary/lf95:none
calgary/pgi:none
CAM tag used for the baseline comparison tests if different than previous
Summarize any changes to answers: bfb
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
Tag name: cam3_6_59
Originator(s): jet, jedwards
Date: Tue Sep 15 15:24:42 MDT 2009
One-line Summary:Bug fixes for scam and CAM column output
Purpose of changes:fixes a problem with restarts when using column output
and bug fixes for scam aerosol deposition
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:na
Describe any changes made to the namelist:na
List any changes to the defaults for the boundary datasets:na
Describe any substantial timing or memory changes:na
Code reviewed by:jet
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/src/control/cam_history.F90
fix for column output
M models/atm/cam/src/utils/spmd_utils.F90
fix for column output
M models/atm/cam/src/chemistry/mozart/mo_drydep.F90
fix for scam dry deposition
M models/atm/cam/src/dynamics/eul/inidat.F90
fix for scam dry deposition
M models/atm/cam/src/dynamics/eul/iop.F90
fix for scam dry deposition
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:none
calgary/lf95:
calgary/pgi or jaguar/pgi:none
CAM tag used for the baseline comparison tests if different than previous
tag:cam3_6_58
Summarize any changes to answers, i.e.,
Default model is bfb
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
Tag name: cam3_6_58
Originator(s): eaton, jedwards
Date: Thu Sep 3 11:44:33 MDT 2009
One-line Summary: Update externals; add cloud fraction tuning parameters to namelist.
Purpose of changes:
. Update externals to be consistent with ccsm4_0_beta22:
cice4_0_20090805 --> cice4_0_20090831
clm3_6_47 --> clm3_6_51
drvseq3_0_24 --> drvseq3_0_29
share3_090706b --> share3_090825
scripts4_090722b --> scripts4_090901a
. Move the cloud fraction tuning parameters rhminl, rhminh to the
namelist. The code that previously set these parameters in cldfrc_init
has been replaced by entries in the namelist_defaults_cam.xml file. The
logic for supported tunings has been simplified by making use of the
high level physics package attribute. So now the default settings are
available for the track1 and track5 physics packages (corresponding to
the -phys values cam3_5_1 and cam), and some chemistry packages. And of
course the tunings remain resolution dependent.
Most importantly, when experimenting with tunings for any particular
model configuration, the user only needs to add values to the namelist
input rather than having to wade through the quagmire of default settings
in cldfrc_init in order to modify the source code.
. Modify build-namelist to allow the definition, defaults, and use case
files to be located in the user source directories (specified via the
-usr_src argument to configure).
. Some workarounds for the intel compiler.
Bugs fixed (include bugzilla ID):
. Fix problems with column output, including the longstanding restriction
that column output doesn't work for fields with names that contain the
underscore character.
. Remove the setting of NF90_NOCLOBBER in calls to nf90_open (this was a
one line change in cam_history.F90). NF90_NOCLOBBER is not a valid
argument to nf90_open, but most netcdf libraries were silently ignoring
this. We encountered an error when trying to link with netcdf4 libs that
were built with hdf5 enabled.
. Fix bug in build-namelist that resulted in the default setting for forganic
having the inputdata root directory prepended twice. This only occurred
when the inputdata root directory was specified as a relative path.
Describe any changes made to build system: none
Describe any changes made to the namelist:
. add outnc_large_files to clm_inparm group
. remove atm_cdf64, ice_cdf64, lnd_cdf64, ocn_cdf64 -- not used
. add cldfrc_rhminl, cldfrc_rhminh for setting cloud fraction tuning
parameters rhminl and rhminh.
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
SVN_EXTERNAL_DIRECTORIES
. cice4_0_20090805 --> cice4_0_20090831
. clm3_6_47 --> clm3_6_51
. drvseq3_0_24 --> drvseq3_0_29
. share3_090706b --> share3_090825
. scripts4_090722b --> scripts4_090901a
models/atm/cam/bld/Makefile.in
. Intel compiler options: move -132 out of FFLAGS and into FIXEDFLAGS.
This allows the compiler to process lines longer than 132 chars in free
format source code.
. Add -static-intel to the intel LDFLAGS to get around problem with CGD
cluster dublin. Should be temporary.
models/atm/cam/bld/build-namelist
. Fix the default setting for forganic (the inputdata root directory was
being prepended twice).
. add cldfrc_rhminl and cldfrc_rhminh by default.
. Modify to allow the definition, defaults, and use case files to be
located in the user source directories (specified via the -usr_src
argument to configure).
models/atm/cam/bld/configure
. fixed a comment
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. add defaults for cldfrc_rhminl, cldfrc_rhminh
models/atm/cam/bld/namelist_files/namelist_definition.xml
. add outnc_large_files to clm_inparm group
. fix documentation for use_64bit_nc
. remove atm_cdf64, ice_cdf64, lnd_cdf64, ocn_cdf64 -- not used
. add cldfrc_rhminl, cldfrc_rhminh
models/atm/cam/src/chemistry/mozart/mo_usrrxt.F90
. Replace parameter index reference by a variable index (workaround for
intel compiler).
models/atm/cam/src/control/cam_history.F90
. Fix problems with column output, including the longstanding restriction
that column output doesn't work for fields with names that contain the
underscore character.
. Remove pio_noclobber flag from the call to cam_pio_openfile.
models/atm/cam/src/control/gauaw_mod.F90
. replace a sqrt generic function call by ()**0.5_r16. Don't know why this
wasn't working since ifort supports quad precision.
models/atm/cam/src/dynamics/fv/inidat.F90
models/atm/cam/src/dynamics/fv/metdata.F90
. Move pio use association out of subroutines and up to module level
(workaround for intel compiler).
models/atm/cam/src/physics/cam/cloud_fraction.F90
. set tuning parameters rhminl, rhminh via namelist
- remove inline code in cldfrc_init for setting rhminl, rhminh
models/atm/cam/src/physics/cam/phys_buffer.F90
. remove old code for binary restart files
. add ability to break large fields into smaller pieces for writing to
restart. If the global field is larger than 4-GB, then try to split it
up using the last pbuf dimension. This avoids a size limit imposed by
the netcdf 64-bit offset format.
models/atm/cam/src/physics/cam/convect_shallow.F90
models/atm/cam/src/physics/cam/phys_debug.F90
models/atm/cam/src/physics/cam/physics_types.F90
models/atm/cam/src/physics/cam/vertical_diffusion.F90
. replace hardcoded constituent indices for cldliq & cldice
models/atm/cam/test/system/test_driver.sh
. changes for intel compiler and for dublin cluster
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except
004 bl132 TBL.sh e32dh ghgrmp 9s ..................................FAIL! rc= 7
011 bl151 TBL.sh e64h outfrq24h 2d ................................FAIL! rc= 7
015 bl331 TBL.sh f4dh co2rmp 9s ...................................FAIL! rc= 7
019 bl332 TBL.sh f4dh ghgrmp 9s ...................................FAIL! rc= 7
030 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7
033 bl337 TBL.sh f4dh fv2d_8tsk 9s ................................FAIL! rc= 7
037 bl371 TBL.sh f1.9dm fvvp_lb2 9s ...............................FAIL! rc= 7
041 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7
044 bl374 TBL.sh fm1.9dh outfrq3s 9s ..............................FAIL! rc= 7
048 bl387 TBL.sh f1.9c351dh outfrq3s+1850-2005_track1 9s ..........FAIL! rc= 7
056 bl992 TBL_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 7
calgary/lf95: All PASS except
004 bl111 TBL.sh e8t5mdm ttrac 9s .................................FAIL! rc= 7
008 bl112 TBL.sh e8dm ghgrmp 9s ...................................FAIL! rc= 7
014 bl114 TBL.sh e8dm co2rmp 9s ...................................FAIL! rc= 7
021 bl311 TBL.sh f10t5mdm ttrac 9s ................................FAIL! rc= 7
029 bl312 TBL.sh f10dm ghgrmp 9s ..................................FAIL! rc= 7
034 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 7
037 bl315 TBL.sh f10c3dm outfrq3s 9s ..............................FAIL! rc= 7
040 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7
dublin/pgi: All PASS except
004 bl111 TBL.sh e8t5mdm ttrac 9s .................................FAIL! rc= 7
008 bl112 TBL.sh e8dm ghgrmp 9s ...................................FAIL! rc= 7
014 bl114 TBL.sh e8dm co2rmp 9s ...................................FAIL! rc= 7
021 bl311 TBL.sh f10t5mdm ttrac 9s ................................FAIL! rc= 7
029 bl312 TBL.sh f10dm ghgrmp 9s ..................................FAIL! rc= 7
034 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 7
037 bl315 TBL.sh f10c3dm outfrq3s 9s ..............................FAIL! rc= 7
040 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7
The failed baseline tests are due to updating the CLM and CICE components.
The baseline tests in aqua-planet, ideal, and adiabatic modes all pass.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: new climate due to changes in CLM and
CICE components.
===============================================================
===============================================================
Tag name: cam3_6_57
Originator(s): Dani Bundy Coleman
Date: 08/13/09
One-line Summary: Add optional ESMFv3 interfaces in order to use the Gregorian calendar
Purpose of changes:
To use the Gregorian calendar, CAM must link with the actual ESMF library
instead of the default WRF fortran implementation of it.
The configure option to do this remains unchanged:
-esmf_libdir directory
There is a new configure option:
-esmf_libvers [ 2 (default) | 3 ]
that is only necessary if configure is unable to determine the library
version through an automatic test. The default remains esmf_libvers=2 for
compatibility with the WRF implementation of the ESMF library.
This code includes the changes necessary for CAM to run with a Gregorian
calendar (existing namelist variable: calendar = 'GREGORIAN' ) but this
option will fail until the externals (drv,lnd,csm_share) have been updated.
Bugs fixed (include bugzilla ID): None
Describe any changes made to build system:
Added configure option -esmf_libvers [ 2 (default) | 3 ]
Describe any changes made to the namelist: None
List any changes to the defaults for the boundary data set: None
Describe any substantial timing or memory changes: None
Code reviewed by: Eaton
List all subroutines eliminated: None
List all subroutines added and what they do: None
List all existing files that have been modified, and describe the changes:
M models/ocn/dom/ocn_time_manager.F90
M models/ice/csim4/ice_time_manager.F90
Change ESMF variables from intent(in) to intent(inout) in order to use
ESMF version 3. When using a Gregorian calendar, have time managers
fake day 366 by repeating day 365. This allows the models to run on
leap-years without having to modify the solar zenith angle calculations.
M models/atm/cam/src/utils/time_manager.F90
In addition to above, trap for Feb 29 of a non-leap year, which
might be invoked legitimately by a routine reading a dataset for a fixed
year (see below).
M models/atm/cam/bld/configure
Require test of ESMF library when invoked with the existing option
-esmf_libdir in order to test which ESMF version is being used. Required
some re-ordering of flow. Add option -esmf_libvers [ 2 (default) | 3 ] in
case the test fails.
M models/atm/cam/bld/Makefile.in
If using the ESMF library, override default linker with ESMF recommendation
M models/atm/cam/bld/config_files/definition.xml
Add esmf_libvers option
M models/atm/cam/src/chemistry/mozart/mo_flbc.F90
Trap for Feb 29 of a non-leap year, which might be invoked legitimately by
a routine reading a dataset for a fixed year (see below).
Explanation of the Feb 29 trap:
When using a fixed year from a dataset, these routines create an integer
date by combining the fixed year with the current model day-of-year.
This fails on Feb 29 of a leap year:
e.g. fixed year = 1990
model date = 20000229
combined date = 19900229
Since this is an unrealistic date, we trap for it here, using Feb 28
instead of Feb 29.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: None
calgary/lf95: None
calgary/pgi or jaguar/pgi: None
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
Default model is bfb
===============================================================
Tag name: cam3_6_56
Originator(s): Art Mirin
Date: 08/10/2009
One-line Summary: Alternative OpenMP in FV dynamics
Purpose of changes: In FV dynamics, OpenMP is applied in the vertical direction. For example, for
cases with one level per subdomain, OpenMP becomes meaningless. An option has been added to instead
apply OpenMP to horizontal loops throughout sw_core and tp_core. Because this involves whether or
not to include directives, the option is necessarily a compile-time option. In order to apply this
option, one must define the CPP flag INNER_OMP. One should not do this in general, since
for cases where there are enough levels per subdomain to apply OpenMP, there is less OpenMP
overhead applying it vertically. If the user does not define INNER_OMP, the precompiled code
will be identical to the previous tag.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Worley
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
In dynamics/fv, files cd_core.F90 and trac2d.F90 optionally have OpenMP directives removed;
files sw_core.F90 and tp_core.F90 optionally have OpenMP directives added.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
calgary/lf95:
calgary/pgi or jaguar/pgi:
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_6_55
Originator(s): mvr,eaton,hannay
Date: Thu Aug 6 2009
One-line Summary: Completed update of externals to be consistent with ccsm;
modified machine support in test scripts; new aerosol/ozone datasets; track1 tuning
Purpose of changes:
- Updated externals as follows:
clm3_6_46 -> clm3_6_47
cice4_0_20090721 -> cice4_0_20090805
clm needed update to fix bug in hybrid dynamic pft code; cice needed update to fix
bug where lahey detected use of variable with undefined value
- new aerosol and ozone datasets arrived for the 1850-2005 transient runs and for
the 2000 climatology - gives new climate
- linux cluster bangkok was retired and a new cluster came online (dublin)...
the test scripts were modified to reflect this
- tuning parameter rhminl was tweaked for track1:
for 0.9x1.25: .91_r8 to .920_r8
for 1.9x2.5: .92_r8 to .924_r8
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: new aerosol and ozone datasets
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
D models/atm/cam/test/system/tests_pretag_bangkok
- file containing set of required tests removed for retired machine
List all subroutines added and what they do:
A + models/atm/cam/test/system/tests_pretag_calgary
- new file for set of required tests on machine calgary
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/test_driver.sh
M models/atm/cam/test/system/CAM_runcmnd.sh
- test scripts modifed for machine changes (bangkok out, dublin in)
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
M models/atm/cam/bld/namelist_files/use_cases/1850-2005_track1.xml
- mods to reflect new aerosol and ozone datasets
M models/atm/cam/src/physics/cam/cloud_fraction.F90
- new setting for tuning parameter rhminl
M SVN_EXTERNAL_DIRECTORIES
M .
- updated externals as summarized above.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
004 bl132 TBL.sh e32dh ghgrmp 9s ..................................FAIL! rc= 7
006 bl133 TBL.sh e32pdh aqpgro 3s .................................FAIL! rc= 7
011 bl151 TBL.sh e64h outfrq24h 2d ................................FAIL! rc= 7
015 bl331 TBL.sh f4dh co2rmp 9s ...................................FAIL! rc= 7
019 bl332 TBL.sh f4dh ghgrmp 9s ...................................FAIL! rc= 7
021 bl333 TBL.sh f4c351pdh aqpgro 3s ..............................FAIL! rc= 7
030 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7
033 bl337 TBL.sh f4dh fv2d_8tsk 9s ................................FAIL! rc= 7
037 bl371 TBL.sh f1.9dm fvvp_lb2 9s ...............................FAIL! rc= 7
041 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7
048 bl387 TBL.sh f1.9c351dh outfrq3s+1850-2005_track1 9s ..........FAIL! rc= 7
056 bl992 TBL_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 7
- baselines (except adiabatic and ideal physics) failed due to new aerosol
and ozone datasets; bl387 would've failed anyway due to bug fix
dublin/lf95:
004 bl111 TBL.sh e8t5mdm ttrac 9s .................................FAIL! rc= 5
008 bl112 TBL.sh e8dm ghgrmp 9s ...................................FAIL! rc= 5
010 bl113 TBL.sh e8pdm aqpgro 3s ..................................FAIL! rc= 5
014 bl114 TBL.sh e8dm co2rmp 9s ...................................FAIL! rc= 5
017 bl115 TBL.sh e8idm idphys 9s ..................................FAIL! rc= 5
021 bl311 TBL.sh f10t5mdm ttrac 9s ................................FAIL! rc= 5
029 bl312 TBL.sh f10dm ghgrmp 9s ..................................FAIL! rc= 5
031 bl313 TBL.sh f10pdm aqpgro 3s .................................FAIL! rc= 5
034 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 5
037 bl315 TBL.sh f10c3dm outfrq3s 9s ..............................FAIL! rc= 5
040 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 5
- all baselines failed because baseline was not set up to run on dublin;
calgary/pgi: All PASS except:
004 bl111 TBL.sh e8t5mdm ttrac 9s .................................FAIL! rc= 7
008 bl112 TBL.sh e8dm ghgrmp 9s ...................................FAIL! rc= 7
010 bl113 TBL.sh e8pdm aqpgro 3s ..................................FAIL! rc= 7
014 bl114 TBL.sh e8dm co2rmp 9s ...................................FAIL! rc= 7
021 bl311 TBL.sh f10t5mdm ttrac 9s ................................FAIL! rc= 7
029 bl312 TBL.sh f10dm ghgrmp 9s ..................................FAIL! rc= 7
031 bl313 TBL.sh f10pdm aqpgro 3s .................................FAIL! rc= 7
034 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 7
040 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7
- all baselines (except ideal physics) failed due to new aerosol and ozone
datasets
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: New climate due to new aerosol and ozone datasets
===============================================================
===============================================================
Tag name: cam3_6_54
Originator(s): eaton
Date: Mon Aug 3 11:23:46 MDT 2009
One-line Summary: Update externals to be consistent with ccsm4_0_beta20.
Purpose of changes:
. Updated externals as follows:
clm3_6_19 -> clm3_6_46
cice4_0_20090226 -> cice4_0_20090721
scripts4_090428 -> scripts4_090722b
docn7_090107 -> docn7_090720
**NOTE** clm3_6_46 has a known bug in the hybrid dynamic pft code.
Consequently this tag should not be used in hybrid mode for transient
simulations (climatology runs are OK). The fix will be added to the
next CAM commit.
. Mods so that SCAM works with the updated CLM and with trop_mam3.
. Added WACCM use case for 1850 track1 simulations.
. Added use case for 1850 track1 w/ 2X CO2 simulations.
. Updated regression tests to use the CCSM version of cprnc.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do:
models/atm/cam/bld/namelist_files/use_cases/1850_track1_2xco2.xml
. same as use case 1850_track1 except double the CO2 value
models/atm/cam/bld/namelist_files/use_cases/waccm_1850_track1.xml
. use case for waccm 1850 track1 runs
List all existing files that have been modified, and describe the changes:
SVN_EXTERNAL_DIRECTORIES
. updated externals as summarized above.
models/atm/cam/bld/build-namelist
. add code to use sim_year in generating the clm namelist
. update code for clm namelist to be consistent with clm3_6_46
. add -DCO2A to the cppdefs. The clm namelist is now being set with the
variable co2_type='diagnostic' so it always listens to the atm to recieve
the surface CO2 concentration. -DCO2A is needed when building the
coupler to make sure these fields are always allocated.
. added capability to not include the aerosol deposition dataset in the CLM
namelist when CAM will provide the deposition fluxes to the coupler.
Note: due to CLM bug the aerosol deposition dataset currently needs to be
added to the CLM namelist whether it's used or not.
models/atm/cam/bld/configure
. define -DCO2A for building the cpl7 code. This is currently the only BGC
option implemented.
. change -DMAXPATCH_PFT=4 to -DMAXPATCH_PFT=numpft+1 for CLM build.
. add new CLM directory for mct or esmf component interface.
. add new CICE directories for mct or esmf component interface.
. hack to modify output of CICE configure in SCAM mode since the CICE
configure doesn't support SCAM mode.
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. update clm defaults for consistency with clm3_6_43
models/atm/cam/bld/namelist_files/namelist_definition.xml
. add sim_year
. remove old clm variables: csm_doflxave, irad
. add new clm variables: rtm_nsteps, urban_hac, urban_traffic
models/atm/cam/bld/namelist_files/use_cases/1850-2005_track1.xml
models/atm/cam/bld/namelist_files/use_cases/1850_track1.xml
models/atm/cam/bld/namelist_files/use_cases/1870-2000_control.xml
models/atm/cam/bld/namelist_files/use_cases/1870_bam.xml
models/atm/cam/bld/namelist_files/use_cases/1870_control.xml
models/atm/cam/bld/namelist_files/use_cases/1870_prog_aero.xml
models/atm/cam/bld/namelist_files/use_cases/waccm_1850_cntrl.xml
. add setting for sim_year
models/atm/cam/src/control/cam_restart.F90
. move 'use pio' for intel compiler workaround
models/atm/cam/bld/build-namelist
models/atm/cam/src/chemistry/mozart/mo_drydep.F90
models/atm/cam/src/chemistry/utils/tracer_data.F90
models/atm/cam/src/control/history_defaults.F90
models/atm/cam/src/control/scamMod.F90
models/atm/cam/src/dynamics/eul/diag_dynvar_ic.F90
models/atm/cam/src/dynamics/eul/inidat.F90
models/atm/cam/src/dynamics/eul/iop.F90
models/atm/cam/src/dynamics/eul/tfilt_massfix.F90
models/atm/cam/src/physics/cam/initindx.F90
models/atm/cam/test/system/nl_files/scm_prep
models/atm/cam/test/system/nl_files/scm_prep_35
. mods for SCAM
models/atm/cam/test/system/CAM_compare.sh
. update to parse output from cprnc version:
https://svn-ccsm-models.cgd.ucar.edu/tools/cprnc/trunk_tags/cprnc_081022
models/atm/cam/test/system/CAM_runcmnd.sh
. modify run commands for bluefire to use the environment variable
TARGET_CPU_LIST and add absolute paths to "launch" and "hybrid_launch"
models/atm/cam/test/system/test_driver.sh
. update the location of the cprnc executable
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
004 bl132 TBL.sh e32dh ghgrmp 9s ..................................FAIL! rc= 7
011 bl151 TBL.sh e64h outfrq24h 2d ................................FAIL! rc= 7
015 bl331 TBL.sh f4dh co2rmp 9s ...................................FAIL! rc= 7
019 bl332 TBL.sh f4dh ghgrmp 9s ...................................FAIL! rc= 7
030 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7
033 bl337 TBL.sh f4dh fv2d_8tsk 9s ................................FAIL! rc= 7
037 bl371 TBL.sh f1.9dm fvvp_lb2 9s ...............................FAIL! rc= 7
041 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7
044 bl374 TBL.sh fm1.9dh outfrq3s 9s ..............................FAIL! rc= 7
046 er387 TER.sh f1.9c351dh outfrq3s+1850-2005_track1 4+5s ........FAIL! rc= 10
047 br387 TBR.sh f1.9c351dh outfrq3s+1850-2005_track1 6+3s ........FAIL! rc= 12
048 bl387 TBL.sh f1.9c351dh outfrq3s+1850-2005_track1 9s ..........FAIL! rc= 7
056 bl992 TBL_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 5
The baseline failures are all expected due to the new climate of the CLM
tag.
er387, br387, bl387 are all failing due to the bug in clm3_6_46.
calgary/lf95: not done
There are currently problems with the mpich installation under lf95 on
calgary (since the OS upgrade).
calgary/pgi: All PASS except:
004 bl111 TBL.sh e8t5mdm ttrac 9s .................................FAIL! rc= 7
008 bl112 TBL.sh e8dm ghgrmp 9s ...................................FAIL! rc= 7
014 bl114 TBL.sh e8dm co2rmp 9s ...................................FAIL! rc= 7
021 bl311 TBL.sh f10t5mdm ttrac 9s ................................FAIL! rc= 7
029 bl312 TBL.sh f10dm ghgrmp 9s ..................................FAIL! rc= 7
034 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 7
037 bl315 TBL.sh f10c3dm outfrq3s 9s ..............................FAIL! rc= 7
040 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7
These baseline failures are expected due to the new climate of CLM.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: New climate due to updated CLM external.
===============================================================
===============================================================
Tag name: cam3_6_53
Originator(s): Jim Edwards, Mark Taylor, Pat Worley, Jerry Olson
Date: 7-23-09
One-line Summary: Update of homme dynamics,
correction of column output,
improvement of load balance initialization
Purpose of changes:
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: Updated build paths for calgary
Describe any changes made to the namelist: Updated defaults for homme dycore,
reimplimented use_64bit_nc namelist flag
(this flag was disabled in cam3_6_39)
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: originators
List all subroutines eliminated:
- subroutine xzy_to_xyz (xyzbuf, xzybuf, dimind)
- subroutine loc_xzy_to_xyz (xyzbuf, xzybuf, dimind)
- subroutine loc_xyz_to_xzy (xzybuf, xyzbuf, dimind)
dead code in cam_history.F90
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
updated build paths for calgary.
models/atm/cam/test/system/CAM_compare.sh
models/atm/cam/test/system/test_driver.sh
models/atm/cam/test/system/CAM_runcmnd.sh
modified build defaults for homme dycore
models/atm/cam/bld/config_files/horiz_grid.xml
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
Column output is unsupported for unstructured grids, added error checking on
fincllonlat
models/atm/cam/src/control/runtime_opts.F90
Changed default value of use_64bit_nc to true
models/atm/cam/src/control/cam_control_mod.F90
Added error checking for the case of a filename which exceeds the variable length
models/atm/cam/src/control/ioFileMod.F90
added BGP to supported platforms
models/atm/cam/src/control/print_memusage.F90
Corrected location of column data output, new locations were verified by
comparing netcdf column output to global output file using chkcol.ncl
models/atm/cam/src/control/cam_history.F90
models/atm/cam/src/utils/pio_utils.F90
Added code so that if the proc_name count exceeds 256 only the first 256 are printed to the log
models/atm/cam/src/utils/spmd_utils.F90
improved initialization of load balancing scheme, especially for high task counts and
unstructed dycores.
models/atm/cam/src/physics/cam/phys_grid.F90
models/atm/cam/src/dynamics/homme/dyn_grid.F90
development update of homme dycore.
models/atm/cam/src/dynamics/homme/external/prim_advance_mod.F90
models/atm/cam/src/dynamics/homme/external/reduction_mod.F90
models/atm/cam/src/dynamics/homme/external/prim_state_mod.F90
models/atm/cam/src/dynamics/homme/external/cube_mod.F90
models/atm/cam/src/dynamics/homme/external/prim_driver_mod.F90
models/atm/cam/src/dynamics/homme/external/element_mod.F90
models/atm/cam/src/dynamics/homme/external/viscosity_mod.F90
models/atm/cam/src/dynamics/homme/external/prim_advection_mod.F90
models/atm/cam/src/dynamics/homme/external/edge_mod.F90
models/atm/cam/src/dynamics/homme/external/namelist_mod.F90
models/atm/cam/src/dynamics/homme/stepon.F90
models/atm/cam/src/dynamics/homme/inidat.F90
models/atm/cam/src/dynamics/homme/dyn_comp.F90
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
051 bl731 TBL.sh h16x4dm aqua 9s ..................................FAIL! rc= 7
Fails due to update of homme dycore
052 sm991 TSM_ccsm.sh f19_f19 F 2d ................................FAIL! rc= 4
053 er991 TER_ccsm.sh f19_f19 F 1+1d ..............................FAIL! rc= 6
054 sm992 TSM_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 4
055 er992 TER_ccsm.sh f19_g15 E 1+1d ..............................FAIL! rc= 6
056 bl992 TBL_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 4
Preexisting failures
calgary/lf95:
all pass
calgary/pgi or jaguar/pgi:
all pass
CAM tag used for the baseline comparison tests if different than previous
tag: cam3_6_51 (no relevent changes in cam3_6_52)
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_6_52
Originator(s): eaton
Date: Thu Jul 23 11:19:58 MDT 2009
One-line Summary: fix filepath for builds from ccsm scripts
Purpose of changes: fix filepath for builds from ccsm scripts
Bugs fixed (include bugzilla ID):
. CAM's configure was not producing the correct filepath when run from the
CCSM scripts (with -ccsm_seq flag).
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/configure
. fix filepath for ccsm mode
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: not done
calgary/lf95: not done
calgary/pgi or jaguar/pgi: not done
No regression testing was done for this tag. The change only affects the
ccsm tests and they are currently failing for other reasons.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam3_6_51
Originator(s): jedwards, eaton
Date: Tue Jul 21 14:04:05 MDT 2009
One-line Summary: fixes for history buffer restart files
Purpose of changes:
. Finish fixing the bug in the history buffer restart files. The first
part of this fix was made in cam3_6_49.
. Update externals:
drvseq3_0_14 -> drvseq3_0_24
pio50_prod -> pio51_prod
share3_090112 -> share3_090706b
Bugs fixed (include bugzilla ID): see above
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes:
. reduce memory requirement for unstructured grids by writing the global
lat, lon, and area fields one at a time.
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
SVN_EXTERNAL_DIRECTORIES
. update driver, csm_share, and pio
models/atm/cam/src/control/cam_history.F90
. fix for history buffer restart files
. reduce memory requirement for unstructured grids by writing the global
lat, lon, and area fields one at a time.
models/atm/cam/src/utils/pio_utils.F90
. removed some commented code
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
050 er731 TER.sh h16x4dm aqua 4+5s ................................FAIL! rc= 10
051 bl731 TBL.sh h16x4dm aqua 9s ..................................FAIL! rc= 7
052 sm991 TSM_ccsm.sh f19_f19 F 2d ................................FAIL! rc= 4
053 er991 TER_ccsm.sh f19_f19 F 1+1d ..............................FAIL! rc= 6
054 sm992 TSM_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 4
055 er992 TER_ccsm.sh f19_g15 E 1+1d ..............................FAIL! rc= 6
056 bl992 TBL_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 4
The ccsm test failures are pre-existing.
Jim Edwards will address the homme test failures in his upcoming commit.
calgary/lf95: All PASS
calgary/pgi: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam3_6_50
Originator(s): Francis Vitt
Date: 14 July 2009
One-line Summary:
Cleanup the old CAM3 prognostic aerosol code files and use
the chemistry preprocessor only when customizing the chemistry.
Purpose of changes:
The old CAM3 prognostic aerosol routines have been replicated
in the MOZART chemistry framework. The deprecated configure and
run-time options and source code have been removed to improve
maintainability.
The configure script only invokes the chemistry preprocessor only
when the chemistry is customized which is done by providing the code files
generated by the preprocessor for the pre-defined chemistry packages.
The purpose of this is to avoid portability and long path name issues
associated with the chemistry preprocessor and shorten the configure time.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Removed deprecated prog_aero configure option
Describe any changes made to the namelist:
Removed the namelist variables:
aero_carbon
aero_dust
aero_progsslt
aero_sea_salt
bndtvdms
bndtvoxid
bndtvsox
caer_emis
rampyear_prognostic_sulfur
scenario_prognostic_sulfur
Moved namelist variable soil_erod form cam_inparm to chem_inparm
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Brian Eaton
List all subroutines eliminated:
D models/atm/cam/src/physics/cam/progseasalts_intr.F90
D models/atm/cam/src/physics/cam/drydep_mod.F90
D models/atm/cam/src/physics/cam/dust_sediment_mod.F90
D models/atm/cam/src/physics/cam/wetdep.F90
D models/atm/cam/src/physics/cam/dust_intr.F90
D models/atm/cam/src/physics/cam/aerosol_intr.F90
D models/atm/cam/src/physics/cam/sulchem.F90
These have moved to models/atm/cam/src/chemistry/bulk_aero
D models/atm/cam/src/physics/cam/acbnd.F90
D models/atm/cam/src/physics/cam/dmsbnd.F90
D models/atm/cam/src/physics/cam/caer.F90
D models/atm/cam/src/physics/cam/sulfur_intr.F90
D models/atm/cam/src/physics/cam/sulemis.F90
D models/atm/cam/src/physics/cam/sulbnd.F90
D models/atm/cam/src/physics/cam/seasalt_intr.F90
D models/atm/cam/src/physics/cam/soxbnd.F90
D models/atm/cam/src/physics/cam/caerbnd.F90
D models/atm/cam/src/physics/cam/carbon_intr.F90
These deprecated module files have been removed.
D models/atm/cam/src/chemistry/trop_mam3
D models/atm/cam/src/chemistry/trop_bam
These directories have been replaced by pp_trop_mam3 and pp_trop_bam.
List all subroutines added and what they do:
A models/atm/cam/src/chemistry/bulk_aero
A models/atm/cam/src/chemistry/bulk_aero/progseasalts_intr.F90
A models/atm/cam/src/chemistry/bulk_aero/drydep_mod.F90
A models/atm/cam/src/chemistry/bulk_aero/dust_sediment_mod.F90
A models/atm/cam/src/chemistry/bulk_aero/wetdep.F90
A models/atm/cam/src/chemistry/bulk_aero/dust_intr.F90
A models/atm/cam/src/chemistry/bulk_aero/aerosol_intr.F90
A models/atm/cam/src/chemistry/bulk_aero/sulchem.F90
These are the CAM3 aerosol module files that are used within the MOZART
chemistry framework.
A models/atm/cam/src/chemistry/pp_trop_mozart
A models/atm/cam/src/chemistry/pp_trop_mozart/mo_lu_solve.F90
A models/atm/cam/src/chemistry/pp_trop_mozart/mo_setrxt.F90
A models/atm/cam/src/chemistry/pp_trop_mozart/m_rxt_id.F90
A models/atm/cam/src/chemistry/pp_trop_mozart/mo_phtadj.F90
A models/atm/cam/src/chemistry/pp_trop_mozart/mo_nln_matrix.F90
A models/atm/cam/src/chemistry/pp_trop_mozart/mo_adjrxt.F90
A models/atm/cam/src/chemistry/pp_trop_mozart/m_het_id.F90
A models/atm/cam/src/chemistry/pp_trop_mozart/mo_lu_factor.F90
A models/atm/cam/src/chemistry/pp_trop_mozart/mo_indprd.F90
A models/atm/cam/src/chemistry/pp_trop_mozart/mo_sim_dat.F90
A models/atm/cam/src/chemistry/pp_trop_mozart/m_spc_id.F90
A models/atm/cam/src/chemistry/pp_trop_mozart/mo_imp_sol.F90
A models/atm/cam/src/chemistry/pp_trop_mozart/chem_mods.F90
A models/atm/cam/src/chemistry/pp_trop_mozart/mo_prod_loss.F90
A models/atm/cam/src/chemistry/pp_trop_mozart/mo_lin_matrix.F90
These are the preprocessor generated code files for the trop_mozart
chemistry package.
A models/atm/cam/src/chemistry/pp_trop_bam
A models/atm/cam/src/chemistry/pp_trop_bam/mo_lu_solve.F90
A models/atm/cam/src/chemistry/pp_trop_bam/mo_setrxt.F90
A models/atm/cam/src/chemistry/pp_trop_bam/m_rxt_id.F90
A models/atm/cam/src/chemistry/pp_trop_bam/mo_phtadj.F90
A models/atm/cam/src/chemistry/pp_trop_bam/mo_nln_matrix.F90
A models/atm/cam/src/chemistry/pp_trop_bam/mo_adjrxt.F90
A models/atm/cam/src/chemistry/pp_trop_bam/m_het_id.F90
A models/atm/cam/src/chemistry/pp_trop_bam/mo_lu_factor.F90
A models/atm/cam/src/chemistry/pp_trop_bam/mo_indprd.F90
A models/atm/cam/src/chemistry/pp_trop_bam/mo_sim_dat.F90
A models/atm/cam/src/chemistry/pp_trop_bam/m_spc_id.F90
A models/atm/cam/src/chemistry/pp_trop_bam/mo_imp_sol.F90
A models/atm/cam/src/chemistry/pp_trop_bam/chem_mods.F90
A models/atm/cam/src/chemistry/pp_trop_bam/mo_prod_loss.F90
A models/atm/cam/src/chemistry/pp_trop_bam/mo_lin_matrix.F90
These are the preprocessor generated code files for the trop_bam
chemistry package.
A models/atm/cam/src/chemistry/pp_trop_mam3
A models/atm/cam/src/chemistry/pp_trop_mam3/mo_lu_solve.F90
A models/atm/cam/src/chemistry/pp_trop_mam3/mo_setrxt.F90
A models/atm/cam/src/chemistry/pp_trop_mam3/m_rxt_id.F90
A models/atm/cam/src/chemistry/pp_trop_mam3/mo_phtadj.F90
A models/atm/cam/src/chemistry/pp_trop_mam3/mo_nln_matrix.F90
A models/atm/cam/src/chemistry/pp_trop_mam3/mo_adjrxt.F90
A models/atm/cam/src/chemistry/pp_trop_mam3/m_het_id.F90
A models/atm/cam/src/chemistry/pp_trop_mam3/mo_lu_factor.F90
A models/atm/cam/src/chemistry/pp_trop_mam3/mo_indprd.F90
A models/atm/cam/src/chemistry/pp_trop_mam3/mo_sim_dat.F90
A models/atm/cam/src/chemistry/pp_trop_mam3/m_spc_id.F90
A models/atm/cam/src/chemistry/pp_trop_mam3/mo_imp_sol.F90
A models/atm/cam/src/chemistry/pp_trop_mam3/chem_mods.F90
A models/atm/cam/src/chemistry/pp_trop_mam3/mo_prod_loss.F90
A models/atm/cam/src/chemistry/pp_trop_mam3/mo_lin_matrix.F90
These are the preprocessor generated code files for the trop_mam3
chemistry package.
A models/atm/cam/src/chemistry/pp_trop_mam7
A models/atm/cam/src/chemistry/pp_trop_mam7/mo_lu_solve.F90
A models/atm/cam/src/chemistry/pp_trop_mam7/mo_setrxt.F90
A models/atm/cam/src/chemistry/pp_trop_mam7/m_rxt_id.F90
A models/atm/cam/src/chemistry/pp_trop_mam7/mo_phtadj.F90
A models/atm/cam/src/chemistry/pp_trop_mam7/mo_nln_matrix.F90
A models/atm/cam/src/chemistry/pp_trop_mam7/mo_adjrxt.F90
A models/atm/cam/src/chemistry/pp_trop_mam7/m_het_id.F90
A models/atm/cam/src/chemistry/pp_trop_mam7/mo_lu_factor.F90
A models/atm/cam/src/chemistry/pp_trop_mam7/mo_indprd.F90
A models/atm/cam/src/chemistry/pp_trop_mam7/mo_sim_dat.F90
A models/atm/cam/src/chemistry/pp_trop_mam7/m_spc_id.F90
A models/atm/cam/src/chemistry/pp_trop_mam7/mo_imp_sol.F90
A models/atm/cam/src/chemistry/pp_trop_mam7/chem_mods.F90
A models/atm/cam/src/chemistry/pp_trop_mam7/mo_prod_loss.F90
A models/atm/cam/src/chemistry/pp_trop_mam7/mo_lin_matrix.F90
These are the preprocessor generated code files for the trop_mam7
chemistry package.
A models/atm/cam/src/chemistry/pp_trop_ghg
A models/atm/cam/src/chemistry/pp_trop_ghg/mo_lu_solve.F90
A models/atm/cam/src/chemistry/pp_trop_ghg/mo_setrxt.F90
A models/atm/cam/src/chemistry/pp_trop_ghg/m_rxt_id.F90
A models/atm/cam/src/chemistry/pp_trop_ghg/mo_phtadj.F90
A models/atm/cam/src/chemistry/pp_trop_ghg/mo_nln_matrix.F90
A models/atm/cam/src/chemistry/pp_trop_ghg/mo_adjrxt.F90
A models/atm/cam/src/chemistry/pp_trop_ghg/m_het_id.F90
A models/atm/cam/src/chemistry/pp_trop_ghg/mo_lu_factor.F90
A models/atm/cam/src/chemistry/pp_trop_ghg/mo_indprd.F90
A models/atm/cam/src/chemistry/pp_trop_ghg/mo_sim_dat.F90
A models/atm/cam/src/chemistry/pp_trop_ghg/m_spc_id.F90
A models/atm/cam/src/chemistry/pp_trop_ghg/mo_imp_sol.F90
A models/atm/cam/src/chemistry/pp_trop_ghg/chem_mods.F90
A models/atm/cam/src/chemistry/pp_trop_ghg/mo_prod_loss.F90
A models/atm/cam/src/chemistry/pp_trop_ghg/mo_lin_matrix.F90
These are the preprocessor generated code files for the trop_ghg
chemistry package.
A models/atm/cam/src/chemistry/pp_super_fast_llnl
A models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_lu_solve.F90
A models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_setrxt.F90
A models/atm/cam/src/chemistry/pp_super_fast_llnl/m_rxt_id.F90
A models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_phtadj.F90
A models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_nln_matrix.F90
A models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_adjrxt.F90
A models/atm/cam/src/chemistry/pp_super_fast_llnl/m_het_id.F90
A models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_lu_factor.F90
A models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_indprd.F90
A models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_sim_dat.F90
A models/atm/cam/src/chemistry/pp_super_fast_llnl/m_spc_id.F90
A models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_imp_sol.F90
A models/atm/cam/src/chemistry/pp_super_fast_llnl/chem_mods.F90
A models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_prod_loss.F90
A models/atm/cam/src/chemistry/pp_super_fast_llnl/mo_lin_matrix.F90
These are the preprocessor generated code files for the super_fast_llnl
chemistry package.
A models/atm/cam/src/chemistry/pp_waccm_mozart
A models/atm/cam/src/chemistry/pp_waccm_mozart/mo_lu_solve.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart/mo_setrxt.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart/m_rxt_id.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart/mo_phtadj.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart/mo_nln_matrix.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart/mo_adjrxt.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart/m_het_id.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart/mo_lu_factor.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart/mo_indprd.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart/mo_sim_dat.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart/m_spc_id.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart/mo_imp_sol.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart/chem_mods.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart/mo_prod_loss.F90
A models/atm/cam/src/chemistry/pp_waccm_mozart/mo_lin_matrix.F90
These are the preprocessor generated code files for the waccm_mozart
chemistry package.
A models/atm/cam/src/chemistry/pp_waccm_ghg
A models/atm/cam/src/chemistry/pp_waccm_ghg/mo_lu_solve.F90
A models/atm/cam/src/chemistry/pp_waccm_ghg/mo_setrxt.F90
A models/atm/cam/src/chemistry/pp_waccm_ghg/m_rxt_id.F90
A models/atm/cam/src/chemistry/pp_waccm_ghg/mo_phtadj.F90
A models/atm/cam/src/chemistry/pp_waccm_ghg/mo_nln_matrix.F90
A models/atm/cam/src/chemistry/pp_waccm_ghg/mo_adjrxt.F90
A models/atm/cam/src/chemistry/pp_waccm_ghg/m_het_id.F90
A models/atm/cam/src/chemistry/pp_waccm_ghg/mo_lu_factor.F90
A models/atm/cam/src/chemistry/pp_waccm_ghg/mo_indprd.F90
A models/atm/cam/src/chemistry/pp_waccm_ghg/mo_sim_dat.F90
A models/atm/cam/src/chemistry/pp_waccm_ghg/m_spc_id.F90
A models/atm/cam/src/chemistry/pp_waccm_ghg/mo_imp_sol.F90
A models/atm/cam/src/chemistry/pp_waccm_ghg/chem_mods.F90
A models/atm/cam/src/chemistry/pp_waccm_ghg/mo_prod_loss.F90
A models/atm/cam/src/chemistry/pp_waccm_ghg/mo_lin_matrix.F90
These are the preprocessor generated code files for the waccm_ghg
chemistry package.
List all existing files that have been modified, and describe the changes:
U models/atm/cam/bld/configure
Removed prog_aer option
Invoke chem preprocessor only when chem package is customized
U models/atm/cam/bld/config_files/definition.xml
Removed prog_aer configure option
U models/atm/cam/bld/config_files/defaults_trop_mozart.xml
Made 1.9x2.5 the default resolution for trop_mozart chemistry
U models/atm/cam/bld/namelist_files/namelist_definition.xml
Removed the CAM3 prognostic aerosol namelist variables listed above.
Moved soil_erod namelist varible to chem_inparm namelist
U models/atm/cam/bld/build-namelist
Removed the old prognostic aerosol code
U models/atm/cam/bld/perl5lib/t/config_definition.xml
Removed prog_aero configure option
U models/atm/cam/SVN_EXTERNAL_DIRECTORIES
Modified chemistry preprocessor to handle longer path names
U models/atm/cam/src/control/runtime_opts.F90
Removed the namelist variables listed above
Moved soil_erod namelist varible to chem_inparm namelist
U models/atm/cam/src/control/filenames.F90
Remove the deprecated prognostic aerosol input file names
U models/atm/cam/src/physics/cam/tropopause.F90
Changed tropLev from real(r8) to integer since this is used to index arrays.
U models/atm/cam/src/physics/cam/advnce.F90
Removed the deprecated aerosol_time_interp subroutine call
U models/atm/cam/src/physics/cam/tphysac.F90
Removed the deprecated aerosol_srcsnk_intr subroutine call
U models/atm/cam/src/chemistry/utils/prescribed_volcaero.F90
Changed tropLev from real(r8) to integer (see tropopause.F90).
U models/atm/cam/src/chemistry/modal_aero/modal_aero_deposition.F90
Fixed for trop_mam7 configuration.
U models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
Removed code that checks the deprecated prognostic aerosol namelist settings
U models/atm/cam/src/chemistry/mozart/chemistry.F90
Moved soil_erod namelist varible from cam_inparm to chem_inparm namelist
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: pre-existing failures
052 sm991 TSM_ccsm.sh f19_f19 F 2d ................................FAIL! rc= 4
053 er991 TER_ccsm.sh f19_f19 F 1+1d ..............................FAIL! rc= 6
054 sm992 TSM_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 4
055 er992 TER_ccsm.sh f19_g15 E 1+1d ..............................FAIL! rc= 6
056 bl992 TBL_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 4
calgary/lf95: All Pass
calgary/pgi or jaguar/pgi:
034 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 7
This fails due to the difference in molecular masses in the preprocessor generated
code produced by different compilers.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: bit-for-bit
===============================================================
===============================================================
Tag name: cam3_6_49
Originator(s): jedwards, eaton
Date: Wed Jul 8 12:15:07 MDT 2009
One-line Summary: fixes for history buffer restart files
Purpose of changes:
. Some bugs have been discovered in the netcdf history buffer restart
files. Restart failures can occur when multiple time samples are being
written to a history file, and the restart occurs at a point when the
history buffers must be written to a restart file.
This commit fixes part of the problem, i.e., the errors encountered when
trying to write some history buffer restart files. But there is an
outstanding problem which will be fixed in a subsequent commit, i.e.,
some history buffer restart files will contain corrupted data which
impacts the next time sample written in the sequence. Things are OK for
subsequent time samples until another restart occurs. The workaround is
to use mfilt=1 for all history files that use averaging periods which
will require writing a history buffer restart file.
. Update to pio50_prod
. Some logfile cleanup.
Bugs fixed (include bugzilla ID): see above
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
SVN_EXTERNAL_DIRECTORIES
. update pio external to pio50_prod
models/atm/cam/bld/build-namelist
. bugfix: don't fail when the use case file is missing a default value that
has already been supplied by the user via the -namelist or -infile args.
models/atm/cam/bld/namelist_files/namelist_definition.xml
. Add 'B' as a valid value of avgflag_pertape
models/atm/cam/src/chemistry/mozart/spedata.F90
. add some initializers for filename variables
models/atm/cam/src/control/cam_history.F90
. fixes for history buffer restart files
models/atm/cam/src/control/runtime_opts.F90
. logfile output cleanup
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except pre-existing failures:
052 sm991 TSM_ccsm.sh f19_f19 F 2d ................................FAIL! rc= 4
053 er991 TER_ccsm.sh f19_f19 F 1+1d ..............................FAIL! rc= 6
054 sm992 TSM_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 4
055 er992 TER_ccsm.sh f19_g15 E 1+1d ..............................FAIL! rc= 6
056 bl992 TBL_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 4
calgary/lf95: All PASS.
calgary/pgi: All PASS.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam3_6_48
Originator(s): mvr
Date: 6/25/2009
One-line Summary: bug fix for proper selection of cam initial files with ccsm; remove
-Mrecursive from pgi debug build; prep code for intel compilers
Purpose of changes:
intel compilers limit line length to 132 chars; also, code work-arounds were required in
some instances of use association and array subscripting
the use of -Mrecursive was tripping up a cam regression test (with pgi); after discussion
it was deemed unneccessary for debug builds
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: none
Describe any changes made to the namelist:
- replaced a bad T42 cam initial dataset as the default for jan 1 startdates, 26levs
List any changes to the defaults for the boundary datasets:
- replaced a bad T42 cam initial dataset as the default for jan 1 startdates, 26levs
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/test_driver.sh
- fix for invocation of 'showproj' to enable grabbing default charge account on jaguar
M models/atm/cam/test/system/TBR.sh
M models/atm/cam/test/system/TER.sh
- bug fix to setting of tasks/threads when testing restarts and branches
M models/atm/cam/bld/Makefile.in
- removed '-Mrecursive' from pgi debug compilation flags
- made 'icc' the default c compiler for ifort builds
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- replaced a bad T42 cam initial dataset as the default for jan 1 startdates, 26levs
M models/atm/cam/bld/cam.cpl7.template
- bug fix for ccsm to grab correct cam initial file based on start date
M models/atm/cam/src/cpl_mct/atm_comp_mct.F90
M models/atm/cam/src/chemistry/mozart/spedata.F90
M models/atm/cam/src/dynamics/eul/inidat.F90
M models/atm/cam/src/dynamics/eul/restart_dynamics.F90
- workaround for use association of pio module
M models/atm/cam/src/physics/cam/sulchem.F90
M models/atm/cam/src/physics/cam/eddy_diff.F90
M models/atm/cam/src/physics/rrtmg/radiation.F90
M models/atm/cam/src/chemistry/mozart/mo_chm_diags.F90
M models/atm/cam/src/chemistry/mozart/chemistry.F90
M models/atm/cam/src/chemistry/mozart/mz_aerosols_intr.F90
- trim line lengths to <132 chars
M models/atm/cam/src/chemistry/mozart/mo_photoin.F90
M models/atm/cam/src/chemistry/mozart/mo_setaer.F90
M models/atm/cam/src/chemistry/mozart/mo_photo.F90
- workaround for array subscripting
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All pass except
052 sm991 TSM_ccsm.sh f19_f19 F 2d ................................FAIL! rc= 4
053 er991 TER_ccsm.sh f19_f19 F 1+1d ..............................FAIL! rc= 6
054 sm992 TSM_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 4
055 er992 TER_ccsm.sh f19_g15 E 1+1d ..............................FAIL! rc= 6
056 bl992 TBL_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 4
failures are pre-existing
calgary/lf95: All pass
calgary/pgi: All pass except
034 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 5
failed due to problem in baseline code that has been fixed with removal of -Mrecursive flag
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate): b4b
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_6_47
Originator(s): Jim Edwards
Date: 6/15/2009
One-line Summary: Fix model column output problem on restart
Purpose of changes:
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes: New PIO external should improve io performance.
Code reviewed by: self
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
cam_history.F90 - improved support for outputing single columns or block subsets
of model fields. Outputing columns was causing model failure on
restart.
test/system/nl_files/ghgrmp - added code to test column output so that this test
becomes a regular part of precommit and postcommit cam testing
SVN_EXTERNAL_DIRECTORIED - updated pio external
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All pass except
004 bl132 TBL.sh e32dh ghgrmp 9s ..................................FAIL! rc= 7
019 bl332 TBL.sh f4dh ghgrmp 9s ...................................FAIL! rc= 7
052 sm991 TSM_ccsm.sh f19_f19 F 2d ................................FAIL! rc= 4
053 er991 TER_ccsm.sh f19_f19 F 1+1d ..............................FAIL! rc= 6
054 sm992 TSM_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 4
055 er992 TER_ccsm.sh f19_g15 E 1+1d ..............................FAIL! rc= 6
056 bl992 TBL_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 4
calgary/lf95: All pass except
008 bl112 TBL.sh e8dm ghgrmp 9s ...................................FAIL! rc= 7
029 bl312 TBL.sh f10dm ghgrmp 9s ..................................FAIL! rc= 7
calgary/pgi or jaguar/pgi:
008 bl112 TBL.sh e8dm ghgrmp 9s ...................................FAIL! rc= 7
029 bl312 TBL.sh f10dm ghgrmp 9s ..................................FAIL! rc= 7
032 sm314 TSM.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 8
033 er314 TER.sh wg10dm outfrq3s 4+5s .............................FAIL! rc= 5
034 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 4
The ghgrmp baseline tests fail due to the addition of a new history file for column output
all results in the existing files are BFB. Bluefire 052-056 and calgary 032-034 are
legacy failures.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_6_46
Originator(s): Francis Vitt
Date: 12 June 2009
One-line Summary: WACCM updates
Purpose of changes:
Bring the trunk up to date with waccm branch.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Brian Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
A models/atm/cam/bld/namelist_files/use_cases/waccm_1850_cntrl.xml
- for ccsm 1850 "track 1" simulation
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- Corrected RSF photolysis file
M models/atm/cam/src/chemistry/mozart/linoz_data.F90
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- INTERP_MISSING_MONTHS option added
M models/atm/cam/src/chemistry/mozart/mo_sad.F90
- corrected strat aero rad forcing
M models/atm/cam/bld/namelist_files/use_cases/waccm_refb1.xml
M models/atm/cam/bld/build-namelist
M models/atm/cam/src/physics/cam/radsw.F90
M models/atm/cam/src/physics/cam/phys_control.F90
M models/atm/cam/src/physics/cam/solvar_interface.F90
M models/atm/cam/src/physics/cam/cloud_fraction.F90
M models/atm/cam/src/physics/cam/tidal_diag.F90
M models/atm/cam/src/physics/cam/vertical_diffusion.F90
M models/atm/cam/src/physics/waccm/gw_drag.F90
M models/atm/cam/src/chemistry/utils/mo_solar_parms.F90
M models/atm/cam/src/chemistry/mozart/solar_photons.F90
M models/atm/cam/src/chemistry/mozart/mo_chm_diags.F90
M models/atm/cam/src/chemistry/mozart/solvar_interface.F90
M models/atm/cam/src/chemistry/mozart/chemistry.F90
M models/atm/cam/src/chemistry/mozart/solvar_data.F90
M models/atm/cam/src/dynamics/fv/pfixer.F90
M models/atm/cam/src/dynamics/fv/cd_core.F90
- merged in mods from waccm16_cam3_5_48
- enabled stratospheric aerosols radiation feedback
M models/atm/cam/src/chemistry/mozart/spedata.F90
- write spedata log messages only from masterproc
M models/atm/cam/src/control/cam_history.F90
- change to appease Lahey compiler
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
030 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7
expected to fail due to waccm updates
052 sm991 TSM_ccsm.sh f19_f19 F 2d ................................FAIL! rc= 4
053 er991 TER_ccsm.sh f19_f19 F 1+1d ..............................FAIL! rc= 6
054 sm992 TSM_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 4
055 er992 TER_ccsm.sh f19_g15 E 1+1d ..............................FAIL! rc= 6
056 bl992 TBL_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 4
calgary/lf95:
034 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 7
This is expected to fail due to parameter changes in waccm gravity wave routine.
calgary/pgi or jaguar/pgi:
032 sm314 TSM.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 8
033 er314 TER.sh wg10dm outfrq3s 4+5s .............................FAIL! rc= 5
034 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 4
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_6_45
Originator(s): eaton
Date: Thu Jun 11 12:35:34 MDT 2009
One-line Summary: bugfix for branch runs via ccsm scripts
Purpose of changes:
. The camdom component isn't using netcdf for its restart files. Need to
back out the change in cam.cpl7.template that assumed it was.
Bugs fixed (include bugzilla ID): see above
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/cam.cpl7.template
. remove ".nc" from value of nrevsn_dom_nml
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: not done
calgary/lf95: not done
calgary/pgi or jaguar/pgi: not done
No regression testing was done for this tag because the change in
cam.cpl7.template is only tested via the ccsm scripts, and our tests of the
ccsm scripts are currently failing for other reasons.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam3_6_44
Originator(s): eaton
Date: Wed Jun 10 11:29:39 MDT 2009
One-line Summary: Add directories to enable optional esmf component interfaces.
Purpose of changes:
. For each component the directories cpl_mct and cpl_esmf are added. The
current mct based component interfaces (in the modules xxx_comp_mct.F90)
are moved to the cpl_mct directory. The esmf component interfaces will
be added to the cpl_esmf directories.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system:
. Add -comp_intf argument to configure. Once the esmf interfaces are in
place, they will be used by adding "-comp_intf esmf" to the configure
commandline.
. Add BGP and BGL support to Makefile for chemistry preprocessor
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
models/atm/cam/src/control/atm_comp_mct.F90
models/ice/csim4/ice_comp_mct.F90
models/ocn/dom/ocn_comp_mct.F90
. These files were moved to the new cpl_mct directories.
models/ocn/som/mixed_layer.F90
models/ocn/som/ocn_filenames.F90
models/ocn/som/ocn_spmd.F90
models/ocn/som/ocn_types.F90
models/ocn/som/ocn_time_manager.F90
models/ocn/som/somint.F90
models/ocn/som/ocn_constants.F90
models/ocn/som/ocn_comp.F90
models/ocn/som/ocean_data.F90
models/ocn/som/somini.F90
models/ocn/som/ocn_comp_mct.F90
. remove unused som code
List all subroutines added and what they do:
models/atm/cam/src/cpl_mct/atm_comp_mct.F90
models/ice/csim4/cpl_mct/ice_comp_mct.F90
models/ocn/dom/cpl_mct/ocn_comp_mct.F90
. These files were moved into the new cpl_mct directories, and the SEQ_MCT
ifdefs were removed.
List all existing files that have been modified, and describe the changes:
models/atm/cam/SVN_EXTERNAL_DIRECTORIES
. chem_proc -> trunk_tags/chem_proc3_6_01
models/atm/cam/bld/configure
. add new option comp_intf to select which component interfaces to use.
Can be set to either mct or esmf. Default: mct.
. modify filepaths for cam, dom, and csim4 to use the cpl_* directories.
Wait to update the filepaths for other components until updating the
externals to versions that have the cpl_* directories implemented. Once
this is done the -DSEQ_MCT can be removed from Makefile.in.
models/atm/cam/bld/config_files/definition.xml
. add comp_intf parameter w/ valid values mct or esmf. Default is mct.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
052 sm991 TSM_ccsm.sh f19_f19 F 2d ................................FAIL! rc= 4
053 er991 TER_ccsm.sh f19_f19 F 1+1d ..............................FAIL! rc= 6
054 sm992 TSM_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 4
055 er992 TER_ccsm.sh f19_g15 E 1+1d ..............................FAIL! rc= 6
056 bl992 TBL_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 4
These are pre-existing failures.
calgary/lf95: All PASS
calgary/pgi: All PASS except:
032 sm314 TSM.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 8
033 er314 TER.sh wg10dm outfrq3s 4+5s .............................FAIL! rc= 5
034 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 4
These are pre-existing failures.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam3_6_43
Originator(s): eaton, mirin
Date: Thu Jun 4 12:50:57 MDT 2009
One-line Summary: Misc bug fixes.
Purpose of changes:
. Workaround for PGI compiler bug.
. Fix bug in hfilename_spec. User specification of history filenames was
being ignored.
. Change to BG/L section of Makefile.in (Mirin).
Bugs fixed (include bugzilla ID):
. hfilename_spec wasn't being recognized. There was a missing broadcast of
the namelist variable.
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/Makefile.in
. change compilers for BG/L to mpxlf90 and mpxlc
models/atm/cam/src/control/cam_history.F90
. remove LLNL_PELOTON
models/atm/cam/src/control/runtime_opts.F90
. add an mpi broadcast for hfilename_spec
models/atm/cam/src/dynamics/fv/metdata.F90
. remove unneeded use shr_kind_mod declarations. This is a workaround for
a pgi compiler bug, but resulting code is cleaner.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except
052 sm991 TSM_ccsm.sh f19_f19 F 2d ................................FAIL! rc= 4
053 er991 TER_ccsm.sh f19_f19 F 1+1d ..............................FAIL! rc= 6
054 sm992 TSM_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 4
055 er992 TER_ccsm.sh f19_g15 E 1+1d ..............................FAIL! rc= 6
056 bl992 TBL_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 4
These are pre-existing failures.
calgary/lf95: All PASS
calgary/pgi: All PASS except
032 sm314 TSM.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 8
033 er314 TER.sh wg10dm outfrq3s 4+5s .............................FAIL! rc= 5
034 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 4
sm314 is a pre-existing failure.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB.
===============================================================
===============================================================
Tag name: cam3_6_42
Originator(s): eaton, sawyer, mvertens, mvr
Date: Tue Jun 2 09:27:01 MDT 2009
One-line Summary: Fixes for ccsm branch runs, intel compiler, low order FV interpolation.
Purpose of changes:
. Fix bug in cam.cpl7.template to restore branch run functionality to ccsm
scripts (mvertens)
. Fixes for intel compiler.
. Bug fixes from Will Sawyer for low order interpolation options in the FV
dycore.
. Update default 1 degree SST climatology dataset (mvr).
Bugs fixed (include bugzilla ID): 970
. 970 - fix cam.cpl7.template to restore branch runs
. Fixes for intel compiler include reformatting lines longer than 132
characters and replacing the LLNL_PELOTON fix with a more generic one.
. Fix problems in FV dycore with setting iord/jord equal to 1 or 2.
Problems were due to indexing bugs.
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets:
. Update the default 1 degree SST dataset to
sst_HadOIBl_bc_0.9x1.25_clim_c040926a.nc. This is the same as the
c040926 version except that the calendar attribute has been added to the
time coordinate variable.
Describe any substantial timing or memory changes: none
Code reviewed by: eaton, mirin (FV indexing)
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/cam.cpl7.template
. add ".nc" suffixes to nrevsn* values
models/atm/cam/src/dynamics/fv/cd_core.F90
models/atm/cam/src/dynamics/fv/sw_core.F90
models/atm/cam/src/dynamics/fv/tp_core.F90
. fixes for low order interpolation schemes.
models/atm/cam/bld/Makefile.in
models/atm/cam/src/chemistry/utils/tracer_data.F90
models/atm/cam/src/control/hycoef.F90
models/atm/cam/src/dynamics/fv/restart_dynamics.F90
models/atm/cam/src/physics/cam/aer_rad_props.F90
models/atm/cam/src/physics/cam/phys_buffer.F90
models/atm/cam/src/physics/cam/rad_constituents.F90
models/atm/cam/src/physics/cam/restart_physics.F90
models/atm/cam/src/utils/time_manager.F90
. fixes for intel compiler
. remove need to #define LLNL_PELOTON
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
052 sm991 TSM_ccsm.sh f19_f19 F 2d ................................FAIL! rc= 4
053 er991 TER_ccsm.sh f19_f19 F 1+1d ..............................FAIL! rc= 6
054 sm992 TSM_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 4
055 er992 TER_ccsm.sh f19_g15 E 1+1d ..............................FAIL! rc= 6
056 bl992 TBL_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 4
Pre-existing failures due to inconsistency between ccsm scripts and clm tag.
bangkok/lf95: All PASS.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam3_6_41
Originator(s): Jim Edwards, Pat Worley
Date: 05-28-2009
One-line Summary: Additional communications algorythms, cice bug fix
Purpose of changes: Cray XT machines were having mpi buffer issues at large proc counts
this pio update resolves those issues.
Updated PIO to trunk_tags/pio47_prod
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: self
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
Moved call to close initial file from cam_init to cam_finalize in
cam_comp.F90 - the ice component
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: no failures other than ccsm tests carried over from cam3_6_38
bangkok/lf95: no failures
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_6_40
Originator(s): Jim Edwards
Date: May 12, 2009
One-line Summary: bug fixes for cam3_6_39
Purpose of changes: Remedy bugs and performance in cam3_6_39
Bugs fixed (include bugzilla ID): Fixed a postcommit PGI compiler issue in
PIO. Fixed problems with history column output.
Fixed a post commit failure in sld dycore.
NOTE: There are still outstanding issues related to running on
CRAY XT systems that we continue to work on.
Describe any changes made to build system:
PIO updated to pio43_prod
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: self, eaton
List all subroutines eliminated:
Removed dead code from cam_history.F90 : read_hbuf, write_hbuf, readin_hbuf
mpigatherv_hbuf, wrap_put_vara_hbuf,
gather_chunk_to_field_hbuf, scatter_field_to_chunk_hbuf
List all subroutines added and what they do:
Added code for storing iodesc in pio_utils.F90: find_iodesc, clean_iodesc_list
List all existing files that have been modified, and describe the changes:
src/control/ncdio_atm.F90
src/control/startup_initialconds.F90
src/control/cam_restart.F90
src/control/cam_history.F90
src/utils/pio_utils.F90
src/physics/cam/restart_physics.F90
src/physics/cam/phys_buffer.F90
src/dynamics/sld/restart_dynamics.F90
src/dynamics/homme/inidat.F90
The iodesc variable was changed to a pointer so that it could be reused, this
improved performance and reduced memory relative to cam3_6_39 and brought it
back inline with cam3_6_38. Capability to output column subsets of history
variables was restored. Some dead code was removed.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All pass against cam3_6_39 except ccsm tests.
post commit test pf385 was run against cam3_6_38 and also passed.
additional testing with ndens=2 was also conducted
052 sm991 TSM_ccsm.sh f19_f19 F 2d ................................FAIL! rc= 4
053 er991 TER_ccsm.sh f19_f19 F 1+1d ..............................FAIL! rc= 6
054 sm992 TSM_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 4
055 er992 TER_ccsm.sh f19_g15 E 1+1d ..............................FAIL! rc= 6
056 bl992 TBL_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 4
bangkok/lf95: All pass against cam3_6_39
PGI tests were also conducted against cam3_6_38, all pass except
expected failures.
032 sm314 TSM.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 8
033 er314 TER.sh wg10dm outfrq3s 4+5s .............................FAIL! rc= 5
034 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 4
POST test sm371 Fails on jaguar with DEBUG=true - this is a feature of the PGI compiler
which is giving an error when mct passes an array of length 0. This is legal fortran
and should not be trapped.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_6_39
Originator(s): Jim Edwards, Mark Taylor
Date: 01-09-2009
One-line Summary: Add pio support for initial file
Purpose of changes: parallelize IO
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: Removed PIO from CAM build in
CCSM: PIO is now built as a library
Describe any changes made to the namelist: Added pio_input logical to
control use of PIO in reading initial file
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
Added PIO specific versions of inidat for all dycores
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/TBR.sh
Updated to match new netcdf restart name string
M models/atm/cam/test/system/nl_files/off1.9x2.5
M models/atm/cam/test/system/nl_files/ghgrmp
M models/atm/cam/test/system/nl_files/off2x2.5
Removed obsolete pio_restart variable
M models/atm/cam/bld/build-namelist
Clean up homme dycore options, add pio_inparm namelist for driver
M models/ice/csim4/ice_comp.F90
T3 field is used to initialize TBOT when TBOT not present in initial file, however this field was being allocated and read even when
TBOT was present. Refactored to read T3 in parallel and
only when required
M models/atm/cam/src/control/runtime_opts.F90
Broadcast pertlim variable to support parallel IO
M models/atm/cam/src/control/cam_history.F90
Added cell area variable for homme dycore
M models/atm/cam/src/control/readinitial.F90
M models/atm/cam/src/control/ncdio_atm.F90
M models/atm/cam/src/control/startup_initialconds.F90
M models/atm/cam/src/utils/pio_utils.F90
M models/atm/cam/src/physics/cam/restart_physics.F90
M models/atm/cam/src/dynamics/sld/inidat.F90
M models/atm/cam/src/dynamics/eul/inidat.F90
M models/atm/cam/src/dynamics/homme/inidat.F90
M models/atm/cam/src/dynamics/fv/inidat.F90
Added PIO versions of initial file read routines.
M models/atm/cam/src/physics/cam/cam3_aero_data.F90
M models/atm/cam/src/physics/cam/seasalt_intr.F90
M models/atm/cam/src/physics/cam/tracers.F90
M models/atm/cam/src/physics/cam/progseasalts_intr.F90
M models/atm/cam/src/physics/cam/dust_intr.F90
M models/atm/cam/src/physics/cam/tracers_suite.F90
M models/atm/cam/src/physics/cam/stratiform.F90
M models/atm/cam/src/physics/cam/carbon_intr.F90
M models/atm/cam/src/physics/cam/co2_cycle.F90
M models/atm/cam/src/physics/cam/chemistry.F90
M models/atm/cam/src/physics/cam/aerosol_intr.F90
M models/atm/cam/src/physics/cam/sulfur_intr.F90
M models/atm/cam/src/physics/waccm/tracers.F90
M models/atm/cam/src/chemistry/mozart/chemistry.F90
Refactored initialization of Constituents to allow for parallel
initialization.
M models/atm/cam/src/dynamics/fv/dyn_grid.F90
M models/atm/cam/src/dynamics/homme/dyn_grid.F90
Added subroutines get_gcol_lat and get_gcol_lon to map a global
column id to a lat/lon pair
M models/atm/cam/src/dynamics/homme/restart_dynamics.F90
M models/atm/cam/src/dynamics/homme/inital.F90
M models/atm/cam/src/dynamics/fv/dp_coupling.F90
Simplified an indexing equation
M models/atm/cam/src/dynamics/fv/dyn_comp.F90
Added default initialization of dyn_in members to inf for debugging
uninitialized variables
M models/atm/cam/src/dynamics/fv/metdata.F90
Cleaned up stdout
M SVN_EXTERNAL_DIRECTORIES
Updated PIO external and mct serial driver
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All pass except
051 bl731 TBL.sh h16x4dm aqua 9s ..................................FAIL! rc= 7
expected failure due to homme dynamics updates
052 sm991 TSM_ccsm.sh f19_f19 F 2d ................................FAIL! rc= 4
053 er991 TER_ccsm.sh f19_f19 F 1+1d ..............................FAIL! rc= 6
054 sm992 TSM_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 4
055 er992 TER_ccsm.sh f19_g15 E 1+1d ..............................FAIL! rc= 6
056 bl992 TBL_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 4
carried forward from cam3_6_38 - This commit passed these tests as run against
cam3_6_37 prior to the cam3_6_38 merge.
bangkok/lf95: All pass except
037 bl315 TBL.sh f10c3dm outfrq3s 9s ..............................FAIL! rc= 7
fails due to round off differences in polar averaging of TCWAT field
as initialized from T in physpkg.F90 - see cam3_6_35 for details.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
Tag name: cam3_6_38
Originator(s): eaton
Date: Tue May 5 11:34:35 MDT 2009
One-line Summary: Fix for running F compsets from ccsm4_0_beta16 and later.
Purpose of changes:
. modify build-namelist so that it generates a landmask file for the CAMDOM
component namelist that is consistent with the one generated for the CICE
namelist.
Note that the fix is for running the F compsets from the ccsm4_0_beta16
and later tags. Using the ccsm scripts to run F compsets from this cam
tag is broken due to an inconsistency between the scripts and the CLM
tag. We will be updating the CLM external to fix this problem shortly.
. Update to the latest scripts and pio tags. Updating to the latest
scripts tag has temporarily broken the ability to run F compsets via the
ccsm scripts in this CAM tag.
Bugs fixed (include bugzilla ID):
. Fix for running F compsets from ccsm4_0_beta16 and later.
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
SVN_EXTERNAL_DIRECTORIES
. models/utils/pio -> pio38_prod/pio
. scripts -> scripts4_090428
cam/bld/build-namelist
. modify build-namelist to look at the cice namelist defaults to determine
whether the gx1v6 mask is being used. If a gx1v6 mask is used, the
default for the focndomain variable will be determined using the attribue
mask="gx1v6".
. don't write dom_inparm namelist group when it isn't needed, i.e., for
ideal, adiabatic, and aqua-planet modes.
cam/bld/namelist_files/namelist_defaults_cam.xml
. remove defaults for old 1x1.25 and 2x2.5 resolutions which are no longer
supported.
. add focndomain defaults for gx1v6 mask.
cam/src/chemistry/utils/tracer_data.F90
cam/src/physics/cam/polar_avg.F90
. Change name of polar_avg_mod to polar_avg to match the filename (required
by dependency generator).
cam/test/system/TCB_ccsm.sh
. change the xmlchange command to modify env_build.xml rather than
env_mach_generic.xml. The EXEROOT variable was moved to env_build.xml.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: all PASS except
052 sm991 TSM_ccsm.sh f19_f19 F 2d ................................FAIL! rc= 4
053 er991 TER_ccsm.sh f19_f19 F 1+1d ..............................FAIL! rc= 6
054 sm992 TSM_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 4
055 er992 TER_ccsm.sh f19_g15 E 1+1d ..............................FAIL! rc= 6
056 bl992 TBL_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 4
These tests fail due to an inconsistency between the scripts and CLM tags.
bangkok/lf95: all PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam3_6_37
Originator(s): pworley
Date: Mon Apr 27 2009
One-line Summary: more accurate reproducible sum; additional OpenMP
parallelism.
Purpose of changes:
1) Modify repro_sum fixed-point algorithm to improve accuracy.
In certain situations, numerical cancellation can degrade accuracy
of original repro_sum algorithm as compared to floating-point-based
algorithms. New algorithm is nearly exact, employing variable
precision as needed. This results in roundoff level differences.
2) Introduce OpenMP parallelism in physics and chemistry routines
that are not called within the major loop over chunks.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes:
- Cost of new repro_sum algorithm is higher than that of the old
algorithm. Experiments indicate that the additional cost is
not significant.
Code reviewed by: self
List all subroutines eliminated:
List all subroutines added and what they do:
physics/cam/polar_avg.F90 -
moved polar averaging routines out of repro_sum_mod.F90 and
into their own module
List all existing files that have been modified, and describe the changes:
chemistry/utils/prescribed_aero.F90 -
slight restructuring to expose OpenMP parallelism;
addition of OpenMP directives
chemistry/utils/tracer_data.F90 -
'use' polar_avg_mod instead of repro_sum_mod;
addition of OpenMP directives
dynamics/fv/spmd_dyn.F90 -
changed default communication protocol for I/O-related scatter
routines in FV, to address runtime failure on Cray XT system
physics/cam/ghg_data.F90 -
addition of OpenMP directive
physics/cam/phys_grid.F90 -
bug fix to unadvertised load balancing option
utils/repro_sum_mod.F90 -
generalization of fixed-point reproducible sum algorithm to
enable (almost) exact arithmetic
test/system/input_tests_master -
bug fix (at request of Brian Eaton)
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except
015 bl331 TBL.sh f4dh co2rmp 9s ...................................FAIL! rc= 7
019 bl332 TBL.sh f4dh ghgrmp 9s ...................................FAIL! rc= 7
021 bl333 TBL.sh f4c351pdh aqpgro 3s ..............................FAIL! rc= 7
024 bl334 TBL.sh f4adh adia 9s ....................................FAIL! rc= 7
027 bl335 TBL.sh f4idh idphys 9s ..................................FAIL! rc= 7
030 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7
033 bl337 TBL.sh f4dh fv2d_8tsk 9s ................................FAIL! rc= 7
037 bl371 TBL.sh f1.9dm fvvp_lb2 9s ...............................FAIL! rc= 7
041 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7
044 bl374 TBL.sh fm1.9dh outfrq3s 9s ..............................FAIL! rc= 7
048 bl387 TBL.sh f1.9c351dh outfrq3s+1850-2005_track1 9s ..........FAIL! rc= 4
056 bl992 TBL_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 7
bangkok/lf95: All PASS except
021 bl311 TBL.sh f10t5mdm ttrac 9s ................................FAIL! rc= 4
029 bl312 TBL.sh f10dm ghgrmp 9s ..................................FAIL! rc= 4
031 bl313 TBL.sh f10pdm aqpgro 3s .................................FAIL! rc= 4
034 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 4
037 bl315 TBL.sh f10c3dm outfrq3s 9s ..............................FAIL! rc= 4
040 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 4
All baselines using FV fail due to roundoff introduced in changes to
reproducible sum algorithm.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers:
Roundoff diffs in all configurations using FV.
If bitwise differences were observed, how did you show they were no worse
than roundoff?
pergro test on bluefire
===============================================================
Tag name: cam3_6_36
Originator(s): fvitt, eaton
Date: Tue Apr 21 14:56:51 MDT 2009
One-line Summary: new solar data code; update prescribed forcing data.
Purpose of changes:
. A new module (solar_data) has been added with capabilities to read
datasets containing either total solar irradiance (TSI) or spectrally
resolved solar irradiance (SSI). Currently only the TSI code, needed for
the track1 transient runs, is active.
. Updates to prescribed forcing data:
- New TSI and SSI datasets from Caspar Ammann.
- New 1850 and 2000 climatologies for both BAM and ozone from
Jean-Francois Lamarque.
- New GHG data (CO2, CH4, N2O, F11*, F12) from Doug Kinnison.
- New volcanic emissions (eruptive) data from Caspar Ammann.
. A new use case (1850-2005_track1) has been implemented for the 1850-2005
transient runs for track1 physics.
. configure has been modified to invoke the CICE configure with the args
"-ntr_aero 0 -ntr_pond 1 -ntr_iage 0". The -ntr_aero setting was already
being used; the -ntr_pond and -ntr_iage settings are new. This is for
consistency with how the CCSM F cases are configured. The values for
-ntr_pond and -ntr_iage are the defaults, so this doesn't change
answers.
Bugs fixed (include bugzilla ID):
. The chemistry/utils/prescribed_* modules all contained a bug which did
not allow the data_type variable to be set to INTERP_MISSING_MONTHS (the
module variable was declared char*8).
. The interpic_new utility was assuming that all global attributes are text
strings. This was causing failures on input files produced by the NCO
operators which contain a global attribute of type integer.
. build-namelist wasn't recognizing a value of "0" in the use case file
because a conditional statement was testing whether the value was true
rather than whether the value was defined.
Describe any changes made to build system: none
Describe any changes made to the namelist:
. new namelist variables for the solar_data module:
character(len=256) :: solar_data_file = ''
character(len=8) :: solar_data_type = 'SERIAL' ! "FIXED" or "SERIAL"
integer :: solar_data_ymd = 0 ! YYYYMMDD for "FIXED" type
integer :: solar_data_tod = 0 ! seconds of day for "FIXED" type
real(r8) :: solar_const ! constant TSI (W/m2)
. The following namelist variables have been removed:
bndtvscon, rampyear_scon, scenario_scon, scon
List any changes to the defaults for the boundary datasets:
. updates to 2000 climatology for ozone, prescribed BAM
. update fixed GHG values to be consistent with year 2000 from the new GHG
dataset.
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
models/atm/cam/src/physics/cam/ramp_scon.F90
. old solar irradiance module
List all subroutines added and what they do:
models/atm/cam/bld/namelist_files/use_cases/1850-2005_track1.xml
. use case for track1 1850-2005 transient runs
models/atm/cam/src/chemistry/utils/solar_data.F90
. new module to manage solar irradiance specification
models/atm/cam/test/system/config_files/f1.9c351dh
. config params for fv, 1.9x2.5, cam3_5_1, debug, hybrid.
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/build-namelist
. unless the solar_data_file is specified by the user or by a use case, set
the default solar constant.
. fix bug in processing use case values (need to check whether value is
defined rather than whether it's true)
models/atm/cam/bld/namelist_files/namelist_definition.xml
. remove bndtvscon, rampyear_scon, scenario_scon, scon
. add solar_data_file, solar_data_type, solar_data_ymd, solar_data_tod,
solar_const
. change type of prescribed_*_type variables from char*8 to char*32. Add
the appropriate valid values to each.
models/atm/cam/bld/namelist_files/use_cases/1850_track1.xml
models/atm/cam/bld/namelist_files/use_cases/1870_bam.xml
models/atm/cam/bld/namelist_files/use_cases/1870_control.xml
models/atm/cam/bld/namelist_files/use_cases/1870_prog_aero.xml
. convert scon -> solar_const (change units too)
models/atm/cam/bld/namelist_files/use_cases/1990_control.xml
models/atm/cam/bld/namelist_files/use_cases/1990_prog_aero.xml
. remove scon spec so default present day value is used (valid for 1990 or 2000).
models/atm/cam/src/chemistry/mozart/solvar_woods.F90
. use sol_tsi from solar_data instead of scon from cam_control_mod.
models/atm/cam/src/chemistry/utils/prescribed_aero.F90
models/atm/cam/src/chemistry/utils/prescribed_ghg.F90
models/atm/cam/src/chemistry/utils/prescribed_ozone.F90
models/atm/cam/src/chemistry/utils/prescribed_volcaero.F90
. change declaration of datatype variable from char*8 to char*32. Needs to
be long enough for the value INTERP_MISSING_MONTHS.
models/atm/cam/src/control/cam_control_mod.F90
. remove doramp_scon and scon
models/atm/cam/src/control/runtime_opts.F90
. add call to solar_data_readnl
. remove use ramp_scon
. remove scenario_scon, rampyear_scon, scon
models/atm/cam/src/physics/cam/advnce.F90
. add call to solar_data_advance
. remove ramp_scon
models/atm/cam/src/physics/cam/physconst.F90
. add speed of light and Planck's constant.
models/atm/cam/src/physics/cam/physpkg.F90
. add call to solar_data_init
models/atm/cam/src/physics/cam/radiation.F90
. use sol_tsi from solar_data instead of scon from cam_control_mod.
use local scon to convert units to mW/m2.
models/atm/cam/src/physics/rrtmg/radiation.F90
. use sol_tsi from solar_data instead of scon from cam_control_mod.
conversion to mks units no longer necessary.
models/atm/cam/test/system/input_tests_master
. add new tests:
- sm387 TSM.sh f1.9c351dh outfrq3s+1850-2005_track1 9s
- er387 TER.sh f1.9c351dh outfrq3s+1850-2005_track1 4+5s
- br387 TBR.sh f1.9c351dh outfrq3s+1850-2005_track1 6+3s
- bl387 TBR.sh f1.9c351dh outfrq3s+1850-2005_track1 9s
models/atm/cam/test/system/tests_pretag_bluefire
. replace {sm,er,br,bl}382 by {sm,er,br,bl}387
models/atm/cam/tools/interpic_new/fmain.F90
. remove assumption that copied attributes are text strings
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except
004 bl132 TBL.sh e32dh ghgrmp 9s ..................................FAIL! rc= 7
006 bl133 TBL.sh e32pdh aqpgro 3s .................................FAIL! rc= 7
011 bl151 TBL.sh e64h outfrq24h 2d ................................FAIL! rc= 7
015 bl331 TBL.sh f4dh co2rmp 9s ...................................FAIL! rc= 7
019 bl332 TBL.sh f4dh ghgrmp 9s ...................................FAIL! rc= 7
021 bl333 TBL.sh f4c351pdh aqpgro 3s ..............................FAIL! rc= 7
030 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7
033 bl337 TBL.sh f4dh fv2d_8tsk 9s ................................FAIL! rc= 7
037 bl371 TBL.sh f1.9dm fvvp_lb2 9s ...............................FAIL! rc= 7
041 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7
044 bl374 TBL.sh fm1.9dh outfrq3s 9s ..............................FAIL! rc= 7
051 bl731 TBL.sh h16x4dm aqua 9s ..................................FAIL! rc= 7
056 bl992 TBL_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 7
bangkok/lf95: All PASS except
004 bl111 TBL.sh e8t5mdm ttrac 9s .................................FAIL! rc= 7
008 bl112 TBL.sh e8dm ghgrmp 9s ...................................FAIL! rc= 7
010 bl113 TBL.sh e8pdm aqpgro 3s ..................................FAIL! rc= 7
014 bl114 TBL.sh e8dm co2rmp 9s ...................................FAIL! rc= 7
021 bl311 TBL.sh f10t5mdm ttrac 9s ................................FAIL! rc= 7
029 bl312 TBL.sh f10dm ghgrmp 9s ..................................FAIL! rc= 7
031 bl313 TBL.sh f10pdm aqpgro 3s .................................FAIL! rc= 7
034 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 7
037 bl315 TBL.sh f10c3dm outfrq3s 9s ..............................FAIL! rc= 7
040 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: Answers change due to update of
prescribed forcing datasets. There was also a roundoff level change in
waccm_mozart due to changing the units of the solar constant input.
===============================================================
===============================================================
Tag name: cam3_6_35
Originator(s): fvitt, aconley, bardeenc, jedwards, eaton
Date: Mon Apr 13 13:36:28 MDT 2009
One-line Summary: volcano emissions and optics; 1850_track1 updates;
build-namelist mods (improve AMIP setup)
Purpose of changes:
. Add new code for reading volcano emission datasets. This includes a new
feature to mask emissions below the tropopause. Currently the volcanic
emissions are radiatively active by default only when used with camrt and
with BAM (prescribed or prognostic). This functionality is needed for the
track1 transient runs.
. Add module to diagnose the location of the tropopause. Multiple
algorithms are provided, some of which may not be able to identify a
tropopause in all situations. To handle these cases, an analytic
defintion and a climatology are provided that can be used to fill in
when the original algorithm fails. The troppause temperature and
pressure are determined and can be output to the history file. (From
Charles Bardeen.)
. Tuning mod: set rhminl=0.92 for FV 1.9x2.5 w/ cam3 or cam3_5* physics.
. Update the 1850_track1 use case with new 1850 prescribed BAM aerosols
from Jean-Francois Lamarque.
. Update build-namelist to allow setting all namelist variables recognized
by the perf_mod module. Also change default for timing output files to
use a single file.
. Update build-namelist to enforce consistency between the SST and ocean
domain files used by CAM-DOM and the prescribed CICE mode. Also add
variables to the CAM namelist that are needed to specify the first and
last years of a multi-year SST dataset (for AMIP runs).
. Update the FV inidat code to replace global sums by distributed sums in
the polar averaging. This results in roundoff level differences. (From
Jim Edwards.)
. Add support for radius dependent optics calculation for volcanic
aerosols. This code will only be turned on for RRTMG.
Bugs fixed (include bugzilla ID):
. The volcano emissions code used in cam3 was broken when changes were made
to remove assumptions about rectangular lat/lon grids from the physics
code. The new code added with this commit replaces the old code which
has been removed.
Describe any changes made to build system: none
Describe any changes made to the namelist: new volcano emission variables
character(len=16) :: prescribed_volcaero_name ! name of variable containing volcano mmr (default: MMRVOLC)
character(len=256) :: prescribed_volcaero_file ! name of file containing emission data
character(len=256) :: prescribed_volcaero_filelist ! name of file containing list of emission dataset file names
character(len=256) :: prescribed_volcaero_datapath ! absolute pathname of file containing emission data
character(len=8) :: prescribed_volcaero_type ! type of time interpolation (default: SERIAL)
logical :: prescribed_volcaero_rmfile ! true => remove dataset after using (default: false)
integer :: prescribed_volcaero_ymd ! start ymd for interpolation if different from model time
integer :: prescribed_volcaero_tod ! time of day (sec) relative to prescribed_volcaero_ymd
. Remove variables used by old volcano emission modules: bndtvvolc and
strat_volcanic.
. Add support for all namelist variables recognized by the perf_mod module.
. Add the variables stream_year_first and stream_year_last which are used
to specify, respectively, the first and last years of data in a
multi-year SST dataset. This is needed for AMIP runs.
. The new tropopause diagnostic code uses a tropopause climatology file.
The default (which is also used by the trop_mozart code) can be
overridden via the namelist variable tropopause_climo_file.
List any changes to the defaults for the boundary datasets:
. new dataset for 1850 climatology of prescribed bulk aerosols.
. replace sst_HadOIBl_bc_32x64_clim_c030228.nc by
sst_HadOIBl_bc_32x64_clim_c050526.nc because the latter file has a time
coordinate that's recognized by cice.
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
models/atm/cam/src/physics/cam/volcanicmass.F90
models/atm/cam/src/physics/cam/volcemission.F
. old volcanic emission code
List all subroutines added and what they do:
models/atm/cam/src/chemistry/utils/prescribed_volcaero.F90
. module to read prescribed volcano emissions dataset
models/atm/cam/src/physics/cam/tropopause.F90
. new tropopause diagnostic code.
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/Makefile.in
. add -gopt to pgi FFLAGS for non-debug build
models/atm/cam/bld/build-namelist
. modify to make the prescribed volanic aerosols radiatively active if
present.
. change the default so a single file is produced for timing output
. modify how default is added for the fsurdat file
. invoke the cice build-namelist with variables set to insure that the
settings of SST dataset and ocn domain file are consistent with the
CAM-DOM settings.
. Add ability to set the variables stream_year_first and stream_year_last
directly in the CAM namelist as a convenience for setting up AMIP runs.
These settings are passed to the cice build-namelist which is the
component that requires setting them.
. add default for tropopause climatology file.
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. add default for tropopause climatology file.
. change T21 SST dataset.
models/atm/cam/bld/namelist_files/namelist_definition.xml
. Add new namelist variables for prescribed_volcaero module.
. Remove old volcano emissions variables
. Add all namelist variables recognized by the perf_mod module.
. add stream_year_first and stream_year_last. These set the first and last
years respectively in a multi-year SST dataset read by cice in prescribed
mode. They must be set for AMIP style runs.
. add definition for tropopause climatology file.
models/atm/cam/bld/namelist_files/use_cases/1850_track1.xml
. new dataset for 1850 climatology of prescribed bulk aerosols.
models/atm/cam/src/chemistry/utils/prescribed_aero.F90
models/atm/cam/src/chemistry/utils/prescribed_ghg.F90
models/atm/cam/src/chemistry/utils/prescribed_ozone.F90
. remove a save attribute that caused compiler warnings (because there's
already a global save statement)
models/atm/cam/src/chemistry/utils/tracer_data.F90
. fix indexing bug in read_za_trc
models/atm/cam/src/dynamics/fv/inidat.F90
. Replace global sums done on masterproc by distributed sums. Needed for
polar averaging.
models/atm/cam/src/control/runtime_opts.F90
models/atm/cam/src/physics/cam/advnce.F90
models/atm/cam/src/physics/cam/initindx.F90
models/atm/cam/src/physics/cam/physpkg.F90
models/atm/cam/src/physics/cam/restart_physics.F90
. add hooks for new prescribed_volcaero module
models/atm/cam/src/control/filenames.F90
models/atm/cam/src/control/runtime_opts.F90
models/atm/cam/src/physics/cam/advnce.F90
models/atm/cam/src/physics/cam/aerosol_intr.F90
models/atm/cam/src/physics/cam/initindx.F90
models/atm/cam/src/physics/cam/physpkg.F90
models/atm/cam/src/physics/cam/sulfur_intr.F90
. remove references to old volcanicmass and volcemission modules
models/atm/cam/src/chemistry/utils/prescribed_volcaero.F90
. replace simple tropopause expression by a utility that provides multiple
methods for diagnosing the tropopause.
models/atm/cam/src/physics/cam/aer_rad_props.F90
. add code to compute optics for radius-dependent quantities
models/atm/cam/src/physics/cam/cloud_fraction.F90
. set rhminl=0.92 for FV 1.9x2.5 with cam3 or cam3_5* physics
models/atm/cam/src/physics/cam/phys_prop.F90
. add code to read radius-dependent mass-specific quantities: r_sw_ext,
r_sw_scat, r_sw_ascat, r_lw_abs, mu
models/atm/cam/src/physics/cam/physpkg.F90
. add calls to tropopause_init
models/atm/cam/src/physics/cam/rad_constituents.F90
. add access to new properties: r_lw_abs, r_sw_ext, r_sw_scat, r_sw_ascat,
mu
models/atm/cam/src/physics/cam/tphysbc.F90
. add output of tropopause diagnostic calc
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except
004 bl132 TBL.sh e32dh ghgrmp 9s ..................................FAIL! rc= 7
015 bl331 TBL.sh f4dh co2rmp 9s ...................................FAIL! rc= 7
019 bl332 TBL.sh f4dh ghgrmp 9s ...................................FAIL! rc= 7
021 bl333 TBL.sh f4c351pdh aqpgro 3s ..............................FAIL! rc= 7
024 bl334 TBL.sh f4adh adia 9s ....................................FAIL! rc= 7
027 bl335 TBL.sh f4idh idphys 9s ..................................FAIL! rc= 7
030 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7
033 bl337 TBL.sh f4dh fv2d_8tsk 9s ................................FAIL! rc= 7
037 bl371 TBL.sh f1.9dm fvvp_lb2 9s ...............................FAIL! rc= 7
041 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7
044 bl374 TBL.sh fm1.9dh outfrq3s 9s ..............................FAIL! rc= 7
048 bl382 TBL.sh fmgpa1.9dh outfrq3s+1870_bam 9s ..................FAIL! rc= 7
056 bl992 TBL_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 7
bangkok/lf95: All PASS except
021 bl311 TBL.sh f10t5mdm ttrac 9s ................................FAIL! rc= 7
029 bl312 TBL.sh f10dm ghgrmp 9s ..................................FAIL! rc= 7
031 bl313 TBL.sh f10pdm aqpgro 3s .................................FAIL! rc= 7
034 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 7
037 bl315 TBL.sh f10c3dm outfrq3s 9s ..............................FAIL! rc= 7
040 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7
. bl132 fails due to new SST dataset.
. All baselines using FV fail due to roundoff introduced in computing polar
averages at initialization.
. bl315 and bl333 have larger than roundoff differences due to the rhminl
tuning mod for track1, 1.9x2.5 runs.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: Roundoff diffs in all configurations
using FV. Larger than roundoff diffs in FV @ 1.9x2.5 using cam3_5_1
or earlier physics.
===============================================================
===============================================================
Tag name: cam3_6_34
Originator(s): Jim Edwards,mvr
Date: Fri Mar 27 2009
One-line Summary: Extend cam3.5 physics to unstructured grids; bug fix to
test scripts for testing of ccsm and cleanup
Purpose of changes:
. Extend cam3.5 physics to unstructured grids. This involves the code used
to interpolate the prescribed aerosols and ozone datasets. It does not
involve lots of code used by the prognostic chemistry schemes to
interpolate surface emission datasets.
The new interpolation code parallelizes the calculation rather than
interpolating on masterproc and scattering the results.
Bugs fixed (include bugzilla ID):
. The changes made to the interpolation code to support unstructured grids
fixed code bugs related to the treatment of interpolation points at or
near the poles. This introduces larger than roundoff changes in the
interpolated fields.
Describe any changes made to build system:
- moved memory pagesize flags for AIX from testdriver to Makefile
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Brian, mvr, Francis
List all subroutines eliminated:
D models/atm/cam/test/system/TCS.ccsm.sh
D models/atm/cam/test/system/TCB.ccsm.sh
D models/atm/cam/test/system/config_files/f4pdh
- remove test scripts, config file that are no longer used
List all subroutines added and what they do:
A models/atm/cam/bld/namelist_files/use_cases/1850_track1.xml
- new use_case for 1850 runs and track1
A models/atm/cam/test/system/config_files/f4c351pdh
- new configure options test file for testing cam3_5_1 physics
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/tests_pretag_bluefire
M models/atm/cam/test/system/tests_posttag_bluefire
- modified bluefire pretag test list to enable it to finish within 6hr limit
M models/atm/cam/test/system/TCB_ccsm.sh
- minor cleanup to allow reuse of existing test directories
M models/atm/cam/test/system/test_driver.sh
M models/atm/cam/bld/Makefile.in
M models/atm/cam/bld/run-ibm.csh
- moved memory pagesize flags for AIX from testdriver to Makefile
M models/atm/cam/test/system/input_tests_master
- modified definition of pergro test to use cam3_5_1 physics
M models/atm/cam/test/system/TBL_ccsm.sh
- bug fix to ccsm baseline test - was not using baseline code!
M models/atm/cam/src/control/interpolate_data.F90
- added support for direct interpolation from a regular grid to a phys
grid decomposition.
M models/atm/cam/src/utils/repro_sum_mod.F90
- added support for pole averaging fv on the phys grid
M models/atm/cam/src/chemistry/utils/tracer_data.F90
- replaced mo_regrid calls with calls to interpolate_data, this parallelizes
the interpolation and extends it to unstructured grids.
M models/atm/cam/src/chemistry/mozart/tracer_cnst.F90
M models/atm/cam/src/chemistry/mozart/mo_chm_diags.F90
M models/atm/cam/src/chemistry/mozart/mo_aurora.F90
M models/atm/cam/src/chemistry/mozart/mo_airplane.F90
M models/atm/cam/src/chemistry/mozart/mo_apex.F90
M models/atm/cam/src/chemistry/mozart/mo_drydep.F90
- replaced references to commap with calls to dyn_grid
M models/atm/cam/src/dynamics/sld/dyn_grid.F90
M models/atm/cam/src/dynamics/eul/dyn_grid.F90
M models/atm/cam/src/dynamics/fv/dyn_grid.F90
- extended functionality of get_horiz_grid_d to return either regular
grid or unstructured grid output depending on the size of the array(s)
to be returned
M models/atm/cam/src/dynamics/homme/dycore.F90
- added 'UNSTRUCTURED' as a valid string in dycore_is
M .
M SVN_EXTERNAL_DIRECTORIES
- updated to new mct, scripts external tags
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All pass except
004 bl132 TBL.sh e32dh ghgrmp 9s ..................................FAIL! rc= 7
006 bl133 TBL.sh e32pdh aqpgro 3s .................................FAIL! rc= 7
011 bl151 TBL.sh e64h outfrq24h 2d ................................FAIL! rc= 7
015 bl331 TBL.sh f4dh co2rmp 9s ...................................FAIL! rc= 7
019 bl332 TBL.sh f4dh ghgrmp 9s ...................................FAIL! rc= 7
021 bl333 TBL.sh f4c351pdh aqpgro 3s ..............................FAIL! rc= 7
030 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7
033 bl337 TBL.sh f4dh fv2d_8tsk 9s ................................FAIL! rc= 7
037 bl371 TBL.sh f1.9dm fvvp_lb2 9s ...............................FAIL! rc= 7
041 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7
051 bl381 TBL.sh fma1.9h outfrq3s+1870_prog_aero 9s ...............FAIL! rc= 7
055 bl382 TBL.sh fmgpa1.9dh outfrq3s+1870_bam 9s ..................FAIL! rc= 7
061 bl991 TBL_ccsm.sh f19_f19 F....................................FAIL! rc= 7
064 bl992 TBL_ccsm.sh f19_g15 E....................................FAIL! rc= 7
bangkok/lf95:
004 bl111 TBL.sh e8t5mdm ttrac 9s .................................FAIL! rc= 7
008 bl112 TBL.sh e8dm ghgrmp 9s ...................................FAIL! rc= 7
010 bl113 TBL.sh e8pdm aqpgro 3s ..................................FAIL! rc= 7
014 bl114 TBL.sh e8dm co2rmp 9s ...................................FAIL! rc= 7
021 bl311 TBL.sh f10t5mdm ttrac 9s ................................FAIL! rc= 7
029 bl312 TBL.sh f10dm ghgrmp 9s ..................................FAIL! rc= 7
031 bl313 TBL.sh f10pdm aqpgro 3s .................................FAIL! rc= 7
034 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 7
040 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7
all baselines were expected to fail except those using:
-ideal physics
-adiabatic
-cam3_5 physics
-homme
-trop mozart chemistry
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers:
- what code configurations: see above
- what platforms/compilers: all
- nature of change (roundoff; larger than roundoff but same climate; new
climate): Larger than roundoff but same climate.
The new interpolation scheme improves handling of grid points close to the
poles, this change results in a larger than roundoff difference.
If bitwise differences were observed, how did you show they were no worse
than roundoff? 20yr validatin run vs cam3_6_33
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository): cs_cam3_6_33
- platform/compilers: bluefire
- configure commandline:
configure -ntasks 16 -nthreads 4 -phys cam3_5_1 -dyn fv -res 1.9x2.5
- build-namelist command (or complete namelist):
build-namelist -s -case cs_cam3_6_33_F -runtype continue -namelist "&camexp stop_option='nmonths', stop_n=30 /"
- MSS location of output:
mss:/MVR/csm/cs_cam3_6_33_F
MSS location of control simulations used to validate new climate:
mss:/MVR/csm/cam3_6_33_F
URL for AMWG diagnostics output used to validate new climate:
mss:/MVR/csm/cs_cam3_6_33_F/cs_cam3_6_33_F-cam3_6_33_F.tar
===============================================================
Tag name: cam3_6_33
Originator(s): eaton
Date: Fri Mar 13 17:00:02 MDT 2009
One-line Summary: add cam3_5_1 option & tuning; change solar constant; change cice default
Purpose of changes:
. Add a new configure option, "-phys cam3_5_1". This is to configure the
"track1" simulations out of the box. cam3_5_1 is the same build as
cam3_5, but the namelist settings are modified to set the critical Froude
number to 1.0 and to set the FV polar filter to use FFT always (these are
the same values that are used by default in the latest development code).
. Tuning parameter for track1: set rhminl=0.90 for 2 degree runs using RK
microphysics.
. Change to solar constant. Notes from Caspar Ammann:
On Wed, Mar 11, 2009 at 04:44:54PM -0600, Caspar Ammann wrote:
>
> the numbers for mean total solar irradiance (TSI, or scon) for the
> 1850-PreIndustrial as well as PresentDay controls are now computed based
> on the latest information from Judith Lean and her irradiance model. They
> are consistent with SORCE:
>
> PreIndustrial: 1360.89 W/m2 (~1850 AD)
> PresentDay: 1361.27 W/m2 (~1990 but also 2000 AD)
>
> FYI:
> the 1850 mean was computed over 3 roughly stable solar cycles 1834-1866 AD
> the 1990/2000 values were computed over 3 roughly stable solar cycles
> 1975-2007 AD
>
> The cycles were assumed to be 11-year cycles.
>
> The spectral range included is: integral of irradiance at wavelengths
> >200nm
. Turn off the aerosol tracers in CICE:
The CAM "standalone" configuration uses the "prescribed" mode of CICE.
The recent addition of aerosol tracers into CICE is causing problems in
the prescribed mode, so we have decided as a short term fix to turn off
the aerosol tracers in CICE. It is an open science question whether the
aerosol tracers should be turned on in the CICE prescribed mode since
this mode allows only vertical transport of the tracers.
. Update scripts external to scripts4_090312
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system:
. Changes to configure: add valid option "-phys cam3_5_1".
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets:
. added default for 1/4 degree soil_erod dataset
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
SVN_EXTERNAL_DIRECTORIES
. Update scripts external to scripts4_090312
models/atm/cam/bld/cam.cpl7.template
. Add $CAM_NAMELIST_OPTS to the build-namelist command.
models/atm/cam/bld/config_files/definition.xml
. add cam3_5_1 as valid value for phys.
models/atm/cam/bld/configure
. add cam3_5_1 as valid value for phys. The build is identical to cam3_5.
. call the cice configure command with the argument "-ntr_aero 0" to turn
off the aerosol tracer code.
. add check of exit status from the cice configure.
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. change present day value of scon to 1.36127e6
models/atm/cam/bld/namelist_files/use_cases/1870_bam.xml
models/atm/cam/bld/namelist_files/use_cases/1870_control.xml
models/atm/cam/bld/namelist_files/use_cases/1870_prog_aero.xml
. change 1870 value of scon to 1.36089e6
models/atm/cam/src/control/atm_comp_mct.F90
models/atm/cam/src/control/cam_comp.F90
. remove of old gensom code
models/atm/cam/src/physics/cam/cloud_fraction
. set rhminl=.90 for 2 degree w/ RK microphysics
models/atm/cam/src/physics/cam/zm_conv.F90
. Change the dpdry arg from optional to required. dpdry is always
available and changing this allows us to remove a compiler workaround for
the bluegene machines.
models/atm/cam/src/physics/cam/zm_conv_intr.F90
. remove BGP ifdef. always pass dpdry to convtran.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except
004 bl132 TBL.sh e32dh ghgrmp 9s ..................................FAIL! rc= 7
006 bl133 TBL.sh e32pdh aqpgro 3s .................................FAIL! rc= 7
011 bl151 TBL.sh e64h outfrq24h 2d ................................FAIL! rc= 7
015 bl331 TBL.sh f4dh co2rmp 9s ...................................FAIL! rc= 7
019 bl332 TBL.sh f4dh ghgrmp 9s ...................................FAIL! rc= 7
021 bl333 TBL.sh f4pdh aqpgro 3s ..................................FAIL! rc= 7
030 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7
033 bl337 TBL.sh f4dh fv2d_8tsk 9s ................................FAIL! rc= 7
037 bl371 TBL.sh f1.9dm fvvp_lb2 9s ...............................FAIL! rc= 7
041 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7
044 bl374 TBL.sh fm1.9dh outfrq3s 9s ..............................FAIL! rc= 7
047 bl370 TBL.sh fmo1.9dh off1.9x2.5 9s ...........................FAIL! rc= 7
051 bl381 TBL.sh fma1.9h outfrq3s+1870_prog_aero 9s ...............FAIL! rc= 7
055 bl382 TBL.sh fmgpa1.9dh outfrq3s+1870_bam 9s ..................FAIL! rc= 7
058 bl731 TBL.sh h16x4dm aqua 9s ..................................FAIL! rc= 7
061 bl991 TBL_ccsm.sh f19_f19 F 2d ................................FAIL! rc= 4
064 bl992 TBL_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 4
bangkok/lf95: All PASS except
004 bl111 TBL.sh e8t5mdm ttrac 9s .................................FAIL! rc= 7
008 bl112 TBL.sh e8dm ghgrmp 9s ...................................FAIL! rc= 7
010 bl113 TBL.sh e8pdm aqpgro 3s ..................................FAIL! rc= 7
014 bl114 TBL.sh e8dm co2rmp 9s ...................................FAIL! rc= 7
021 bl311 TBL.sh f10t5mdm ttrac 9s ................................FAIL! rc= 7
029 bl312 TBL.sh f10dm ghgrmp 9s ..................................FAIL! rc= 7
031 bl313 TBL.sh f10pdm aqpgro 3s .................................FAIL! rc= 7
034 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 7
037 bl315 TBL.sh f10c3dm outfrq3s 9s ..............................FAIL! rc= 7
040 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7
The baseline failures are expected as summarized below.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: new climate in all configs except
adiabatic and ideal physics due to changing the solar constant. In
addition the tuning parameter change for track1 affects 2 degree FV runs
using the RK microphysics (cam3, cam3_5, and cam3_5_1 physics options).
I suspect that turning off the aerosol tracers in CICE will also change
answers although that change wasn't tested in isolation.
===============================================================
===============================================================
Tag name: cam3_6_32
Originator(s): eaton
Date: Mon Mar 2 12:55:20 MST 2009
One-line Summary: mods to ramp_co2; update externals
Purpose of changes:
. The option to ramp CO2 at a specified annual rate currently required the
user to specify the start date via the namelist variable
ramp_co2_start_ymd. This has been modified so that if this variable is
not specified, the default behavior is to start the ramping at the
beginning of the run.
. Update to the following externals:
drvseq3_0_09
clm3_6_19
cice4_0_20090226
scripts4_090226
** Note ** Updating the CLM and CICE externals results in a new climate
for any configuration that provides deposition of carbon/dust to the
surface (currently trop_mam3, trop_bam, trop_mozart). The new CLM and
CICE versions will use deposition fluxes from the atm whenever they are
provided.
Bugs fixed (include bugzilla ID):
. bug fix from Sungsu in the MG microphyics:
Add the term mnucct+msacwi to the calculation of tlat.
These terms do not appear in the cam3_6 version of MG microphysics, so
this fix will not affect the "track2b" simulations. It does change
answers in the version of MG microphysics currently on CAM's trunk which
came from the end of the modal branch (modal12_cam3_6_26).
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: not checked
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
SVN_EXTERNAL_DIRECTORIES
. update to include drvseq3_0_09, clm3_6_19, cice4_0_20090226, scripts4_090226
models/atm/cam/bld/build-namelist
. restore untruncated names of the following variables to the fincl2 list
used for the camiop test: afixcldliq, afixcldice,dqfxcldliq,dqfxnumliq,
dqfxnumice,afixnumice,afixnumliq,dqfxcldice (bug fix for SCAM).
models/atm/cam/src/dynamics/eul/iop.F90
. change declaration of lowername from char*8 to char*16 (bug fix for SCAM).
models/atm/cam/src/physics/cam/chem_surfvals.F90
. modify how ramp_co2_start_ymd works. Rather than failing if the user
doesn't set it, let the default be to start the co2 ramping at the
beginning of the run.
models/atm/cam/src/physics/cam/cldwat2m.F90
. bug fix from sungsu, line 2879, add the terms mnucct(k)+msacwi(k) to tlat calc.
((bergs(k)+psacws(k)+mnuccc(k)+mnucct(k)+msacwi(k))*cldm(i,k)+(mnuccr(k)+ &
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except
004 bl132 TBL.sh e32dh ghgrmp 9s ..................................FAIL! rc= 7
006 bl133 TBL.sh e32pdh aqpgro 3s .................................FAIL! rc= 7
011 bl151 TBL.sh e64h outfrq24h 2d ................................FAIL! rc= 7
015 bl331 TBL.sh f4dh co2rmp 9s ...................................FAIL! rc= 7
019 bl332 TBL.sh f4dh ghgrmp 9s ...................................FAIL! rc= 7
021 bl333 TBL.sh f4pdh aqpgro 3s ..................................FAIL! rc= 7
030 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7
033 bl337 TBL.sh f4dh fv2d_8tsk 9s ................................FAIL! rc= 7
037 bl371 TBL.sh f1.9dm fvvp_lb2 9s ...............................FAIL! rc= 7
041 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7
044 bl374 TBL.sh fm1.9dh outfrq3s 9s ..............................FAIL! rc= 7
047 bl370 TBL.sh fmo1.9dh off1.9x2.5 9s ...........................FAIL! rc= 7
051 bl381 TBL.sh fma1.9h outfrq3s+1870_prog_aero 9s ...............FAIL! rc= 7
055 bl382 TBL.sh fmgpa1.9dh outfrq3s+1870_bam 9s ..................FAIL! rc= 7
061 bl991 TBL_ccsm.sh f19_f19 F 2d ................................FAIL! rc= 7
064 bl992 TBL_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 7
065 sc001 TSC.sh e64bfbiop scm_prep scm64bfbiop scm_b4b_o1 7s .....FAIL! rc= 6
bangkok/lf95: All PASS except
004 bl111 TBL.sh e8t5mdm ttrac 9s .................................FAIL! rc= 7
008 bl112 TBL.sh e8dm ghgrmp 9s ...................................FAIL! rc= 7
010 bl113 TBL.sh e8pdm aqpgro 3s ..................................FAIL! rc= 7
014 bl114 TBL.sh e8dm co2rmp 9s ...................................FAIL! rc= 7
021 bl311 TBL.sh f10t5mdm ttrac 9s ................................FAIL! rc= 7
029 bl312 TBL.sh f10dm ghgrmp 9s ..................................FAIL! rc= 7
031 bl313 TBL.sh f10pdm aqpgro 3s .................................FAIL! rc= 7
034 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 7
040 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7
The baseline failures are all expected (see below)
The SCAM failure is pre-existing.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: New climate for configurations that
include the default MG microphysics (bug fix) or prognostic aerosols (due
to the new CICE and CLM externals responding to deposition fluxes.
BFB for ideal and adiabatic physics, and for cam3.
===============================================================
===============================================================
Tag name: cam3_6_31
Originator(s): eaton
Date: Fri Feb 27 13:45:55 MST 2009
One-line Summary: option to use "cam3_6" physics; add deposition fluxes for
: prognostic BAM
** N.B. ** The current "out of the box" configuration matches that of the
end of the modal branch, modal12_cam3_6_26, plus the bug fix mentioned
below. This is not a tuned configuration, and is not one of the configs
being evaluated as the possible cam4. The candidate cam4 configurations
require setting various arguments to the configure command. There is
also a large amount of candidate code still on a branch which will be
merged to the trunk in the near future.
Purpose of changes:
. Add capability to run with "cam3_6" physics, defined as cam3_5 + a
version of MG microphysics prior to adding the new treatment for ice +
prognostic BAM. cam3_6 retains the camrt radiation package, the HB
vertical diffusion code, the Hack Shallow convection, and the ZM deep
convection w/ Neale/Richter modifications. This option is configured by
setting the argument "-phys cam3_6" to configure. To run with prescribed
instead of prognostic BAM add the argument "-chem none" to configure.
. Send the carbon and dust deposition fluxes to the coupler when prognostic
BAM is active.
Bugs fixed (include bugzilla ID): in stratiform.F90
. Fix bug in stratiform.F90 (remove conditional surrounding the statement
that sets the tendency of cloud ice number due to detrainment of Nc).
Describe any changes made to build system:
. Changes to configure: add valid option "-phys cam3_6".
Describe any changes made to the namelist:
. Move cloud_fraction namelist variables into their own namelist group.
Change names: freeze_dry -> cldfrc_freeze_dry
rhminl_31 -> cldfrc_rhminl_31
Add cldfrc_ice to turn on/off the calculation of ice cloud fraction.
. Add variable atm_dep_flux to turn on sending deposition fluxes to the
coupler. This is set by build-namelist when the chemistry option is one
that provides the desired fluxes.
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: not checked
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do:
models/atm/cam/src/physics/cam3_6/cldwat2m.F90
. This is the version of cldwat2m being used for cam3_6. It came
originally from modal03_cam3_6_24 and has been modified for the "track2"
runs. This is temporary. The physics that has been backed out of
cldwat2m needs to be implemented with conditional logic in
physics/cam/cldwat2m.F90.
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/build-namelist
. add defaults for cldfrc_freeze_dry and cldfrc_ice
. add default for atm_dep_flux
models/atm/cam/bld/config_files/defaults_waccm_ghg.xml
models/atm/cam/bld/config_files/defaults_waccm_mozart.xml
. remove setting phys=waccm
models/atm/cam/bld/config_files/definition.xml
. add cam3_6 as valid value for phys
. remove waccm as a valid value for phys. This removes an unnecessary
dependence between the phys and chem parameters.
. remove setting the default value for phys. configure needs to determine
the physics package before the configure object is initialized.
models/atm/cam/bld/configure
. add cam3_6 option to -phys
. Move the determination of the physics package ahead of determining the
chemistry package in configure. The physics packages have different
default chemistry packages, so the physics package needs to be set first.
Also remove waccm as a valid physics package value. This created an
unnecessary dependence between the chemistry and physics package settings
that was causing problems as more options are added. configure can do
whatever needs to be done for waccm by checking the value of the
chemistry package.
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. remove the phys="waccm" attributes. In the gw_drag_file element the attribute
was not needed. In the solar_parms_file element it was replaced by
duplicating the element and inserting chem="waccm_ghg" in one and
chem="waccm_mozart" in the other.
. add defaults for cldfrc_freeze_dry and cldfrc_ice.
models/atm/cam/bld/namelist_files/namelist_definition.xml
. move cloud_fraction namelist variables into namelist group cldfrc_nl
. rename freeze_dry -> cldfrc_freeze_dry
rhminl_31 -> cldfrc_rhminl_31
. add cldfrc_ice to switch whether the ice cloud fraction calculation is done
. add atm_dep_flux to indicate when CAM will send deposition fluxes to the
coupler.
models/atm/cam/src/chemistry/mozart/chemistry.F90
. chem_timestep_tend - change intent of cam_out arg from in to inout.
. copy drydep for CB1 to cam_out%bcphodry
copy drydep for CB2 to cam_out%bcphidry
copy drydep for OC1 to cam_out%ocphodry
copy drydep for OC2 to cam_out%ocphidry
models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
. add intent(out) arg drydepflx to gas_phase_chemdr, and set it.
models/atm/cam/src/chemistry/mozart/mz_aerosols_intr.F90
. copy wetdep for OC2 to cam_out%ocphiwet
copy wetdep for CB2 to cam_out%bcphiwet
models/atm/cam/src/control/camsrfexch_types.F90
. use phys_control module to query whether atm will send deposition fluxes
to the coupler.
models/atm/cam/src/control/runtime_opts.F90
. remove cloud fraction namelist vars from cam_inparm
. add call to cldfrc_readnl
models/atm/cam/src/physics/cam/aerosol_intr.F90
. add cam_out to calling args for dust_wet_intr and dust_drydep_intr
models/atm/cam/src/physics/cam/aer_rad_props.F90
. add AEROD_V to default history output
models/atm/cam/src/physics/cam/cloud_fraction.F90
. add method to read namelist cldfrc_nl
. use new namelist variable cldfrc_ice to control whether ice cloud
fraction is computed, not the microphysics type
models/atm/cam/src/physics/cam/dust_intr.F90
. copy wetdep for 4 dust bins to cam_out%dstwet[1-4]
. copy drydep for 4 dust bins to cam_out%dstdry[1-4]
models/atm/cam/src/physics/cam/phys_control.F90
. add namelist variable atm_dep_flux. access via the phys_getopts method.
models/atm/cam/src/physics/cam/stratiform.F90
. remove conditional around setting the tendency of cloud ice number due to
detrainment of Nc.
. misc cleanup
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
004 bl132 TBL.sh e32dh ghgrmp 9s ..................................FAIL! rc= 7
006 bl133 TBL.sh e32pdh aqpgro 3s .................................FAIL! rc= 7
011 bl151 TBL.sh e64h outfrq24h 2d ................................FAIL! rc= 7
015 bl331 TBL.sh f4dh co2rmp 9s ...................................FAIL! rc= 7
019 bl332 TBL.sh f4dh ghgrmp 9s ...................................FAIL! rc= 7
021 bl333 TBL.sh f4pdh aqpgro 3s ..................................FAIL! rc= 7
030 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7
033 bl337 TBL.sh f4dh fv2d_8tsk 9s ................................FAIL! rc= 7
037 bl371 TBL.sh f1.9dm fvvp_lb2 9s ...............................FAIL! rc= 7
041 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7
044 bl374 TBL.sh fm1.9dh outfrq3s 9s ..............................FAIL! rc= 7
047 bl370 TBL.sh fmo1.9dh off1.9x2.5 9s ...........................FAIL! rc= 7
051 bl381 TBL.sh fma1.9h outfrq3s+1870_prog_aero 9s ...............FAIL! rc= 7
055 bl382 TBL.sh fmgpa1.9dh outfrq3s+1870_bam 9s ..................FAIL! rc= 7
058 bl731 TBL.sh h16x4dm aqua 9s ..................................FAIL! rc= 7
061 bl991 TBL_ccsm.sh f19_f19 F 2d ................................FAIL! rc= 7
064 bl992 TBL_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 7
065 sc001 TSC.sh e64bfbiop scm_prep scm64bfbiop scm_b4b_o1 7s .....FAIL! rc= 6
bangkok/lf95: all PASS except:
004 bl111 TBL.sh e8t5mdm ttrac 9s .................................FAIL! rc= 7
008 bl112 TBL.sh e8dm ghgrmp 9s ...................................FAIL! rc= 7
010 bl113 TBL.sh e8pdm aqpgro 3s ..................................FAIL! rc= 7
014 bl114 TBL.sh e8dm co2rmp 9s ...................................FAIL! rc= 7
021 bl311 TBL.sh f10t5mdm ttrac 9s ................................FAIL! rc= 7
029 bl312 TBL.sh f10dm ghgrmp 9s ..................................FAIL! rc= 7
031 bl313 TBL.sh f10pdm aqpgro 3s .................................FAIL! rc= 7
034 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 7
037 bl315 TBL.sh f10c3dm outfrq3s 9s ..............................FAIL! rc= 7
040 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7
All baselines fail except adiabatic and ideal physics due to changes in the
cam3 configuration and due to the bug fix in MG microphysics which is the
default. The SCAM test failure is pre-existing (fix coming soon).
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: different climate due to bug fix in MG
microphysics and change in cam3 configuration (returned the freeze_dry
setting to it's cam3 value).
===============================================================
===============================================================
Tag name: cam3_6_30
Originator(s): mvr
Date: Fri Feb 20 2009
One-line Summary: added testing of ccsm from with cam's test driver; updated
run template scripts; new externals for timing and docn7; other small fixes
Purpose of changes:
will now test F (stand-alone cam default, prescribed ocn/ice) and E (cam mode
with fully active ice and som ocean) configurations during required pretag testing
via the ccsm scripts...
the run template scripts provided in the ~bld directory had not been updated since
cice became the default sea ice model, requiring changes to the configure calls...
latest timing tag eliminated the need for an extra env var in cppdefs in order to
limit the log output of the timing code...however, a change to the performance test
script was required to account for unrelated format changes in the output...
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system:
- removed '-DHAVE_GETTIMEOFDAY' from cppdefs - no longer needed with new timing code
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
D models/atm/cam/test/system/TSB.ccsm.sh
D models/atm/cam/test/system/TCT.ccsm.sh
- outdated scripts for testing ccsm from cam's test driver
List all subroutines added and what they do:
A models/atm/cam/test/system/TCB_ccsm.sh
- new test script to test the configure and build of a ccsm executable
A models/atm/cam/test/system/TEQ_ccsm.sh
- new test script to test the equivalence of a ccsm run vs a cam run
A models/atm/cam/test/system/TER_ccsm.sh
- new test script to test that a ccsm run restarts b4b
A models/atm/cam/test/system/TSM_ccsm.sh
- new test script to test that a ccsm runs to completion (smoke test)
A models/atm/cam/test/system/TBL_ccsm.sh
- new test script to test that a ccsm run is b4b with a run using baseline code
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/tests_pretag_bluefire
M models/atm/cam/test/system/tests_posttag_bluefire
- new ccsm tests will now be part of default/required pretag testing on bluefire
M models/atm/cam/test/system/TPF.sh
- performance test script modified to reflect format changes in timing output
M models/atm/cam/test/system/input_tests_master
- test definitions modified with new ccsm tests;
- bug fix to 1870 test to use proper 'use' case
M models/atm/cam/bld/run-pc.csh
M models/atm/cam/bld/run-ibm.csh
M models/atm/cam/bld/run-lightning.csh
- run template scripts brought up to date to account for recent changes to configure
M models/atm/cam/bld/Makefile.in
- removed '-DHAVE_GETTIMEOFDAY' from cppdefs - no longer needed with new timing code
M SVN_EXTERNAL_DIRECTORIES
M .
- new docn7 tag for som runs to work; new timing tag to eliminate need for
'-DHAVE_GETTIMEOFDAY' in cppdefs to cut down log output
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
061 bl991 TBL_ccsm.sh f19_f19 F 2d ................................FAIL! rc= 5
064 bl992 TBL_ccsm.sh f19_g15 E 2d ................................FAIL! rc= 5
065 sc001 TSC.sh e64bfbiop scm_prep scm64bfbiop scm_b4b_o1 7s .....FAIL! rc= 6
sc001 is a pre-existing failure; bl991 and bl992 failed because these tests did
not exist in the baseline code
bangkok/lf95: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam3_6_29
Originator(s): eaton
Date: Thu Feb 19 11:44:19 MST 2009
One-line Summary: fixes for cam3_5 version; no chem preproc for trop_bam;
: update emissions
Purpose of changes:
. Apply fixes to restore the cam3_5 version (use "-phys cam3_5" option to
configure). This is a combination of adding some control logic in the
source code, and appropriately setting the default values of some
namelist variables which is accomplished via mods to build-namelist. The
changes are:
- cloud_fraction fixed so the new algorithms for ice cloud fraction are
only invoked for the MG microphysics package.
- set the default value of fcrit2 in gw_drag to 0.5 for CAM3 and CAM3_5
physics. The default in the current code is 1.0.
- set fft_flt=0 for CAM3 and CAM3_5 versions. This only applies to the FV
dycore. The current default is fft_flt=1 (which required building with
ALT_PFT defined in the CAM3 and CAM3_5 versions of the code).
Note 1: There are still some differences from the CAM3_5 version of the physics
present in the current code:
- The radiation code now uses a mass based relative humidity (rh) calc
rather than a mole based one. This is for consistency with other calcs
of rh in the moist physics.
- The radiation code assumes that the constituents are dry mixing ratios
rather than moist when computing optical depths.
- the sea salt optics for camrt are now computed using two bins (coarse
and accumulation modes) instead of one. These optics calculations now
depend on the available optics files. We'd need some code modification
along with a new optics file to enable a 1-bin version of sea salt
optics used by cam3_5.
Note 2: The above changes also apply to the cam3 physics. But cam3
physics is still not working due to the Neale/Richter convection mods not
being optional. This is still on the to do list.
. Remove the need to run the chemistry preprocessor for prognostic BAM
(trop_bam).
. Update srf emission and external forcing files for both BAM and MAM to use
the AEROCOM files for year 2000 (trop_mam3 was already using these files)
. Update CICE to cice4_0_20090211
. Update the 1870_prog_aero use case to use the aerocom emissions.
Update the 1870_control use case to remove the trop_mam3 attribute from
the emission datasets. These datasets are used by either BAM or MAM
prognostic aerosols.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist:
. have build-namelist set the default values of the variables in the
phys_control module, i.e., deep_scheme, eddy_scheme, shallow_scheme,
srf_flux_avg, and microp_scheme, so that CAM version specific defaults
will be set. build-namelist also made responsible for setting fcrit2 and
fft_flt.
List any changes to the defaults for the boundary datasets:
. use aerocom 2000 files for BAM and MAM
atm/cam/chem/trop_mozart_aero/emis/aerocom_DMS_2000.c080417.nc
atm/cam/chem/trop_mozart_aero/emis/aerocom_CB1_2000.c080807.nc
atm/cam/chem/trop_mozart_aero/emis/aerocom_OC1_2000.nosoa.c080807.nc
atm/cam/chem/trop_mozart_aero/emis/aerocom_SOAG_2000.c080926.nc
atm/cam/chem/trop_mozart_aero/emis/aerocom_SO2_surface_2000.c080807.nc
atm/cam/chem/trop_mozart_aero/emis/aerocom_SO4_surface_2000.c080807.nc
atm/cam/chem/trop_mozart_aero/emis/aerocom_SO2_vertical_2000.c080807.nc
atm/cam/chem/trop_mozart_aero/emis/aerocom_SO4_vertical_2000.c080807.nc
Describe any substantial timing or memory changes: not checked
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do:
Files from chemistry preprocessor added to src/chemistry/trop_bam/
A chem_mods.F90
A m_het_id.F90
A mo_adjrxt.F90
A mo_imp_sol.F90
A mo_indprd.F90
A mo_lin_matrix.F90
A mo_lu_factor.F90
A mo_lu_solve.F90
A mo_nln_matrix.F90
A mo_phtadj.F90
A mo_prod_loss.F90
A mo_setrxt.F90
A mo_sim_dat.F90
A m_rxt_id.F90
A m_spc_id.F90
List all existing files that have been modified, and describe the changes:
SVN_EXTERNAL_DIRECTORIES
. update CICE to cice4_0_20090211
models/atm/cam/bld/build-namelist
. add defaults for deep_scheme, eddy_scheme, shallow_scheme, srf_flux_avg
. add default for fcrit2 (non-WACCM runs only)
. add default for fft_flt
. fix bug in sub add_default. Needed to check for a defined return value
from get_default_value rather than a true or false return value. The
valid return value "0" was causing the check to fail.
models/atm/cam/bld/configure
. add section analogous to the one for trop_mam3 in the section that
invokes the chem preprocessor.
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. add default values for deep_scheme, eddy_scheme, shallow_scheme,
srf_flux_avg
. add default values for fcrit2 (depends of cam version)
. add default values for fft_flt (depends on cam version)
. use aerocom 2000 files for BAM and MAM
models/atm/cam/bld/namelist_files/namelist_definition.xml
. move the phys_control namelist variables into group phys_ctl_nl
models/atm/cam/bld/namelist_files/use_cases/1870_control.xml
models/atm/cam/bld/namelist_files/use_cases/1870_prog_aero.xml
. Update the 1870_prog_aero use case to use the aerocom emissions.
Update the 1870_control use case to remove the trop_mam3 attribute from
the emission datasets. These datasets are used by either BAM or MAM
prognostic aerosols. Currently these use cases are identical.
models/atm/cam/src/control/runtime_opts.F90
. replace phys_{default,set}opts methods by phys_ctl_readnl. phys_control
module now reads it's own namelist.
. remove unused indforce variable
models/atm/cam/src/physics/cam/cloud_fraction.F90
. add logic to only do new cloud ice fraction calculation for MG
microphysics and to return the cloud fraction for cam3_5 physics
overwise.
models/atm/cam/src/physics/cam/gw_drag.F90
. remove setting default value of fcrit2
models/atm/cam/src/physics/cam/phys_control.F90
. replace the phys_{default,set}opts methods with phys_ctl_readnl so this
module reads it's own namelist.
. Since the default physics packages depend on what physics version we want
to run, allow build-namelist to set the defaults rather than hardcoding
them. Remove the hardcoding of initial values in the module variables.
models/atm/cam/src/physics/cam/radiation.F90
. Remove radforce, indforce, and associated diagnostics. This isn't
supported by the new aerosol-radiation interfaces.
models/atm/cam/src/physics/rrtmg/radiation.F90
. Remove radforce and indforce module variables. This isn't
supported by the new aerosol-radiation interfaces.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
051 bl381 TBL.sh fma1.9h outfrq3s+1870_prog_aero 9s ...............FAIL! rc= 7
058 bl731 TBL.sh h16x4dm aqua 9s ..................................FAIL! rc= 7
059 sc001 TSC.sh e64bfbiop scm_prep scm64bfbiop scm_b4b_o1 7s .....FAIL! rc= 6
bl381 fails due to changing the emission datasets in the 1870_prog_aero use case
bl731 fails because it uses cam3 physics which changed (bug fixes)
sc001 is a pre-existing failure
bangkok/lf95: All PASS except:
037 bl315 TBL.sh f10c3dm outfrq3s 9s ..............................FAIL! rc= 7
bl315 fails due to changes in cam3 physics (bug fixes)
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB except for configs using cam3 or
cam3_5 physics, and for prognostic BAM due to updating the emission
datasets (both present day (year 2000) and pre-industrial (year 1750)
emissions).
===============================================================
===============================================================
Tag name: cam3_6_28
Originator(s): eaton, mflanner
Date: Tue Feb 10 09:23:00 MST 2009
One-line Summary: send trop_mam3 aerosol deposition fluxes to the coupler
Purpose of changes:
. Add the aerosol deposition fluxes to CAM's export state. These fields
are copied to the coupler. Currently this is only being done for the
trop_mam3 aerosols. Eventually CAM will always be responsible for
providing these fluxes, even in prescribed aerosol modes.
. Update CICE tag to cice4_0_20090209
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none known
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do:
models/atm/cam/src/chemistry/modal_aero/modal_aero_deposition.F90
. This module contains the subroutines that are responsible for
partitioning the deposition fluxes of the modal aerosol components into
the fluxes passed through the coupler.
List all existing files that have been modified, and describe the changes:
SVN_EXTERNAL_DIRECTORIES
. Update CICE tag to cice4_0_20090209
models/atm/cam/bld/build-namelist
. use cice config cache filepath from config_cache file when issuing CICE
build-namelist command.
models/atm/cam/bld/config_files/definition.xml
. add cice_config_cache_file
models/atm/cam/bld/configure
. write the filepath for the CICE config_cache file to CAM's config_cache
file so that CAM's configure knows where it is when it invokes the CICE
build-namelist.
models/atm/cam/src/chemistry/mozart/mz_aerosols_intr.F90
. add cam_out to arg list of mz_aero_{wet,dry}_intr
. add dry/wet fluxes for each modal constituent to local var
. call new methods, set_srf_drydep and set_srf_wetdep, to partition
contributions from modal constituents into fields passed to coupler.
models/atm/cam/src/control/atm_comp_mct.F90
. copy fluxes from cam export state into coupler buffer
models/atm/cam/src/control/camsrfexch_types.F90
. add aerosol deposition fluxes to cam export type
. initialize values to 0. for trop_mam3, to shr_const_spval otherwise.
models/atm/cam/src/physics/cam/aerosol_intr.F90
. add cam_out to actual args of mz_aero_wet_intr and mz_aero_dry_intr
. add cam_out to dummy args of aerosol_wet_intr and aerosol_drydep_intr
models/atm/cam/src/physics/cam/cam_diagnostics.F90
. add new method, diag_export, to make outfld calls for the deposition
fluxes that are being passed to the coupler
models/atm/cam/src/physics/cam/physpkg.F90
. modify calling args to diag_surf (cleanup)
models/atm/cam/src/physics/cam/restart_physics.F90
. add 8 drydep components of cam_out to both the pio and serial code
sections.
models/atm/cam/src/physics/cam/tphysac.F90
. add cam_out to actual args of aerosol_drydep_intr
models/atm/cam/src/physics/cam/tphysbc.F90
. add cam_out to actual args of aerosol_wet_intr
. call diag_export at end of subroutine. This is the state that is copied
to the coupler.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
059 sc001 TSC.sh e64bfbiop scm_prep scm64bfbiop scm_b4b_o1 7s .....FAIL! rc= 6
This failure is pre-existing.
bangkok/lf95: All PASS.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB.
===============================================================
Tag name: cam3_6_27
Originator(s): eaton,andrew,mvr
Date: 02 Feb 2009
One-line Summary: Make trop_mam3 the default chemistry; integrate MAM with the MG microphysics.
Purpose of changes:
. Make trop_mam3 the default chemistry.
- Don't require running the chemistry preprocessor for predefined chemistry
schemes; currently only trop_mam3 has been reimplemented this way.
- Add tuning to modal aerosols from Xiaohong.
- Add boundary layer nucleation from Xiaohong.
. MG microphysics changes
- add MG microphysics bug fix for diagnostic field used by wet dep codes
- ice microphysics mods; MAM tuning
- hook immersion freezing to modal dust
. Update RRTMG external
- update rrtmg to latest tag, rrtmg_081203, which includes change to sw
surface fluxes needed to balance the surface energy budget.
. changes for regression testing
- modified the offline driver test to run at 1.9x2.5
- change mo_apex.F90 to calculate latitude midpoints inline rather than
relying on a quantity (clat_staggered) that's not available from all the
dycores. This is to get the regression tests working for the spectral
dycores.
- Removed the gensom and som config files from the regression tests.
. new external tags for almost all components to emulate current ccsm beta tags
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
. default chemistry set to trop_mam3
. Don't require running the chemistry preprocessor for predefined chemistry
schemes; currently only trop_mam3 has been reimplemented this way.
. Refactor configure's -chem option. This needed to be done to provide a way
to turn off the default trop_mam3 chemistry. At the same time, the
overloaded use of "-chem none" to imply that the CAM3 prescribed aerosols
and ozone were being used was removed. Now "-chem none" is used to turn
off all prognostic chemistry without implying which prescribed aerosols or
ozone are used. To do that new valid values, cam3 and cam3_5, have been
added to the -phys argument. "-phys cam3" now implies "-chem none" and
that the CAM3 prescribed aerosols and ozone will be used. This is not yet
a completely implemented backwards compatibility switch for CAM3 physics
because the Richter/Neale convection mods are still hardwired. But the
intention is to have the "-phys cam3" option eventually provide the
complete CAM3 physics package. "-phys cam3_5" also implies "-chem none",
but instead of the CAM3 aerosols and ozone, the newer prescribed BAM
aerosols and the ozone dataset from J.F. Lamarque will be used.
In addition to the new configure functionality, quite a bit of new error
checking has been added to try and prevent building the model with
inconsistent physics packages.
. The -gensom and "-ocn som" options were removed from configure. The new
SOM functionality is currently only available via the CCSM scripts.
Describe any changes made to the namelist:
. option added to allow cam's build-namelist to call build-namelist for cice
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
. trop_mam3 is much more expensive than prescribed BAM
. the build will be faster when the chemistry preprocessor isn't needed
Code reviewed by: Brian Eaton
List all subroutines eliminated:
D models/atm/cam/test/system/config_files/f10sdm
D models/atm/cam/test/system/config_files/f4gdh
D models/atm/cam/test/system/config_files/f4sdh
D models/atm/cam/test/system/config_files/e8sdm
D models/atm/cam/test/system/config_files/e32sdh
D models/atm/cam/test/system/config_files/s8sdm
D models/atm/cam/test/system/config_files/e64sm
D models/atm/cam/test/system/config_files/s32sdh
. obsolete test files containing configure options
D models/atm/cam/test/system/tests_posttag_bluevista
. removed file for obsolete machine
List all subroutines added and what they do:
A + models/atm/cam/test/system/config_files/fmo1.9dh
- new test file containing configuration options for offline driver test
A + models/atm/cam/test/system/config_files/s8dm
. replacement test file containing configuration options
A + models/atm/cam/test/system/nl_files/off1.9x2.5
. new test file containing namelist options for offline driver test
A + models/atm/cam/bld/namelist_files/use_cases/1870_bam.xml
. new use case for prescribed BAM aerosols
A + models/atm/cam/src/chemistry/trop_mam3
A + models/atm/cam/src/chemistry/trop_mam3/mo_lu_solve.F90
A + models/atm/cam/src/chemistry/trop_mam3/mo_setrxt.F90
A + models/atm/cam/src/chemistry/trop_mam3/mo_adjrxt.F90
A + models/atm/cam/src/chemistry/trop_mam3/mo_imp_sol.F90
A + models/atm/cam/src/chemistry/trop_mam3/chem_mods.F90
A + models/atm/cam/src/chemistry/trop_mam3/mo_prod_loss.F90
A + models/atm/cam/src/chemistry/trop_mam3/mo_lin_matrix.F90
A + models/atm/cam/src/chemistry/trop_mam3/m_rxt_id.F90
A + models/atm/cam/src/chemistry/trop_mam3/mo_phtadj.F90
A + models/atm/cam/src/chemistry/trop_mam3/mo_nln_matrix.F90
A + models/atm/cam/src/chemistry/trop_mam3/m_het_id.F90
A + models/atm/cam/src/chemistry/trop_mam3/mo_lu_factor.F90
A + models/atm/cam/src/chemistry/trop_mam3/mo_indprd.F90
A + models/atm/cam/src/chemistry/trop_mam3/mo_sim_dat.F90
A + models/atm/cam/src/chemistry/trop_mam3/m_spc_id.F90
. preprocessor output for trop_mam3
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/TCB.sh
. added -cice_decomptype to configure call
M models/atm/cam/test/system/tests_pretag_bluefire
. modified list of default tests for pretag testing of machine bluefire
M models/atm/cam/test/system/config_files/f10c3dm
. use "-phys cam3" in place of "-chem none -microphys rk -rad camrt"
M models/atm/cam/test/system/config_files/h16x4dm
. modified the test file used for configuration options for homme test
M models/atm/cam/test/system/config_files/scm64bfbiop
M models/atm/cam/test/system/config_files/e64bfbiop
M models/atm/cam/test/system/config_files/scmarmiop
- added -chem "none" to configuration options of scam tests as temporary fix
M models/atm/cam/test/system/tests_pretag_bangkok
. modified list of pretag tests to fit in time constraints
M models/atm/cam/test/system/TBR.sh
. set the cice_ic var with the restart filename only (not absolute path)
. modified call to build-namelist for cice changes; cleanup
M models/atm/cam/test/system/test_driver.sh
. added env var to allow specification of batch queue
M models/atm/cam/test/system/TER.sh
. modified call to build-namelist for cice changes
M models/atm/cam/test/system/nl_files/scm_prep
. changed the names in the fincl2 list to match the untruncated names
M models/atm/cam/test/system/input_tests_master
. change the *382 tests to use the new use case 1870_bam.xml. This
preserves the test of prescribed bam aerosols.
. change e32sdh to e32dh
. change e64sm to e64m
. change e8sdm to e8dm
. change f10sdm to f10dm
. change f4gdh to f4dh
. change f4sdh to f4dh
. change s8sdm to s8dm
M models/atm/cam/test/system/TSM.sh
. modified call to build-namelist for cice changes
M models/atm/cam/bld/configure
. make trop_mam3 the default value for chem.
. add conditional logic to not call chem_preprocess when the chem package
is trop_mam3. This requires correctly setting $chem_nadv, $chem_cppdefs,
and the 'chem_src_dir' entry in the configure object.
. fix bug in setting chem_pkg when phys is ideal or adiabatic
. modify how the default chemistry package is set.
. add error checking for inconsistent settings of physics and chemistry
packages.
. add error checking for inconsistent settings of radiation and
microphysics packages. The defaults for both these packages has moved
from the definition file to configure since they depend on the setting of
the physics package.
. remove -gensom and "-ocn som" options. SOM is only supported via the
CCSM scripts.
. mods needed for cice build changes; additions for ocn,lnd,glc stub code
M models/atm/cam/bld/Makefile.in
. -DHAVE_GETTIMEOFDAY added to cppdefs (required to control output of new
timing code)
M models/atm/cam/bld/config_files/definition.xml
. change valid value for phys: cam1 -> cam, and add cam3 and cam3_5.
. mods in connection with stub code additions
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. update filenames of aerocom emissions for 2000
. add dst_0.47x0.63_c081217.nc for 1/2 degree soil_erod default
. update the CLM entries to be as consistent with the CLM version as
possible.
. Add the entries for fsnowoptics, fsnowaging, faerdep.
M models/atm/cam/bld/namelist_files/namelist_definition.xml
. add definitions for fsnowoptics, fsnowaging, faerdep.
M models/atm/cam/bld/namelist_files/use_cases/1870_control.xml
. update for trop_mam3 chemistry
M models/atm/cam/bld/build-namelist
. add defaults for fsnowoptics, fsnowaging, faerdep.
. add co2_ppmv. Take this from the value set for CAM and convert vmr to
ppmv.
. change the way CAM3 physics is specified. Currently uses $chem eq 'none'
change to $phys eq 'cam3'. The cam3_5 option should use the prescribed
BAM aerosols and JF's prescribed ozone.
. Don't try to create CLM or CICE namelists if the physics mode doesn't use
an active land or ice component. Currently that's true for the ideal,
adiabatic, and aqua_planet modes.
. option added to allow cam's build-namelist to call build-namelist for cice
M models/atm/cam/bld/camdom.cpl7.template
M models/atm/cam/bld/cam.cpl7.template
. update .template files for ccsm4_0_beta07 (and later) tags
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
. from rrtmg_081029 to rrtmg_081203
M models/atm/cam/src/control/scamMod.F90
. change declarations for alphanam(pcnst) and dqfxnam(pcnst) from char*8 to
char*16 so the names aren't truncated.
M models/atm/cam/src/control/atm_comp_mct.F90
. compute nextsw_cday at init time for startup runs only
M models/atm/cam/src/physics/cam/cldwat2m.F90
. remove temp vars nevapr1, evapsnow1, prain1, prodsnow1, cmeout1. They
were being zeroed incorrectly inside the iteration where they were
supposed to be accumulating quantities over sub-steps.
. replace DOS with UNIX endlines
. immersion freezing hooks in modal dust
M models/atm/cam/src/physics/cam/stratiform.F90
M models/atm/cam/src/physics/cam/cloud_fraction.F90
M models/atm/cam/src/physics/cam/wv_saturation.F90
. ice microphysics mods from Andrew Gettleman
. immersion freezing hooks in modal dust
M models/atm/cam/src/physics/cam/aerosol_intr.F90
. immersion freezing hooks in modal dust
M models/atm/cam/src/chemistry/modal_aero/modal_aero_coag.F90
M models/atm/cam/src/chemistry/modal_aero/modal_aero_newnuc.F90
. fix declaration of dotend -- was pcnstxx, changed to pcnst.
. mods for boundary layer nucleation from xiaohong
M models/atm/cam/src/physics/cam/tphysac.F90
M models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
M models/atm/cam/src/chemistry/mozart/chemistry.F90
. mods for boundary layer nucleation from xiaohong
M models/atm/cam/src/chemistry/mozart/mo_apex.F90
. replace reference to clat_staggered by inline calculation of latitude
midpoints.
M models/atm/cam/src/chemistry/mozart/mo_drydep.F90
. In subroutine interp_map change veg_ext from 10 to 20.
M models/atm/cam/src/chemistry/mozart/mz_aerosols_intr.F90
. Add tuning to modal aerosols from Xiaohong.
M SVN_EXTERNAL_DIRECTORIES
. new tags for almost all components to emulate current ccsm beta tags
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
004 bl132 TBL.sh e32dh ghgrmp 9s ..................................FAIL! rc= 7
006 bl133 TBL.sh e32pdh aqpgro 3s .................................FAIL! rc= 7
011 bl151 TBL.sh e64h outfrq24h 2d ................................FAIL! rc= 7
015 bl331 TBL.sh f4dh co2rmp 9s ...................................FAIL! rc= 7
019 bl332 TBL.sh f4dh ghgrmp 9s ...................................FAIL! rc= 7
021 bl333 TBL.sh f4pdh aqpgro 3s ..................................FAIL! rc= 7
030 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7
033 bl337 TBL.sh f4dh fv2d_8tsk 9s ................................FAIL! rc= 7
037 bl371 TBL.sh f1.9dm fvvp_lb2 9s ...............................FAIL! rc= 7
041 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7
044 bl374 TBL.sh fm1.9dh outfrq3s 9s ..............................FAIL! rc= 7
047 bl370 TBL.sh fmo1.9dh off1.9x2.5 9s ...........................FAIL! rc= 5
051 bl381 TBL.sh fma1.9h outfrq3s+1870_prog_aero 9s ...............FAIL! rc= 7
055 bl382 TBL.sh fmgpa1.9dh outfrq3s+1870_bam 9s ..................FAIL! rc= 5
058 bl731 TBL.sh h16x4dm aqua 9s ..................................FAIL! rc= 7
059 sc001 TSC.sh e64bfbiop scm_prep scm64bfbiop scm_b4b_o1 7s .....FAIL! rc= 4
- all baselines failed except for adiabatic and ideal physics (this was expected)
- scam test has been broken, a fix is in the works
bangkok/lf95:
004 bl111 TBL.sh e8t5mdm ttrac 9s .................................FAIL! rc= 7
008 bl112 TBL.sh e8dm ghgrmp 9s ...................................FAIL! rc= 7
010 bl113 TBL.sh e8pdm aqpgro 3s ..................................FAIL! rc= 7
014 bl114 TBL.sh e8dm co2rmp 9s ...................................FAIL! rc= 7
021 bl311 TBL.sh f10t5mdm ttrac 9s ................................FAIL! rc= 7
029 bl312 TBL.sh f10dm ghgrmp 9s ..................................FAIL! rc= 7
031 bl313 TBL.sh f10pdm aqpgro 3s .................................FAIL! rc= 7
034 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 7
037 bl315 TBL.sh f10c3dm outfrq3s 9s ..............................FAIL! rc= 7
040 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7
- all baselines failed except for adiabatic and ideal physics (this was expected)
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: new climate
. There are answer changes due to all of the following
- Updates in both the MAM and MG microphysics, as well as integrating those
packages to work together.
- New RRTMG tag with modified SW surface fluxes.
- Answer changes from updated CLM and CICE tags.
===============================================================
Tag name: cam3_6_26
Originator(s): Francis Vitt
Date: 24 Dec 2008
One-line Summary: Addition of the LLNL super fast chemistry
Purpose of changes:
Provide a less expensive option for interactive chemistry
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Added configure option:
-chem super_fast_llnl
Describe any changes made to the namelist:
Replaced "photopath" namelist variable with "exo_coldens_file",
"tuv_xsect_file", and "o2_xsect_file".
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Brian Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
A models/atm/cam/bld/config_files/defaults_super_fast_llnl.xml
- configure defaults for the super_fast_llnl chem
A models/atm/cam/src/chemistry/mozart/llnl_O1D_to_2OH_adj.F90
A models/atm/cam/src/chemistry/mozart/llnl_set_chem_trop.F90
- for super_fast_llnl chem
A models/atm/cam/src/chemistry/mozart/lin_strat_chem.F90
A models/atm/cam/src/chemistry/mozart/linoz_data.F90
- for linoz ozone speicification in the stratosphere
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/configure
M models/atm/cam/bld/config_files/definition.xml
- configure -chem super_fast_llnl option added
M models/atm/cam/bld/chem_preprocess.pl
- changed location of the chemistry preprocessor to models/atm/cam/chem_proc
- super_fast_llnl option added
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- new chem IC files added
- defaults for super_fast_llnl chemistry
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- replaced "photopath" namelist variable with
"exo_coldens_file", "tuv_xsect_file", and "o2_xsect_file".
- linoz options added
- tropopause level options added
M models/atm/cam/bld/build-namelist
- changes for super_fast_llnl chem
- replaced "photopath" namelist variable with "exo_coldens_file",
"tuv_xsect_file", and "o2_xsect_file".
M models/atm/cam/src/physics/cam/eddy_diff.F90
- change to appease lf95 compiler
MM models/atm/cam/src/chemistry/utils/tracer_data.F90
- mods to read zonal-averaged LINOZ data
M models/atm/cam/src/chemistry/mozart/mo_aero_settling.F90
- fixed potential array out-of-boounds error
M models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
- changes for super_fast_llnl and linoz
M models/atm/cam/src/chemistry/mozart/mo_seto2.F90
- changed photopath to o2_xsect_file
M models/atm/cam/src/chemistry/mozart/mo_tropopause.F90
- cleanup
M models/atm/cam/src/chemistry/mozart/mo_exp_sol.F90
- added ability to specify top pressure level to apply solver
M models/atm/cam/src/chemistry/mozart/mo_chemini.F90
- chemistry options change
M models/atm/cam/src/chemistry/mozart/mo_synoz.F90
M models/atm/cam/src/chemistry/mozart/mo_fstrat.F90
- SYNOZ bug fix
M models/atm/cam/src/chemistry/mozart/mo_setinv.F90
- change for super_fast_llnl
M models/atm/cam/src/chemistry/mozart/chemistry.F90
- chemistry options change
- changes for super_fast_llnl and linoz
M models/atm/cam/src/chemistry/mozart/mo_tuv_inti.F90
M models/atm/cam/src/chemistry/mozart/mo_photo.F90
- replaced "photopath" namelist variable with "exo_coldens_file",
"tuv_xsect_file", and "o2_xsect_file".
M models/atm/cam/src/chemistry/mozart/efield.F90
- waccm change provided by HAO
M models/atm/cam/src/chemistry/mozart/mo_usrrxt.F90
- changes for super_fast_llnl and linoz
M models/atm/cam/SVN_EXTERNAL_DIRECTORIES
M SVN_EXTERNAL_DIRECTORIES
- changed location of the chemistry preprocessor to models/atm/cam/chem_proc
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
030 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7
A science change in efield.F90 caused this failure
bangkok/lf95: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_6_25
Originator(s): pworley
Date: Wed Dec 18 10:25:45 MST 2008
One-line Summary: generalize and optimize load balancing options
Purpose of changes:
1) Generalize the definition of load balancing options
0, 1, 3, and 4 to be able to use more processes in the physics than in
the dynamics.
Previously there was support for using additional
processes in the physics only when specifying the most aggressive
load balancing (phys_loadbalance==2). The other options have been
generalized to assign work to all physics processes.
For example, phys_loadbalance==0 will now distribute work
to otherwise idle processes, but will (still) not trade work with
other processes active during the dynamics. When the number of
physics and dynamics processes are the same, the behavior of
the load balancing options is unchanged.
2) Introduce new default communication option for physics
load balancing (phys_alltoall = -1) that uses information about
nature of the communication to choose between a point-to-point
algorithm and an MPI collective.
When using more physics processes than dynamics processes, and
when using load balancing options other than phys_loadbalance=2,
most processes send and receive data to/from only a small number of
the other processes. In these situations, a point-to-point implementation
of the data arrangment is often faster than calling MPI_Alltoallv.
This observation is now "implemented" in the code. It can be overridden
by setting phys_alltoall to 0 (to use collective) or 1 (to use
point-to-point algorithm).
3) Performance optimizations in FV (mod_comm and geopk_d)
and dynamics/physics MPI communications (spmd_util).
Eliminated sending/receiving zero length messages in one
of the communication options in both the FV transposes and
in the physics/dynamics transposes. Also added support for
the mpi_isend/mpi_send option (modc_send_cdcore) to the
geopk_d algoritm.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist:
- Changed the range of phys_alltoall to include -1, and made -1 the default.
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes:
- Performance improved at scale, especially when using more physics
processes than dynamics processes
Code reviewed by: Mirin
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/src/physics/cam:
phys_grid.F90 - generalized load balancing options;
changed phys_alltoall default
models/atm/cam/src/utils:
spmd_utils.F90 - eliminated zero-length messages when not using
handshaking protocol
models/utils/pilgrim:
mod_comm.F90 - eliminated zero-length messages when not using
handshaking protocol
models/atm/cam/src/dynamics/fv:
cd_core.F90 - changed parameters in call to geopk_d
geopk.F90 - added modc_send_cdcore option to geopk_d
dynamics_vars.F90 - updated comments
spmd_dyn.F90 - updated comments
dp_coupling.F90 - zeroed out uninitialized array elements being
communicated during load balancing
models/atm/cam/src/dynamics/eul:
spmd_dyn.F90 - initialized work array used in load balancing
models/atm/cam/src/dynamics/sld:
spmd_dyn.F90 - initialized work array used in load balancing
models/atm/cam/bld/namelist_files:
namelist_definition.xml - updated descriptions
models/atm/cam/src/control:
runtime_opts.F90 - updated comments
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: ALL PASS
bangkok/lf95: ALL PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate): NO CHANGES (bfb)
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
Tag name: cam3_6_24
Originator(s): eaton
Date: Fri Dec 12 15:02:11 MST 2008
One-line Summary: remove PIO as part of CAM component when building for CCSM
Purpose of changes:
. Remove PIO as part of CAM component when building for CCSM. This backs
out a hack added in cam3_6_19 for the CCSM build.
*** This change doesn't affect the CAM standalone build and has not been
run through all the usual pretag regression testing.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system:
. As noted above when building for CCSM.
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/configure
. Don't build PIO as part of CAM component for CCSM build.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: not done.
bangkok/lf95: not done.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam3_6_23
Originator(s): eaton, pworley, mvertens
Date: Tue Dec 9 12:35:36 MST 2008
One-line Summary: update interpic_new, scripts tag; misc
Purpose of changes:
. Substantial update to interpic_new code. See tools/interpic_new/README
for details. This includes mods from Mark Taylor to deal with large files.
. Mod to phys_grid.F90 that was left out of the cam3_6_21 commit (from
Pat Worley).
. Mods to run on BGP.
. Cleanup: put config_* files into config_files/*
. update scripts external to scripts4_081207.
. updated cam.cpl7.template file (from Mariana Vertens).
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
models/atm/cam/bld/config_defaults_eul.xml
models/atm/cam/bld/config_defaults_fv.xml
models/atm/cam/bld/config_defaults_homme.xml
models/atm/cam/bld/config_defaults_sld.xml
models/atm/cam/bld/config_defaults_trop_bam.xml
models/atm/cam/bld/config_defaults_trop_ghg.xml
models/atm/cam/bld/config_defaults_trop_mam3.xml
models/atm/cam/bld/config_defaults_trop_mam7.xml
models/atm/cam/bld/config_defaults_trop_mozart.xml
models/atm/cam/bld/config_defaults_waccm_ghg.xml
models/atm/cam/bld/config_defaults_waccm_mozart.xml
models/atm/cam/bld/config_definition.xml
models/atm/cam/bld/config_definition.xsl
models/atm/cam/bld/config_horiz_grid.xml
models/atm/cam/bld/config_sys_defaults.xml
. moved to subdirectory config_files/ and renamed
models/atm/cam/tools/interpic_new/interp_driver.F90
models/atm/cam/tools/interpic_new/varspecs_mod.F90
List all subroutines added and what they do:
models/atm/cam/bld/config_files/defaults_eul.xml
models/atm/cam/bld/config_files/defaults_fv.xml
models/atm/cam/bld/config_files/defaults_homme.xml
models/atm/cam/bld/config_files/defaults_sld.xml
models/atm/cam/bld/config_files/defaults_trop_bam.xml
models/atm/cam/bld/config_files/defaults_trop_ghg.xml
models/atm/cam/bld/config_files/defaults_trop_mam3.xml
models/atm/cam/bld/config_files/defaults_trop_mam7.xml
models/atm/cam/bld/config_files/defaults_trop_mozart.xml
models/atm/cam/bld/config_files/defaults_waccm_ghg.xml
models/atm/cam/bld/config_files/defaults_waccm_mozart.xml
models/atm/cam/bld/config_files/definition.xml
models/atm/cam/bld/config_files/definition.xsl
models/atm/cam/bld/config_files/horiz_grid.xml
models/atm/cam/bld/config_files/sys_defaults.xml
. moved from the bld/ subdirectory and renamed (remove config_ prefix)
models/atm/cam/tools/interpic_new/interp.F90
models/atm/cam/tools/interpic_new/interpolate_data.F90
models/atm/cam/tools/interpic_new/shr_kind_mod.F90
. interpic_new refactoring
List all existing files that have been modified, and describe the changes:
SVN_EXTERNAL_DIRECTORIES
. change scripts external to scripts4_081207.
models/atm/cam/bld/build-namelist
. replace instances of $defaults->get_value by get_default_value. This is
so that the use case setting affects the default values.
models/atm/cam/bld/cam.cpl7.template
. cam's dtime is set based on $ATM_NCPL
. mods to incorporate $CAM_CONFIG_OPTS arg to configure. This replaces the
more specific $CAM_CHEM.
. remove broken CO2 cycle code. Will reimplement as required.
. $CAM_USE_CASE replaced by $CAM_NML_USE_CASE.
models/atm/cam/bld/Makefile.in
. mods to the BGP section
models/atm/cam/bld/configure
. change names of config_*.xml files to config_files/*.xml
models/atm/cam/bld/namelist_files/use_cases/1870-2000_control.xml
. Remove tracer_cnst_file from use case xml file since the default chem is
not using prognostic aerosols. Also remove the scon setting since that
now comes from the general defaults.
models/atm/cam/test/system/test_driver.sh
. increase queue time on bluefire to 4:28 (from 3:28)
models/atm/cam/src/control/cam_history.F90
. remove endrun condition if LOGNAME isn't found in environment
models/atm/cam/src/control/ioFileMod.F90
. remove unused "use shr_sys_mod" and "use shr_file_mod
models/atm/cam/src/control/print_memusage.F90
. add BGP to the ifdef that avoids calling get_memusage
models/atm/cam/src/physics/cam/phys_grid.F90
. replace some mpigather* calls by fc_gather*
models/atm/cam/tools/interpic_new/Makefile
models/atm/cam/tools/interpic_new/README
models/atm/cam/tools/interpic_new/compare_var.F90
models/atm/cam/tools/interpic_new/control.F90
models/atm/cam/tools/interpic_new/cpvar.F90
models/atm/cam/tools/interpic_new/dimensions.F90
models/atm/cam/tools/interpic_new/driver.F90
models/atm/cam/tools/interpic_new/fill_positions.F90
models/atm/cam/tools/interpic_new/fmain.F90
models/atm/cam/tools/interpic_new/handle_special_cases.F90
models/atm/cam/tools/interpic_new/is_special_case.F90
models/atm/cam/tools/interpic_new/wrap_nf.F90
. update interpic_new
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS
bangkok/lf95: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam3_6_22
Originator(s): Xiaohong Liu, Francis Vitt
Date: 5 Dec 2008
One-line Summary: Incorporate the Modal Aerosol Model (MAM)
Purpose of changes:
Provide more aerosol model options. In addition the the bulk
aerosol model, 3-mode and 7-mode modal models have been
incorporated with radiation feed-backs on the climate.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Changes to the configure utility:
-chem options added:
trop_mam3 (3-mode model)
trop_mam7 (7-mode model)
-chem option change:
trop_aero --> trop_bam (bulk aerosol model)
Describe any changes made to the namelist:
New namelist variables:
modal_optics (char*256)
dataset for modal aerosol optics
aer_drydep_list (char*16(1000))
list of modal aerosol species that undergo dry deposition
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Brian Eaton
List all subroutines eliminated:
D models/atm/cam/bld/config_defaults_trop_aero.xml
List all subroutines added and what they do:
A models/atm/cam/bld/config_defaults_trop_mam7.xml
A models/atm/cam/bld/config_defaults_trop_mam3.xml
A models/atm/cam/bld/config_defaults_trop_bam.xml
- added configure options "-chem trop_mam7" and "-chem trop_mam3"
- replaced the "-chem trop_aero" with "-chem trop_bam"
A models/atm/cam/src/physics/cam/miesubs.F
A models/atm/cam/src/physics/cam/modal_aer_opt.F90
A models/atm/cam/src/physics/cam/ndrop.F90
A models/atm/cam/src/chemistry/modal_aero/modal_aero_rename.F90
A models/atm/cam/src/chemistry/modal_aero/modal_aero_coag.F90
A models/atm/cam/src/chemistry/modal_aero/modal_aero_data.F90
A models/atm/cam/src/chemistry/modal_aero/modal_aero_gasaerexch.F90
A models/atm/cam/src/chemistry/modal_aero/modal_aero_initialize_data.F90
A models/atm/cam/src/chemistry/modal_aero/modal_aero_wateruptake.F90
A models/atm/cam/src/chemistry/modal_aero/modal_aero_calcsize.F90
A models/atm/cam/src/chemistry/modal_aero/modal_aero_newnuc.F90
- files added for MAM
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/config_files/fma1.9h
M models/atm/cam/test/system/config_files/fma1.9m
- change "-chem trop_aero" to "-chem trop_bam"
M models/atm/cam/bld/configure
M models/atm/cam/bld/chem_preprocess.pl
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
M models/atm/cam/bld/namelist_files/namelist_definition.xml
M models/atm/cam/bld/build-namelist
M models/atm/cam/bld/config_definition.xml
- implemented configure and build-namelist changes for MAM
M models/atm/cam/src/control/runtime_opts.F90
M models/atm/cam/src/control/wrap_nf.F90
M models/atm/cam/src/control/atm_comp_mct.F90
M models/atm/cam/src/control/filenames.F90
M models/atm/cam/src/physics/cam/progseasalts_intr.F90
M models/atm/cam/src/physics/cam/radiation.F90
M models/atm/cam/src/physics/cam/dust_intr.F90
M models/atm/cam/src/physics/cam/tphysbc.F90
M models/atm/cam/src/physics/cam/cldwat2m.F90
M models/atm/cam/src/physics/cam/aer_rad_props.F90
M models/atm/cam/src/physics/cam/radsw.F90
M models/atm/cam/src/physics/cam/phys_prop.F90
M models/atm/cam/src/physics/cam/stratiform.F90
M models/atm/cam/src/physics/cam/radlw.F90
M models/atm/cam/src/physics/cam/rad_constituents.F90
M models/atm/cam/src/physics/cam/wetdep.F90
M models/atm/cam/src/physics/cam/cloud_fraction.F90
M models/atm/cam/src/physics/cam/carbon_intr.F90
M models/atm/cam/src/physics/cam/sulchem.F90
M models/atm/cam/src/physics/cam/wv_saturation.F90
M models/atm/cam/src/physics/cam/aerosol_intr.F90
M models/atm/cam/src/physics/cam/tphysac.F90
M models/atm/cam/src/physics/cam/qneg3.F90
M models/atm/cam/src/physics/cam/vertical_diffusion.F90
M models/atm/cam/src/physics/rrtmg/radlw.F90
M models/atm/cam/src/physics/rrtmg/radiation.F90
M models/atm/cam/src/physics/rrtmg/radsw.F90
M models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
M models/atm/cam/src/chemistry/mozart/mo_chm_diags.F90
M models/atm/cam/src/chemistry/mozart/mo_extfrc.F90
M models/atm/cam/src/chemistry/mozart/chemistry.F90
M models/atm/cam/src/chemistry/mozart/mo_drydep.F90
M models/atm/cam/src/chemistry/mozart/mz_aerosols_intr.F90
M models/atm/cam/src/chemistry/mozart/mo_usrrxt.F90
M models/atm/cam/src/chemistry/mozart/mo_srf_emissions.F90
M models/atm/cam/src/chemistry/mozart/mo_setsox.F90
- source code files changed to implement MAM
M models/atm/cam/src/chemistry/mozart/mo_photo.F90
M models/atm/cam/src/chemistry/mozart/mo_photoin.F90
- changes needed to use fast-tuv photolyis in MAM
M models/atm/cam/src/chemistry/mozart/mo_jlong.F90
M models/atm/cam/src/chemistry/mozart/mo_chemini.F90
- reduced the memory overhead for the table-lookup photolysis method
M SVN_EXTERNAL_DIRECTORIES
- new chem_proc tag (chemistry preprocessor)
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
030 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7
044 bl374 TBL.sh fm1.9dh outfrq3s 9s ..............................FAIL! rc= 7
047 bl355 TBL.sh fmo2dh off2x2.5 9s ...............................FAIL! rc= 7
These baseline failures are expected. These are caused by changes in
the photolysis routines and input datasets.
bangkok/lf95: All Pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_6_21
Originator(s): Art Mirin, Pat Worley
Date:
One-line Summary: Improvements to communication, other misc. improvements
Purpose of changes: Improve communication performance at scale.
Changes are bit-for-bit.
MPI performance in CAM for runs with large numbers of processes has proven
to be erratic in certain circumstances, sometimes even leading to failures
due to exhaustion of system buffer space. Aspects of this behavior have been
observed on three different platforms (Atlas at LLNL, Cray XT, and IBM BG/P),
and is likely to occur with some process count on all systems.
Through empirical experiments, we identified a number of communication protocols
that were implemented in the swapm routine in the spmd_utils module that eliminate
these problems. However, no one protocol was best in all situations, so a number
of protocols were imported into the mod_comm irregular communication routines
(which are used for intra-FV-dynamics transposes, gathers and scatters,
and geopotential communications) and the choice of what to use can be
specified at runtime. Reasonable defaults were determined, however.
As part of this exercise, we also cleaned up the swapm routine,
removing the options that are never exercised, so the logic in mod_comm and
swapm are now very similar.
(Fun fact: swapm communication protocols were originally developed for PSTSWM
and were used in CCM/MP2D.)
The ALTGATHER logic, used to avoid problems on the Cray XT when too many "unexpected"
messages arrive at the root of the gather, is now supported within mod_comm as one
of the options, and is enabled by default. Similarly, new routines with robust
behavior but without any performance degradation were made the defaults for I/O-related
gathers for the spectral dycores.
The new communication options, implemented in the mod_comm mp_sendirr
and mp_recvirr irregular communication routines, include:
(1) option to handshake; that is, the source will not post its send until the
corresponding receive has already been posted; this mitigates against unnecessary
contention and buffering; in this case, ready send is used instead of send.
(2) option to use blocking rather than nonblocking send.
(3) option to limit the number of outstanding message requests; this mitigates against
exhausting buffer space and performance degradation when the number of sources and
destinations for a process during a collective operation becomes large.
(4) option to call MPI_Alltoallv. (Note, this is typically not the best choice within
the FV dycore on Atlas or on the Cray XT, but it usually is best on the IBM BG/P.)
(5) option to perform certain intra-FV-dynamics transposes in pairs.
(6) option to perform intra-FV-dynamics tracer transposes N at a time; irregular transpose
structures were replaced with 4-D arrays; also, the partial real*4 tracer support was
removed.
We also changed ordering of target tasks to dimensional-exchange ordering, as used in swapm,
rather than 1 to N; this mitigates against contention.
Changes were also made to the logic for error checking with the fast reproducible sum
calculation; testing and warning message are now run-time options, off by default; in
particular, the collective sums are computed once only unless the user requests error
checking.
Bugs fixed (include bugzilla ID): Temporary 1-line conditionally compilable patches
were added to six pio-related files, as workaround for ifort compiler bug.
Describe any changes made to build system:
Describe any changes made to the namelist: The following were added:
modc_sw_dynrun - mod_comm irregular underlying communication method for dyn_run/misc
modc_hs_dynrun - mod_comm irregular communication handshaking for dyn_run/misc (T/F)
modc_send_dynrun - mod_comm irregular communication send protocol for dyn_run/misc (T/F)
modc_mxreq_dynrun - mod_comm irregular communication nonblocking request throttle for dyn_run/misc
modc_sw_cdcore - mod_comm irregular underlying communication method for cd_core/geopk
modc_hs_cdcore - mod_comm irregular communication handshaking for cd_core/geopk (T/F)
modc_send_cdcore - mod_comm irregular communication send protocol for cd_core/geopk (T/F)
modc_mxreq_cdcore - mod_comm irregular communication nonblocking request throttle for cd_core/geopk
modc_sw_gather - mod_comm irregular underlying communication method for gather
modc_hs_gather - mod_comm irregular communication handshaking for gather (T/F)
modc_send_gather - mod_comm irregular communication send protocol for gather (T/F)
modc_mxreq_gather - mod_comm irregular communication nonblocking request throttle for gather
modc_sw_scatter - mod_comm irregular underlying communication method for scatter
modc_hs_scatter - mod_comm irregular communication handshaking for scatter (T/F)
modc_send_scatter - mod_comm irregular communication send protocol for scatter (T/F)
modc_mxreq_scatter - mod_comm irregular communication nonblocking request throttle for scatter
modc_sw_tracer - mod_comm irregular underlying communication method for multiple tracers
modc_hs_tracer - mod_comm irregular communication handshaking for multiple tracers (T/F)
modc_send_tracer - mod_comm irregular communication send protocol for multiple tracers (T/F)
modc_mxreq_tracer - mod_comm irregular communication nonblocking request throttle for multiple tracers
modc_onetwo - one or two simultaneous mod_comm irregular communications (excl. tracers)
modc_tracers - max number of tracers for simultaneous mod_comm irregular communications;
if 0, then use mp_sendirr/mp_recvirr to do one at a time
fc_gather_flow_cntl - tuning option for flow-controlled gathers used by spectral dycores.
Default should be sufficient, but if performance is suffering can try a
different level of message request throttling.
List any changes to the defaults for the boundary datasets: NONE
Describe any substantial timing or memory changes: Certain configurations on
certain platforms run a lot faster at scale.
Code reviewed by: Eaton
List all subroutines eliminated:
utils/pilgrim/mod_comm.F90:
mp_swapmirr, mp_sendirr_fc, mp_sendirr_fc_r4, mp_sendirr_fc_i4
List all subroutines added and what they do:
utils/pilgrim/mod_comm.F90:
mp_sendtrirr,
mp_recvtrirr - versions of mp_sendirr/mp_recvirr for use with tracer advection
mp_swapirr,
mp_swapirr_i4,
mp_swaptrirr - implements new flow-controlled and MPI_collective options, called
within mp_sendirr, mp_sendirr_i4, and mp_sendtrirr, respectively
atm/cam/src/utils/spmd_utils.F90:
fc_gatherv,
fc_gathervr4,
fc_gathervint - flow-controlled gather algorithms, based on ALTGATHER code
formerly in control/wrap_mpi.F90 in the routines mpigatherv,
mpigathervr4, and mpigathervint, respectively.
fc_gather_setopts - set fc_gather_flow_cntl, the maximum number of preposted
receive requests allowed in the flow-controlled gather algorithm.
fc_gather_defaultopts - return default value for fc_gather_flow_cntl (64)
atm/cam/src/utils/repro_sum_mod.F90:
repro_sum_tol_exceeded - reports whether difference between fast reproducible
collective sum and fast nonreproducible collective sum
exceeds the specified tolerance.
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/namelist_files:
namelist_definition.xml - registered new tuning parameters
models/atm/cam/src/chemistry/utils:
tracer_data.F90 - included pio patch for ifort on atlas
models/atm/cam/src/control:
binary_io.F90 - referenced renamed gather routines now included in spmd_utils.F90
cam_history.F90 - referenced renamed gather routines now included in spmd_utils.F90
hycoef.F90 - included pio patch for ifort on atlas
interpolate_data.F90 - fixed syntax
runtime_opts.F90 - included new tuning parameters
wrap_mpi.F90 - eliminated gather and altalltoallv routines (now relocated in
spmd_utils.F90). wrap_mpi.F90 is now just a wrapper for MPI calls.
models/atm/cam/src/dynamics/eul:
realloc4.F90 - added reference to altalltoallv in spmd_utils module
realloc7.F90 - added reference to altalltoallv in spmd_utils module
scan2.F90 - added reference to altalltoallv in spmd_utils module
spmd_dyn.F90 - included FV-relevant performance options (for spmd_dyn.F90 consistency)
models/atm/cam/src/dynamics/fv:
advect_tend.F90 - adopted new tracer array structure
benergy.F90 - adopted new tracer array structure
cd_core.F90 - generalized interface to mod_comm, provided option to call multiple
simultaneous transposes
d2a3dikj.F90 - incorporated summation modifications
diag_dynvar_ic.F90 - adopted new tracer array structure
diag_module.F90 - generalized interface to mod_comm
dp_coupling.F90 - adopted new tracer array structure, generalized interface to mod_comm,
fixed syntax
dryairm.F90 - adopted new tracer array structure
dyn_comp.F90 - adopted new tracer array structure, generalized interface to mod_comm,
provided option to call multiple simultaneous transposes
dynamics_vars.F90 - added performance parameters to FV datatype, adopted new tracer array
structure
fv_prints.F90 - incorporated summation modifications
geopk.F90 - generalized interface to mod_comm
inidat.F90 - adopted new tracer array structure, generalized interface to mod_comm
io_dist.F90 - generalized interface to mod_comm
mapz_module.F90 - adopted new tracer array structure
mean_module.F90 - incorporated summation modifications
p_d_adjust.F90 - adopted new tracer array structure, incorporated summation modifications
par_xsum.F90 - incorporated summation modifications
pfixer.F90 - generalized interface to mod_comm
restart_dynamics.F90 - adopted new tracer array structure, eliminated real*4 tracer option,
included pio patch for ifort on atlas
spmd_dyn.F90 - included new tuning parameters
stepon.F90 - adopted new tracer array structure
te_map.F90 - adopted new tracer array structure
trac2d.F90 - adopted new tracer array structure
models/atm/cam/src/dynamics/homme:
spmd_dyn.F90 - included FV-relevant performance options (for spmd_dyn.F90 consistency)
models/atm/cam/src/dynamics/sld:
realloc4.F90 - added reference to altalltoallv in spmd_utils module
realloc7.F90 - added reference to altalltoallv in spmd_utils module
scan2.F90 - added reference to altalltoallv in spmd_utils module
spmd_dyn.F90 - included FV-relevant performance options (for spmd_dyn.F90 consistency)
models/atm/cam/src/physics/cam:
phys_buffer.F90 - included pio patch for ifort on atlas
phys_gmean.F90 - incorporated summation modifications
phys_grid.F90 - generalized interface to mod_comm
restart_physics.F90 - included pio patch for ifort on atlas
models/atm/cam/src/utils:
repro_sum_mod.F90 - incorporated summation modifications
spmd_utils.F90 - incorporated gather and altalltoallv routines from wrap_mpi.F90;
introduced new flow-controlled gather routines;
modified swap software
time_manager.F90 - included pio patch for ifort on atlas
models/utils/pilgrim:
debugutilitiesmodule.F90 - trivial change
decompmodule.F90 - trivial change
mod_comm.F90 - incorporated major communication options
parutilitiesmodule.F90 - generalized interface to mod_comm
pilgrim.F90 - trivial change
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: ALL PASS
bangkok/lf95: ALL PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_6_20
Originator(s): eaton
Date: Wed Nov 26 14:18:10 MST 2008
One-line Summary: Make RRTMG default; update default radiative forcings;
misc fixes
Purpose of changes:
. Make RRTMG the default radiation package.
. Update the namelist defaults so that the "out of the box" configuration
uses the same forcings as the CCSM B_PRESENT_DAY cases. This includes
the solar constant, prescribed ozone, aerosols, and GHGs.
. misc fixes and compiler workarounds
- Fix new logic in getfil to make PGI happy.
- Fix PIO namelist variable definition (iotype_name)
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist:
. default values of forcings changed as described above
List any changes to the defaults for the boundary datasets:
. default values of forcings changed as described above
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/config_definition.xml
. set default value for "rad" to "rrtmg"
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. update defaults from 1990 to 2000
models/atm/cam/bld/namelist_files/namelist_definition.xml
. fix pio namelist variable iotype_name
models/atm/cam/bld/namelist_files/use_cases/1870_control.xml
. remove the solar constant. now build-namelist will use the value in
namelist_defaults_cam.xml.
models/atm/cam/src/chemistry/mozart/tgcm_forcing.F90
. change dimension of dimids from 3 to 4 in tgcm_init
models/atm/cam/src/control/ioFileMod.F90
. fix getfil logic (test on optional arg iflag) to make PGI happy
models/atm/cam/src/physics/rrtmg/radiation.F90
. add :ncol bounds to array expression to make lf95/debug happy
models/atm/cam/test/system/config_files/f10c3dm
. add -rad camrt to get at cam3.5 configuration
models/atm/cam/test/system/nl_files/ghgrmp
. remove "doisccp=.true."
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
004 bl132 TBL.sh e32sdh ghgrmp 9s .................................FAIL! rc= 7
006 bl133 TBL.sh e32pdh aqpgro 3s .................................FAIL! rc= 7
011 bl151 TBL.sh e64h outfrq24h 2d ................................FAIL! rc= 7
015 bl331 TBL.sh f4gdh co2rmp 9s ..................................FAIL! rc= 7
019 bl332 TBL.sh f4sdh ghgrmp 9s ..................................FAIL! rc= 7
021 bl333 TBL.sh f4pdh aqpgro 3s ..................................FAIL! rc= 7
030 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7
033 bl337 TBL.sh f4dh fv2d_8tsk 9s ................................FAIL! rc= 7
037 bl371 TBL.sh f1.9dm fvvp_lb2 9s ...............................FAIL! rc= 7
041 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7
044 bl374 TBL.sh fm1.9dh outfrq3s 9s ..............................FAIL! rc= 7
047 bl355 TBL.sh fmo2dh off2x2.5 9s ...............................FAIL! rc= 7
051 bl381 TBL.sh fma1.9h outfrq3s+1870_prog_aero 9s ...............FAIL! rc= 7
055 bl382 TBL.sh fmgpa1.9dh outfrq3s+1870_control 9s ..............FAIL! rc= 7
058 bl731 TBL.sh h16x4dm aqua 9s ..................................FAIL! rc= 7
bangkok/lf95: All PASS except:
004 bl111 TBL.sh e8t5mdm ttrac 9s .................................FAIL! rc= 7
008 bl112 TBL.sh e8sdm ghgrmp 9s ..................................FAIL! rc= 7
010 bl113 TBL.sh e8pdm aqpgro 3s ..................................FAIL! rc= 7
014 bl114 TBL.sh e8dm co2rmp 9s ...................................FAIL! rc= 7
019 bl153 TBL.sh e64m outfrq24h 2d ................................FAIL! rc= 7
023 bl311 TBL.sh f10t5mdm ttrac 9s ................................FAIL! rc= 7
031 bl312 TBL.sh f10sdm ghgrmp 9s .................................FAIL! rc= 7
033 bl313 TBL.sh f10pdm aqpgro 3s .................................FAIL! rc= 7
036 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 7
039 bl315 TBL.sh f10c3dm outfrq3s 9s ..............................FAIL! rc= 7
042 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7
045 bl377 TBL.sh f1.9m outfrq24h 2d ...............................FAIL! rc= 7
049 bl511 TBL.sh s8t5mdm ttrac 9s .................................FAIL! rc= 7
053 bl512 TBL.sh s8sdm ghgrmp 9s ..................................FAIL! rc= 7
055 bl513 TBL.sh s8pdm aqpgro 3s ..................................FAIL! rc= 7
057 bl553 TBL.sh s64m outfrq24h 2d ................................FAIL! rc= 7
These baseline comparisons are expected to fail due to making RRTMG the
default radiation package, and due to new default radiative forcings
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: New climate due to changing the default
radiation and the default forcings.
===============================================================
===============================================================
Tag name: cam3_6_19
Originator(s): eaton
Date: Mon Nov 17 18:51:53 MST 2008
One-line Summary: CCSM/PIO build; BGP mods; getfil; extend build-namelist use_case
Purpose of changes:
. Modify CAM's configure so that when building the CAM component via the
CCSM build scripts, PIO is built as part of CAM. This is a temporary
hack!
. Add standalone CAM build support for BGP.
. Modify getfil so that the default is to abort when a requested file isn't
found. This makes it easier to detect this common error.
. Extend the functionality of build-namelist so that use case
specifications can contain variables that are not namelist variables just
like the namelist_defaults_cam.xml file does. This is needed to support
building complex namelist values like those used by the chemistry code to
specify the emission datasets for multiple constituents.
Bugs fixed (include bugzilla ID):
. fix initialization problem in FV offline mode that was causing sm355 to
fail.
Describe any changes made to build system:
. modified as described above for CCSM builds.
. support for BGP
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/build-namelist
. Check the element names in the use case defaults file and if the name
isn't in the definition file then don't add the name to the namelist.
. The namelist defaults object corresponding to the use case needs to be
saved, and searched for default values before the base defaults file is
searched. Encapsulate this in a new routine, get_default_values.
models/atm/cam/bld/configure
. add utils/pio to Filepath produced for ccsm build.
. add -D_NETCDF macro to the $cfg_cppdefs for a ccsm build.
. add -D_USEBOX to $cfg_cppdefs
models/atm/cam/bld/Makefile.in
. remove -D_USEBOX
models/atm/cam/bld/cam.cpl7.template
. remove old unicos code
models/atm/cam/bld/config_sys_defaults.xml
. add bgp support (default is pure MPI)
models/atm/cam/bld/namelist_files/use_cases/1870_control.xml
. remove tracer_cnst_file specifier which is inconsistent with the default
prescribed BAM aerosols. This was causing the failure of sm386.
models/atm/cam/src/control/iofilemod.F90
. change default behavior of getfil so that it aborts when requested file
can't be found.
models/atm/cam/src/dynamics/fv/inidat.F90
. initialize dyn_in%u3s and dyn_in%v3s to zero for FV offline mode
models/atm/cam/src/physics/waccm/gw_drag.F90
. use getfil before trying to open a file
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
047 bl355 TBL.sh fmo2dh off2x2.5 9s ...............................FAIL! rc= 5
This baseline is failing because the smoke test sm355 fails for cam3_6_18.
bangkok/lf95: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam3_6_18
Originator(s): aconley, eaton
Date: Mon Nov 3 18:57:32 MST 2008
One-line Summary: RRTMG fixes and new cloud diagnostics
Purpose of changes:
. Update RRTMG external to
- Fixes bug in restart (some wrong intents)
- Fixes problem with minimal solar zenith angle for WACCM.
. RRTMG diagnostics
- Cleanup of addfld calls
- Output aerosol masses as seen by radiation
. Cloud diagnostics moved into a new module, cloud_diagnostics. This
module is currently only implemented for RRTMG.
. Add capability to compute optical properties for volcanic aerosols.
. Extend the rad_constituents interface to allow setting gas species mixing
ratios to zero. This is done by using the specifier 'Z' in the
appropriate namelist variable in place of 'P' or 'D'.
. Bugfix in new ISCCP simulator.
. Modify how radiation package is specified to configure. The -rrtmg
switch has been replaced by the argument -rad which takes either rrtmg or
camrt as values. The current default is camrt.
. Update test_driver.sh to use pgi-7.2-5 on bangkok and calgary.
Bugs fixed (include bugzilla ID):
. fix an index problem in the new ISCCP simulator
. change char*80 to 256 in filename declaration in mo_flbc.F90
Describe any changes made to build system:
. remove -rrtmg switch from configure. replaced by -rad argument which
takes either rrtmg or camrt as values. The current default is camrt.
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
models/atm/cam/src/physics/rrtmg/param_cldoptics.F90
. not used -- much of this moved to cloud_diagnostics
List all subroutines added and what they do:
models/atm/cam/src/physics/cam/cloud_diagnostics.F90
. stub function for so that cam can separate cloud physical diagnostics
from radiation diagnostics
models/atm/cam/src/physics/rrtmg/cloud_diagnostics.F90
. write cloud physical diagnostics (non-radiative diagnostics) to history
file.
List all existing files that have been modified, and describe the changes:
models/atm/cam/SVN_EXTERNAL_DIRECTORIES
. update RRTMG external to rrtmg_081029
models/atm/cam/bld/build-namelist
. fixes related to changing configuration parameter from rrtmg to rad
. add check for doisccp. use it to set doisccp_38 if rrtmg is rad pkg
models/atm/cam/bld/config_definition.xml
. remove rrtmg and add generic rad
models/atm/cam/bld/configure
. remove the -rrtmg flag and add a more full featured -rad arg
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. change rrtmg attribute of physprop files
models/atm/cam/bld/namelist_files/use_cases/1870-2000_control.xml
models/atm/cam/bld/namelist_files/use_cases/1870_control.xml
models/atm/cam/bld/namelist_files/use_cases/1870_prog_aero.xml
. add tracer_cnst_file setting so the use case doesn't break when the
defaults in namelist_defaults_cam.xml change.
models/atm/cam/bld/perl5lib/Build/Namelist.pm
. fix code that limits the length of strings written to namelist output
files
models/atm/cam/src/chemistry/mozart/mo_flbc.F90
. change declaration of filename from char*80 to char*256
models/atm/cam/src/physics/cam/aer_rad_props.F90
. add radiation hooks for volcanic aerosol.
. fix indexing for output of SW aerosol OD's.
. update names in code to match names in aerosol files (lw_ext renamed to lw_abs)
models/atm/cam/src/physics/cam/icarus_scops_38.F90
. fix indexing bug
models/atm/cam/src/physics/cam/phys_prop.F90
. add functionality for volcanics
. update names in code to match names in aerosol files (lw_ext renamed to lw_abs)
models/atm/cam/src/physics/cam/physpkg.F90
. add call for cloud_diagnostics_init
. remove call for param_cldoptics_init
models/atm/cam/src/physics/cam/rad_constituents.F90
. specify masses that are not included in diagnostic radiation calls as
having zero mass path.
. add new props for volcanics to rad_cnst_get_clim_aer_props()
models/atm/cam/src/physics/cam/radae.F90
. make code robust for zero ozone concentrations (from Philip Cameron-Smith
and Dan Bergmann)
models/atm/cam/src/physics/cam/radiation.F90
. move param_cldoptics initialization into camrt radiation.F90, since it is
now a radiation utility
models/atm/cam/src/physics/cam/tphysbc.F90
. add call to cloud_diagnostics
models/atm/cam/src/physics/rrtmg/cloud_rad_props.F90
. clean up diagnostic history tape calls
. remove unused pointers to CLD in physics buffer.
models/atm/cam/src/physics/rrtmg/radiation.F90
. remove obsolete "addfld" calls
. fix intent of fsds. Was intent(out), but should be intent(inout) since
it is only set on radiation timesteps.
. add calls to output aerosol masses as seen by radiation
. remove old isccp simulator -- can replace if we decide we really want to
run this with rrtmg.
models/atm/cam/src/physics/rrtmg/radsw.F90
. change solvar_interface names for waccm
models/atm/cam/test/system/config_files/h16x4dm
models/atm/cam/test/system/config_files/h5x8adm
models/atm/cam/test/system/config_files/h5x8dm
models/atm/cam/test/system/config_files/h5x8idm
. add -hgrid values
models/atm/cam/test/system/CAM_runcmnd.sh
. remove hpmcount from bluefire (left in by mistake on previous commit)
models/atm/cam/test/system/test_driver.sh
. update to pgi-7.2-5 on bangkok and calgary
. return default queue on bluefire to regular
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
045 sm355 TSM.sh fmo2dh off2x2.5 9s ...............................FAIL! rc= 8
046 er355 TER.sh fmo2dh off2x2.5 4+5s .............................FAIL! rc= 5
047 bl355 TBL.sh fmo2dh off2x2.5 9s ...............................FAIL! rc= 4
The smoke test for these tests fails in cam3_6_17.
bangkok/lf95: All PASS except:
019 bl153 TBL.sh e64m outfrq24h 2d ................................FAIL! rc= 7
045 bl377 TBL.sh f1.9m outfrq24h 2d ...............................FAIL! rc= 7
057 bl553 TBL.sh s64m outfrq24h 2d ................................FAIL! rc= 7
There is a roundoff difference in these results. Not sure why. Validated
with a pergro test (need to set -microphys rk).
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB, except roundoff for non-debug
production configurations under lf95.
===============================================================
===============================================================
Tag name: cam3_6_17
Originator(s): Jim Edwards
Date:
One-line Summary: Refactor restart files from binary to netcdf using PIO
Purpose of changes: Improved memory profile allow parallel IO for restart
in NetCDF format
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: Added external PIO library
Describe any changes made to the namelist: Added pio_ctl namelist
default allowed values
io_stride 1 1 - n
iotype_name 'netcdf' 'netcdf', 'pnetcdf'
pio_restart .false.
num_iotasks n 1 - n
where n is the total number of compute tasks
io_stride and num_iotasks control the number of tasks participating in IO, If only io_stride is set
num_iotasks will be adjusted automatically.
pio_restart can be used to turn off the pio facility and use the old binary restart.
If you find that you need to set this please let me know so that I can address the problem, we plan to make PIO
the only option in an upcoming release.
iotype_name allows a choice of netcdf or pnetcdf libraries - currently netcdf appears to perform better but
that may change depending on platform, decomposition and problem size.
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: self
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
models/atm/cam/test/system/TSB.ccsm.sh
models/atm/cam/test/system/TCT.ccsm.sh
models/atm/cam/test/system/TBL.sh
models/atm/cam/test/system/TBR.sh
models/atm/cam/test/system/test_driver.sh
models/atm/cam/test/system/CAM_runcmnd.sh
models/atm/cam/test/system/TEQ.sh
removed bluevista, added firefly. updated files_to_compare so that only history files are listed.
models/atm/cam/test/system/nl_files/ghgrmp
models/atm/cam/test/system/nl_files/off2x2.5
added pio_restart=.true. so that pio is tested with all dycores as part of precommit tests.
models/atm/cam/test/system/nl_files/aqua
clean up of test namelist
models/atm/cam/test/system/input_tests_master
added a homme ideal phys test
models/atm/cam/bld/configure
models/atm/cam/bld/namelist_files/namelist_definition.xml
models/atm/cam/bld/Makefile.in
added pio support
models/atm/cam/src/control/rgrid.F90
modified initialization so that default values are used instead of invalid values.
models/atm/cam/src/physics/cam/physpkg.F90
models/atm/cam/src/physics/cam/physics_types.F90
added initialization of phys_state and tend data structures for improved error checking
models/atm/cam/src/physics/cam/convect_shallow.F90
changed cush declaration from a 3D variable (pverp) to a 2D.
models/atm/cam/src/physics/cam/tphysac.F90
added homme to dycores which need moist physics
models/atm/cam/src/dynamics/eul/dyn_grid.F90
new values in get_dyn_grid_parm for compatability
models/atm/cam/src/dynamics/homme/external/prim_state_mod.F90
models/atm/cam/src/dynamics/homme/external/edge_mod.F90
models/atm/cam/src/dynamics/homme/dp_coupling.F90
models/atm/cam/src/dynamics/homme/inidat.F90
models/atm/cam/src/dynamics/homme/dyn_comp.F90
models/atm/cam/src/dynamics/homme/restart_dynamics.F90
models/atm/cam/src/dynamics/homme/inital.F90
refactored model flow to better match other dycores.
models/atm/cam/src/control/cam_restart.F90
models/atm/cam/src/control/hycoef.F90
models/atm/cam/src/control/cam_comp.F90
models/atm/cam/src/utils/pio_utils.F90
models/atm/cam/src/utils/time_manager.F90
models/atm/cam/src/physics/cam/restart_physics.F90
models/atm/cam/src/physics/cam/chemistry.F90
models/atm/cam/src/physics/cam/buffer.F90
models/atm/cam/src/physics/cam/phys_buffer.F90
models/atm/cam/src/chemistry/utils/prescribed_ghg.F90
models/atm/cam/src/chemistry/utils/prescribed_aero.F90
models/atm/cam/src/chemistry/utils/tracer_data.F90
models/atm/cam/src/chemistry/utils/prescribed_ozone.F90
models/atm/cam/src/chemistry/mozart/tracer_cnst.F90
models/atm/cam/src/chemistry/mozart/tracer_srcs.F90
models/atm/cam/src/chemistry/mozart/chemistry.F90
models/atm/cam/src/dynamics/sld/restart_dynamics.F90
models/atm/cam/src/dynamics/eul/restart_dynamics.F90
models/atm/cam/src/dynamics/fv/restart_dynamics.F90
models/atm/cam/src/dynamics/fv/metdata.F90
added support for PIO restart method
SVN_EXTERNAL_DIRECTORIES
updated PIO tag
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
All tests were done twice, once with each restart format.
bluefire: All pass except
058 bl731 TBL.sh h16x4dm aqua 9s ..................................FAIL
Fails due to continuing dycore development.
bangkok/lf95: All tests pass.
CAM tag used for the baseline comparison tests if different than previous
tag:
===============================================================
===============================================================
Tag name: cam3_6_16
Originator(s): Francis Vitt
Date:
One-line Summary: Combine CAM-Chem and WACCM chemistry codes.
Purpose of changes: Ease of chemistry code mantainance and
provide more flexability in the aerosol and gas phase chemistry
configurations.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Changes to configure:
- new configure options :
-prog_species [SO4[,DST[,SSLT[,OC[,BC[,GHG[,CARBON16]]]]]]]
Configure will generate a preprocessor input file
for any combination of these predefined prognostics
aerosol and ghg packages.
-edit_chem_mech
This allows the user to edit the mechanism file before
it is processed. Default editor is vi. The user can
specify another editor via the CAMCHEM_EDITOR environment
variable.
-usr_mech_infile $mechanism_file
This allows the user to specify a preprocessor input file.
- configure invokes the chemstry preprocessor to generate chemistry
solver code
- predefined chemistry options:
-chem trop_mozart | trop_ghg | trop_aero | waccm_mozart | waccm_ghg
- when configured with the option -chem trop_mozart | waccm_mozart
the preprocessor will use a predefined "standard" preprocessor input file
- when configured with the option -chem trop_ghg | trop_aero | waccm_ghg
or with the -prog_species SO4 | DST | SSLT | OC | BC | GHG | CARBON16
a preprocessor input file is created
New features:
- The ability to specify short-lived chemical tracers which will
not be transported. These species are specified via a "ShortLived"
species designation in the preprocessor input file.
- The ability to skip solving the chemical equations some time
steps. The chemical equation are solved each time step by
default. This can be changed via the chem_freq namelist variable.
- The prescribed mozart ozone and aerosols for the radiation can now be used
with any chemistry configuration (and without any chemistry).
Describe any changes made to the namelist:
Chemistry options are now read from "chem_inparm" namelist in atm_in.
Pass "chem_rad_passive = .true." to build-namelist to set the rad_climate variable
so that the chemical tracers are radiatively passive.
emis_date --> srf_emis_date
emis_type --> srf_emis_type
WACCM
now uses srf_emis_specifier as with CAM-chem
lbc_file --> flbc_file
lbc_fixed_date --> flbc_date
lbc_senario --> flbc_type
sad_fixed_date --> sad_date
sad_scenario --> sad_type
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Brian Eaton
List all subroutines eliminated:
D models/atm/cam/src/chemistry/trop_mozart
D models/atm/cam/src/chemistry/trop_mozart_aero
D models/atm/cam/src/chemistry/waccm_mozart
D models/atm/cam/src/chemistry/trop_mozart_prescribed_aero
D models/atm/cam/src/chemistry/trop_mozart_ghg_paero
The source code in these directories are combined. The preprocessor
generated code is now generated by configure via the the chemistry
preprocessor.
D models/atm/cam/bld/config_defaults_trop_mozart_aero.xml
D models/atm/cam/bld/config_defaults_trop_mozart_prescribed_aero.xml
D models/atm/cam/bld/config_defaults_trop_mozart_ghg_paero.xml
The -chem options trop_mozart_aero, trop_mozart_prescribed_aero, and
trop_mozart_ghg_paero were removed.
D models/atm/cam/src/physics/waccm/iondrag.F90
D models/atm/cam/src/physics/waccm/chemistry.F90
D models/atm/cam/src/physics/waccm/mo_solar_parms.F90
D models/atm/cam/src/physics/waccm/mo_msis_ubc.F90
D models/atm/cam/src/physics/waccm/upper_bc.F90
D models/atm/cam/src/physics/waccm/msise00.F90
D models/atm/cam/src/physics/waccm/tgcm_forcing.F90
These were moved to the shared chemistry directory.
List all subroutines added and what they do:
A models/atm/cam/bld/config_defaults_trop_aero.xml
A models/atm/cam/bld/config_defaults_trop_ghg.xml
Files added for the new -chem options "trop_aero" and "trop_ghg".
A models/atm/cam/bld/chem_preprocess.pl
Perl file added to invoke the chemistry preprocessor.
A models/atm/cam/bld/namelist_files/use_cases/waccm_refb1.xml
A models/atm/cam/bld/namelist_files/use_cases/1990_prog_aero.xml
WACCM use cases added for the build-namelist util.
A models/atm/cam/src/physics/waccm/gravity_waves_sources.F90
new WACCM gravity wave drag code.
A models/atm/cam/src/chemistry/utils/mo_constants.F90
A models/atm/cam/src/chemistry/utils/prescribed_ghg.F90
A models/atm/cam/src/chemistry/utils/mo_util.F90
A models/atm/cam/src/chemistry/utils/prescribed_aero.F90
A models/atm/cam/src/chemistry/utils/tracer_data.F90
A models/atm/cam/src/chemistry/utils/prescribed_ozone.F90
A models/atm/cam/src/chemistry/utils/mo_solar_parms.F90
A models/atm/cam/src/chemistry/utils/mo_regrider.F90
A models/atm/cam/src/chemistry/utils/time_utils.F90
A models/atm/cam/src/chemistry/utils/mo_msis_ubc.F90
A models/atm/cam/src/chemistry/utils/msise00.F90
A models/atm/cam/src/chemistry/utils
A models/atm/cam/src/chemistry/mozart/mo_aero_settling.F90
A models/atm/cam/src/chemistry/mozart/mo_ghg_chem.F90
A models/atm/cam/src/chemistry/mozart/mo_wavelen.F90
A models/atm/cam/src/chemistry/mozart/mo_mean_mass.F90
A models/atm/cam/src/chemistry/mozart/mo_airglow.F90
A models/atm/cam/src/chemistry/mozart/mo_jlong.F90
A models/atm/cam/src/chemistry/mozart/mo_setext.F90
A models/atm/cam/src/chemistry/mozart/mo_airmas.F90
A models/atm/cam/src/chemistry/mozart/mo_gas_phase_chemdr.F90
A models/atm/cam/src/chemistry/mozart/charge_neutrality.F90
A models/atm/cam/src/chemistry/mozart/mo_photoin.F90
A models/atm/cam/src/chemistry/mozart/mo_setaer.F90
A models/atm/cam/src/chemistry/mozart/mo_lightning.F90
A models/atm/cam/src/chemistry/mozart/mo_tgcm_ubc.F90
A models/atm/cam/src/chemistry/mozart/mo_trislv.F90
A models/atm/cam/src/chemistry/mozart/spedata.F90
A models/atm/cam/src/chemistry/mozart/mo_setair.F90
A models/atm/cam/src/chemistry/mozart/mo_tracname.F90
A models/atm/cam/src/chemistry/mozart/mo_waveall.F90
A models/atm/cam/src/chemistry/mozart/mo_seto2.F90
A models/atm/cam/src/chemistry/mozart/clybry_fam.F90
A models/atm/cam/src/chemistry/mozart/solar_photons.F90
A models/atm/cam/src/chemistry/mozart/mo_tropopause.F90
A models/atm/cam/src/chemistry/mozart/mo_schu.F90
A models/atm/cam/src/chemistry/mozart/mo_rtlink.F90
A models/atm/cam/src/chemistry/mozart/upper_bc.F90
A models/atm/cam/src/chemistry/mozart/mo_flbc.F90
A models/atm/cam/src/chemistry/mozart/wei96.F90
A models/atm/cam/src/chemistry/mozart/mo_jshort.F90
A models/atm/cam/src/chemistry/mozart/mo_exp_sol.F90
A models/atm/cam/src/chemistry/mozart/mo_sulf.F90
A models/atm/cam/src/chemistry/mozart/mo_cph.F90
A models/atm/cam/src/chemistry/mozart/mo_strato_rates.F90
A models/atm/cam/src/chemistry/mozart/mo_chemini.F90
A models/atm/cam/src/chemistry/mozart/tracer_cnst.F90
A models/atm/cam/src/chemistry/mozart/mo_chm_diags.F90
A models/atm/cam/src/chemistry/mozart/tracer_srcs.F90
A models/atm/cam/src/chemistry/mozart/mo_lymana.F90
A models/atm/cam/src/chemistry/mozart/set_cp.F90
A models/atm/cam/src/chemistry/mozart/mo_setz.F90
A models/atm/cam/src/chemistry/mozart/mo_calcoe.F90
A models/atm/cam/src/chemistry/mozart/mo_aerosols.F90
A models/atm/cam/src/chemistry/mozart/mo_zadj.F90
A models/atm/cam/src/chemistry/mozart/mo_aurora.F90
A models/atm/cam/src/chemistry/mozart/mo_airplane.F90
A models/atm/cam/src/chemistry/mozart/mo_extfrc.F90
A models/atm/cam/src/chemistry/mozart/solvar_interface.F90
A models/atm/cam/src/chemistry/mozart/mo_pchem.F90
A models/atm/cam/src/chemistry/mozart/mo_ps2str.F90
A models/atm/cam/src/chemistry/mozart/mo_apex.F90
A models/atm/cam/src/chemistry/mozart/mo_heatnirco2.F90
A models/atm/cam/src/chemistry/mozart/woods.F90
A models/atm/cam/src/chemistry/mozart/mo_fstrat.F90
A models/atm/cam/src/chemistry/mozart/mo_sphers.F90
A models/atm/cam/src/chemistry/mozart/mo_snoe.F90
A models/atm/cam/src/chemistry/mozart/mo_sad.F90
A models/atm/cam/src/chemistry/mozart/mo_setinv.F90
A models/atm/cam/src/chemistry/mozart/euvac.F90
A models/atm/cam/src/chemistry/mozart/chemistry.F90
A models/atm/cam/src/chemistry/mozart/mo_setozo.F90
A models/atm/cam/src/chemistry/mozart/m_types.F90
A models/atm/cam/src/chemistry/mozart/solvar_woods.F90
A models/atm/cam/src/chemistry/mozart/mo_drydep.F90
A models/atm/cam/src/chemistry/mozart/iondrag.F90
A models/atm/cam/src/chemistry/mozart/mo_photo.F90
A models/atm/cam/src/chemistry/mozart/mo_sethet.F90
A models/atm/cam/src/chemistry/mozart/efield.F90
A models/atm/cam/src/chemistry/mozart/mo_setsoa.F90
A models/atm/cam/src/chemistry/mozart/apex_subs.F90
A models/atm/cam/src/chemistry/mozart/mz_aerosols_intr.F90
A models/atm/cam/src/chemistry/mozart/mo_waccm_hrates.F90
A models/atm/cam/src/chemistry/mozart/mo_mass_xforms.F90
A models/atm/cam/src/chemistry/mozart/mo_tuv_inti.F90
A models/atm/cam/src/chemistry/mozart/short_lived_species.F90
A models/atm/cam/src/chemistry/mozart/mo_strato_sad.F90
A models/atm/cam/src/chemistry/mozart/mo_waveo3.F90
A models/atm/cam/src/chemistry/mozart/solvar_data.F90
A models/atm/cam/src/chemistry/mozart/neckel.F90
A models/atm/cam/src/chemistry/mozart/tgcm_forcing.F90
A models/atm/cam/src/chemistry/mozart/mo_usrrxt.F90
A models/atm/cam/src/chemistry/mozart/mo_jeuv.F90
A models/atm/cam/src/chemistry/mozart/mag_parms.F90
A models/atm/cam/src/chemistry/mozart/spehox.F90
A models/atm/cam/src/chemistry/mozart/exbdrift.F90
A models/atm/cam/src/chemistry/mozart/mo_setcld.F90
A models/atm/cam/src/chemistry/mozart/mo_xsections.F90
A models/atm/cam/src/chemistry/mozart/mo_inter.F90
A models/atm/cam/src/chemistry/mozart/m_sad_data.F90
A models/atm/cam/src/chemistry/mozart/mo_drydep_tables.F90
A models/atm/cam/src/chemistry/mozart/mo_wavelab.F90
A models/atm/cam/src/chemistry/mozart/mo_solarproton.F90
A models/atm/cam/src/chemistry/mozart/mo_negtrc.F90
A models/atm/cam/src/chemistry/mozart/mo_synoz.F90
A models/atm/cam/src/chemistry/mozart/mo_params.F90
A models/atm/cam/src/chemistry/mozart/mo_chem_utls.F90
A models/atm/cam/src/chemistry/mozart/mo_srf_emissions.F90
A models/atm/cam/src/chemistry/mozart/sv_decomp.F90
A models/atm/cam/src/chemistry/mozart/mo_setsox.F90
A models/atm/cam/src/chemistry/mozart/mo_jpl.F90
A models/atm/cam/src/chemistry/mozart
trop_mozart and waccm_mozart source code directories combined
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/configure
M models/atm/cam/bld/config_defaults_waccm_ghg.xml
M models/atm/cam/bld/build-namelist
M models/atm/cam/bld/config_definition.xml
Implement new chem configure options and invoke chem preprocessor.
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
M models/atm/cam/bld/namelist_files/namelist_definition.xml
Chem namelist cleanup.
M models/atm/cam/bld/namelist_files/use_cases/waccm_1950_ramped.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1950_smax.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1953_ramped_qbo.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1995_climo.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1995_smin.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1995_smax.xml
M models/atm/cam/bld/namelist_files/use_cases/waccm_1950_smin.xml
M models/atm/cam/bld/namelist_files/use_cases/1870_prog_aero.xml
M models/atm/cam/bld/namelist_files/use_cases/1870-2000_control.xml
Modified for namelist variables cleanup.
M models/atm/cam/src/control/runtime_opts.F90
Chem namelist cleanup.
M models/atm/cam/src/physics/cam/initindx.F90
Prescribed aerosols and ozone independent of chemistry.
WACCM gravity wave drag.
M models/atm/cam/src/physics/cam/trb_mtn_stress.F90
WACCM change
M models/atm/cam/src/physics/cam/radiation.F90
Make restarts bit-for-bit regardless of number of steps in initial run.
M models/atm/cam/src/physics/cam/tphysac.F90
M models/atm/cam/src/physics/cam/tphysbc.F90
Cleanup. Change for short-lived waccm tracers.
M models/atm/cam/src/physics/cam/phys_prop.F90
Change in error message string.
M models/atm/cam/src/physics/cam/advnce.F90
M models/atm/cam/src/physics/cam/restart_physics.F90
Changes for prescribed rad tracers independent of chem.
MM models/atm/cam/src/physics/cam/chem_surfvals.F90
Remove lbc* (old waccm) namelist variables.
M models/atm/cam/src/physics/cam/physpkg.F90
Change for short-lived chem tracers.
Init prescribed rad tracer modules.
M models/atm/cam/src/physics/cam/pkg_cld_sediment.F90
WACCM adjustment
M models/atm/cam/src/dynamics/fv/dp_coupling.F90
M models/atm/cam/src/physics/cam/physics_types.F90
M models/atm/cam/src/physics/cam/convect_deep.F90
M models/atm/cam/src/physics/waccm/gw_drag.F90
Changes for new waccm gravity wave drag.
M models/atm/cam/src/physics/cam/chemistry.F90
Change in namelist variable interface --> read "chem_parm" namelist.
M models/atm/cam/src/physics/waccm/radheat.F90
Change for waccm lower upper boundary (without ions)
M models/atm/cam/src/dynamics/fv/cd_core.F90
M models/atm/cam/src/dynamics/fv/sw_core.F90
Changes for offline waccm (minor cleanup).
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
030 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7
044 bl374 TBL.sh fm1.9dh outfrq3s 9s ..............................FAIL! rc= 7
047 bl355 TBL.sh fmo2dh off2x2.5 9s ...............................FAIL! rc= 7
051 bl381 TBL.sh fma1.9h outfrq3s+1870_prog_aero 9s ...............FAIL! rc= 7
Sharing chemistry code changes answers for these configurations
WACCM default aerosols for radiation are now the mozart prescribed aerosols
055 bl382 TBL.sh fmgpa1.9dh outfrq3s+1870_control 9s ..............FAIL! rc= 7
Water vapor production from methane oxidation is now included
057 er731 TER.sh h16x4dm aqua 4+5s ................................FAIL! rc= 10
This also failed in cam3_6_15
bangkok/lf95:
036 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 7
This is expected to fail due to change in waccm_ghg implementation --
now uses the mozart chemical equation solver.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: trop_mozart and waccm_mozart codes are combined
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_6_15
Originator(s): eaton, aconley
Date: Sun Oct 12 16:31:36 MDT 2008
One-line Summary: RRTMG uses 3D CO2 and O2 fields. Update externals for
consistency with CCSM tags.
Purpose of changes:
. Modify RRTMG to accept 3D fields of CO2 and O2.
. Remove output of the vertically integrated cloud diagnostics from RRTMG
since those diagnostics aren't available.
. Fix logic so new ISCCP simulator is available for use with RRTMG.
. update driver code to drv/seq_mct/trunk_tags/drvseq2_0_33
. update cice code to cice/branch_tags/cice4_0_20080930b_brnchT_20081008
. update csm_share code to csm_share/trunk_tags/share3_081009
- this includes larger than roundoff changes to the surface fluxes
resulting from a bug fix introduce at share3_080811
. update the scripts code to scripts/trunk_tags/scripts4_081009
. Move the external definition for the RRTMG code into a new file,
models/atm/cam/SVN_EXTERNAL_DIRECTORIES. This was done to avoid the need
for separate external definition for RRTMG in the CCSM tags. The tag
used for the CAM src will now automatically include the RRTMG external.
Bugs fixed (include bugzilla ID): described above
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets:
. 1/2 and 1/4 degree versions of fatmlndfrc changed to use gx1v5
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do:
models/atm/cam/SVN_EXTERNAL_DIRECTORIES
. add external for RRTMG
List all existing files that have been modified, and describe the changes:
SVN_EXTERNAL_DIRECTORIES
. update driver, cice, csm_share, and scripts externals.
. remove external for RRTMG
models/atm/cam/bld/configure
. make commandline arg values of dynamics/physics/chemistry options case insensitive
. add check that RRTMG radiation only be turned on with MG microphysics.
. update CPP defines needed for new CICE tag
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. add gx1v5 versions of fatmlndfrc files for 1/2 and 1/4 degree FV grids
models/atm/cam/bld/perl5lib/Build/Config.pm
models/atm/cam/bld/perl5lib/Build/Namelist.pm
models/atm/cam/bld/perl5lib/Build/NamelistDefaults.pm
models/atm/cam/bld/perl5lib/Build/NamelistDefinition.pm
. mods to make the interfaces case insensitive with respect to the namelist
variable names.
. add some checking that the types of supplied namelist variable values
match the types declared in the definition file.
models/atm/cam/src/physics/rrtmg/radiation.F90
models/atm/cam/src/physics/rrtmg/radlw.F90
models/atm/cam/src/physics/rrtmg/radsw.F90
. mods to pass 3D fields of O2 and CO2 into RRTMG
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
004 bl132 TBL.sh e32sdh ghgrmp 9s .................................FAIL! rc= 7
006 bl133 TBL.sh e32pdh aqpgro 3s .................................FAIL! rc= 7
011 bl151 TBL.sh e64h outfrq24h 2d ................................FAIL! rc= 7
015 bl331 TBL.sh f4gdh co2rmp 9s ..................................FAIL! rc= 7
019 bl332 TBL.sh f4sdh ghgrmp 9s ..................................FAIL! rc= 7
021 bl333 TBL.sh f4pdh aqpgro 3s ..................................FAIL! rc= 7
030 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7
033 bl337 TBL.sh f4dh fv2d_8tsk 9s ................................FAIL! rc= 7
037 bl371 TBL.sh f1.9dm fvvp_lb2 9s ...............................FAIL! rc= 7
041 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7
044 bl374 TBL.sh fm1.9dh outfrq3s 9s ..............................FAIL! rc= 7
047 bl355 TBL.sh fmo2dh off2x2.5 9s ...............................FAIL! rc= 7
051 bl381 TBL.sh fma1.9h outfrq3s+1870_prog_aero 9s ...............FAIL! rc= 7
055 bl382 TBL.sh fmgpa1.9dh outfrq3s+1870_control 9s ..............FAIL! rc= 7
057 er731 TER.sh h16x4dm aqua 4+5s ................................FAIL! rc= 10
058 bl731 TBL.sh h16x4dm aqua 9s ..................................FAIL! rc= 7
bangkok/lf95: All PASS except:
004 bl111 TBL.sh e8t5mdm ttrac 9s .................................FAIL! rc= 7
008 bl112 TBL.sh e8sdm ghgrmp 9s ..................................FAIL! rc= 7
010 bl113 TBL.sh e8pdm aqpgro 3s ..................................FAIL! rc= 7
014 bl114 TBL.sh e8dm co2rmp 9s ...................................FAIL! rc= 7
019 bl153 TBL.sh e64m outfrq24h 2d ................................FAIL! rc= 7
023 bl311 TBL.sh f10t5mdm ttrac 9s ................................FAIL! rc= 7
031 bl312 TBL.sh f10sdm ghgrmp 9s .................................FAIL! rc= 7
033 bl313 TBL.sh f10pdm aqpgro 3s .................................FAIL! rc= 7
036 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 7
039 bl315 TBL.sh f10c3dm outfrq3s 9s ..............................FAIL! rc= 7
042 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7
045 bl377 TBL.sh f1.9m outfrq24h 2d ...............................FAIL! rc= 7
049 bl511 TBL.sh s8t5mdm ttrac 9s .................................FAIL! rc= 7
053 bl512 TBL.sh s8sdm ghgrmp 9s ..................................FAIL! rc= 7
055 bl513 TBL.sh s8pdm aqpgro 3s ..................................FAIL! rc= 7
057 bl553 TBL.sh s64m outfrq24h 2d ................................FAIL! rc= 7
. All baseline tests except for adiabatic and ideal physics fail as
expected due to bugfix in csm_share code
. Test er731 is an unexpected failure which will be resolved in a future
tag.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: There are larger than roundoff changes
due to a bugfix in the surface fluxes that was incorporated via updating
the csm_share external.
===============================================================
===============================================================
Tag name: cam3_6_14
Originator(s): mvertens
Date: Sun Sep 28 20:15:13 MDT 2008
One-line Summary: changes to cam.cpl7.template and camdom.cpl7.template for new scripts build
Purpose of changes: Bug fixes to cam.cpl7.template and camdom.cpl7.template to get it
to work correctly with new build
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none (does not apply to cam build yet)
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: myself
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
M bld/camdom.cpl7.template
M bld/cam.cpl7.template
bug fixes to the way gmake is invoked for the new build
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: no testing since cam build/makefile was not touched
bangkok/lf95: no testing since cam build/makefile was not touched
Summarize any changes to answers: bfb in the cam build
===============================================================
Tag name: cam3_6_13
Originator(s): aconley, eaton
Date: Thu Sep 25 18:42:12 MDT 2008
One-line Summary: new cloud optics for RRTMG
Purpose of changes:
. Add new cloud optics for RRTMG.
- Note 1: This code has been added to the trunk to provide early
access for developers. However, the port of the code from the Andrew's
branch to the trunk has not yet been scientifically validated.
- Note 2: The RRTMG code is not yet the default. It is enabled by
specifying -rrtmg on the configure commandline. When this is done, the
default for the cloud optics is set to be the mitchell ice cloud optics
and the conley liquid cloud optics. There is currently a capability to
switch the cloud optics back to CAM3 versions by setting the namelist
variables liqcldoptics='slingo' and icecldoptics='ebertcurry'.
- Note 3: There is still alot of work to be done to clean up the new
cloud optics code. But exactly what needs to be done will depend on
whether both the new and old cloud optics schemes need to be available
for use with RRTMG. The old optics will remain available for use in a
CAM3 backwards compatibility mode in any case.
. Add new version (3.8) of the ISCCP cloud simulator. This version only
works with RRTMG. The old cloud simulator is maintained to work with
CAMRT.
Bugs fixed (include bugzilla ID):
. Some bug fixes in cam.cpl7.template needed for CCSM builds
- fix the logic for the CAM_CHEM variable so that setting CAM_CHEM='none'
results in setting the arg "-chem none" for configure.
- remove the override that forced CAM to always be built with threading on.
Describe any changes made to build system: none
Describe any changes made to the namelist:
. Add the following variables for cloud optics:
- liqcldoptics type="char*32" valid_values="slingo,conley"
filepath and name for ice optics data for rrtmg
- icecldoptics type="char*32" valid_values="ebertcurry,mitchell"
filepath and name for ice optics data for rrtmg
- iceopticsfile type="char*256"
filepath and name for ice optics data for rrtmg
- liqopticsfile type="char*256"
filepath and name for liquid cloud (gamma distributed) optics data for
rrtmg
. Add new variable, doisccp_38, to turn on the new version of the isccp
cloud simulator. This is a logical variable and is false by default.
List any changes to the defaults for the boundary datasets:
. New physprop files from Steve Ghan. New variables have been added to the
new set of files (*_c080918.nc), and all the existing variables have the
same values as in the previous set of files (*_c080819.nc).
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do:
models/atm/cam/src/physics/cam/cloud_rad_props.F90
. this is currently stub code
models/atm/cam/src/physics/cam/cloudsimulator_38.F90
models/atm/cam/src/physics/cam/icarus_scops_38.F90
. new version (3.8) of the ISCCP cloud simulator
models/atm/cam/src/physics/rrtmg/cloud_rad_props.F90
models/atm/cam/src/physics/rrtmg/ebert_curry.F90
models/atm/cam/src/physics/rrtmg/oldcloud.F90
models/atm/cam/src/physics/rrtmg/param_cldoptics.F90
models/atm/cam/src/physics/rrtmg/slingo.F90
. code and interfaces to allow switching between new and old cloud optics
for RRTMG only.
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/build-namelist
. add defaults for new cloud optics variables when running with rrtmg.
. modify the filepaths produced for the -inputdata option to retain leading
shell variables (e.g., $DIN_LOC_ROOT).
models/atm/cam/bld/cam.cpl7.template
. fix the logic for the CAM_CHEM variable so that setting CAM_CHEM='none'
results in setting the arg "-chem none" for configure.
. remove the override that forced CAM to always be built with threading on.
. mods for using unified Makefile and Macros files for CCSM build.
models/atm/cam/bld/configure
. create stub versions of the misc.h and preproc.h files which are still
reference by CLM code, but not used.
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. new physprop files
. add defaults for new cloud optics variables.
models/atm/cam/bld/namelist_files/namelist_definition.xml
. new namelist variables oldcldoptics, liqcldoptics, icecldoptics,
iceopticsfile, liqopticsfile, doisccp_38
. update the documentation for the input_pathname attribute, and remove the
attribute from the variables used to specify the restart files used for
branch runs.
models/atm/cam/src/control/runtime_opts.F90
. add doisccp_38
models/atm/cam/src/physics/cam/aer_rad_props.F90
. change idxvis to idx_sw_diag
models/atm/cam/src/physics/cam/cldwat2m.F90
. add new diagnostic fields
models/atm/cam/src/physics/cam/phys_prop.F90
. change idxVIS to idx_sw_diag
models/atm/cam/src/physics/cam/physpkg.F90
. add call to cloud_rad_props_init
. remove init for cloudsimulator
models/atm/cam/src/physics/cam/rad_constituents.F90
. add public data for ice/liq optics files?
models/atm/cam/src/physics/cam/radconstants.F90
. wavelength band indices for diagnostic fields
models/atm/cam/src/physics/cam/radiation.F90
. add initialization of cloud simulator to radiation_init
models/atm/cam/src/physics/cam/stratiform.F90
. add microphysics fields to pbuf
. Change limiters on in-cloud ice and liquid mixing ratios. These fields
are only set by MG microphysics. They are used in the new cloud optics
routines, which means the new cloud optics won't work with RK
microphysics.
models/atm/cam/src/physics/rrtmg/radconstants.F90
. add wavelength band indexes for diagnostic fields
models/atm/cam/src/physics/rrtmg/radiation.F90
. mods for diagnostic calcs
. mods for cloud simulator.
. call cloudsimulator_init from radiation_init
. add diagnostic calc specific versions of output fields
. mods for cloud optics
. remove cicewp, cliqwp, rei, and rel from the args passed to
rad_rrtmg_{sw,lw}.
. add cloud optics fields as input args to the RT calls
. remove old cloud cover diagnostics that aren't consistent with mcica.
models/atm/cam/src/physics/rrtmg/radlw.F90
. change dummy arg from cldtau to tauc_lw which has a
wavelength dependence (nbndlw). Previously tauc_lw was a local var.
. remove input dummy args E_cicewp, E_cliqwp, E_rei, E_rel. CAM is no
longer providing these fields as input
models/atm/cam/src/physics/rrtmg/radsw.F90
. get idx_sw_diag from radconstants
. add args for cld optics to rad_rrtmg_sw
. remove input dummy args E_cicewp, E_cliqwp, E_rei, E_rel. CAM is no
longer providing these fields as input
. code for different versions of cloud optics -- maintain it all for now
. Remove public subroutine radsw_cldoptics (original implementation of old
cloud optics for rrtmg)
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS
bangkok/lf95: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB. Note that there are currently no
regression tests for RRTMG. The answers with RRTMG have changed due to new
cloud optics.
===============================================================
===============================================================
Tag name: cam3_6_12
Originator(s): Jim Edwards
Date: 9-23-08
One-line Summary: Update of Homme dycore, code cleanup to prepare for IO changes
Purpose of changes: Remove access to module pmgrid variables plat and plon and
require access through the dyn_grid module.
Bugs fixed (include bugzilla ID): Fixed a bug in boundary data, this bug
manafest in the homme dycore only.
Describe any changes made to build system: Removed macros DYN_STATE_INTERFACE
and LSMLON, LSMLAT (except for the scam build) from configure.
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
models/atm/cam/src/dynamics/homme/external/flops_mod.F90
models/atm/cam/src/dynamics/homme/external/forcing_mod.F90
models/atm/cam/src/dynamics/homme/external/surfaces_mod.F90
models/atm/cam/src/dynamics/homme/external/baroclinic_inst_mod.F90
models/atm/cam/src/dynamics/homme/external/checksum_mod.F90
models/atm/cam/src/dynamics/homme/external/timer_mod.F90
Homme code not used in cam interface.
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
models/atm/cam/test/system/config_files/h16x4dm
removed reference to lapack lib - this is handled in configure
models/atm/cam/bld/configure
removed obsolete macros DYN_STATE_INTERFACE and LSMLON, LSMLAT
(except for scam case)
models/atm/cam/bld/Makefile.in
added LAPACK libs to Linux build (needed for homme)
added NO_MPI2 macro to Linux build (needed for PIO)
models/atm/cam/src/control/startup_initialconds.F90
models/atm/cam/src/control/cam_restart.F90
models/atm/cam/src/control/cam_comp.F90
Removed DYN_STATE_INTERFACE clauses
models/atm/cam/src/physics/cam/boundarydata.F90
Fixed a bug in interpolation code used for homme dycore only
models/atm/cam/src/physics/cam/progseasalts_intr.F90
models/atm/cam/src/physics/cam/acbnd.F90
models/atm/cam/src/physics/cam/carbon_intr.F90
models/atm/cam/src/physics/cam/spmd_phys.F90
models/atm/cam/src/physics/cam/sulfur_intr.F90
models/atm/cam/src/chemistry/trop_mozart/mo_gas_phase_chemdr.F90
models/atm/cam/src/chemistry/trop_mozart/mo_lightning.F90
models/atm/cam/src/chemistry/trop_mozart/mo_tropopause.F90
models/atm/cam/src/chemistry/trop_mozart/mo_flbc.F90
models/atm/cam/src/chemistry/trop_mozart/mo_sulf.F90
models/atm/cam/src/chemistry/trop_mozart/mo_chemini.F90
models/atm/cam/src/chemistry/trop_mozart/mo_aerosols.F90
models/atm/cam/src/chemistry/trop_mozart/mo_airplane.F90
models/atm/cam/src/chemistry/trop_mozart/mo_extfrc.F90
models/atm/cam/src/chemistry/trop_mozart/mo_regrider.F90
models/atm/cam/src/chemistry/trop_mozart/mo_fstrat.F90
models/atm/cam/src/chemistry/trop_mozart/chemistry.F90
models/atm/cam/src/chemistry/trop_mozart/tracer_data.F90
models/atm/cam/src/chemistry/trop_mozart/mo_drydep.F90
models/atm/cam/src/chemistry/trop_mozart/mo_photo.F90
models/atm/cam/src/chemistry/trop_mozart/mo_seasalt.F90
models/atm/cam/src/chemistry/trop_mozart/mz_aerosols_intr.F90
models/atm/cam/src/chemistry/trop_mozart/mo_strato_sad.F90
models/atm/cam/src/chemistry/trop_mozart/mo_synoz.F90
models/atm/cam/src/chemistry/trop_mozart/mo_srf_emissions.F90
Removed reference to pmgrid module variables plon and plat and
replaced with calls to get_dyn_grid_parm
models/atm/cam/src/dynamics/sld/pmgrid.F90
models/atm/cam/src/dynamics/sld/inidat.F90
models/atm/cam/src/dynamics/sld/dyn_comp.F90
models/atm/cam/src/dynamics/sld/inital.F90
models/atm/cam/src/dynamics/sld/stepon.F90
models/atm/cam/src/dynamics/sld/restart_dynamics.F90
models/atm/cam/src/dynamics/eul/inidat.F90
models/atm/cam/src/dynamics/eul/dyn_comp.F90
models/atm/cam/src/dynamics/eul/inital.F90
models/atm/cam/src/dynamics/eul/stepon.F90
models/atm/cam/src/dynamics/eul/restart_dynamics.F90
Added empty dyn state structure so that dyn interface
matchs homme and fv and DYN_STATE_INTERFACE macro could
be removed
models/atm/cam/src/dynamics/homme/external/prim_advance_mod.F90
models/atm/cam/src/dynamics/homme/external/parallel_mod.F90
models/atm/cam/src/dynamics/homme/external/reduction_mod.F90
models/atm/cam/src/dynamics/homme/external/prim_state_mod.F90
models/atm/cam/src/dynamics/homme/external/dimensions_mod.F90
models/atm/cam/src/dynamics/homme/external/control_mod.F90
models/atm/cam/src/dynamics/homme/external/prim_driver_mod.F90
models/atm/cam/src/dynamics/homme/external/interpolate_mod.F90
models/atm/cam/src/dynamics/homme/external/element_mod.F90
models/atm/cam/src/dynamics/homme/external/viscosity_mod.F90
models/atm/cam/src/dynamics/homme/external/time_mod.F90
models/atm/cam/src/dynamics/homme/external/hybvcoord_mod.F90
models/atm/cam/src/dynamics/homme/external/prim_advection_mod.F90
models/atm/cam/src/dynamics/homme/external/edge_mod.F90
models/atm/cam/src/dynamics/homme/external/dof_mod.F90
models/atm/cam/src/dynamics/homme/external/physical_constants.F90
models/atm/cam/src/dynamics/homme/external/namelist_mod.F90
models/atm/cam/src/dynamics/homme/external/bndry_mod.F90
Ported to lf95 compiler and merged development from homme.
models/atm/cam/src/dynamics/homme/stepon.F90
models/atm/cam/src/dynamics/homme/dyn_comp.F90
Ported to lf95 compiler
SVN_EXTERNAL_DIRECTORIES
Updated mct, drv, and scripts tags.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: all pass except
058 bl731 TBL.sh h16x4dm aqua 9s ..................................FAIL
expected failure due to homme dycore updates.
bangkok/lf95: all pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_6_11
Originator(s): eaton, fvitt
Date: Thu Sep 18 14:27:10 MDT 2008
One-line Summary: wet deposition bug fix
Purpose of changes:
. This bug fix removes the double counting of snow in both the total
precipitation and total evaporation fields that were being passed to the
gas phase chemistry and used in wet deposition.
Bugs fixed (include bugzilla ID):
. described above
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/src/chemistry/trop_mozart/chemistry.F90
. use physics buffer fields NEVAPR instead of TOTEVAPR, and PRAIN instead
of TOTPRECP as actual args passed through the gas_phase_chemdr interface.
models/atm/cam/src/physics/cam/cldwat2m.F90
. fix the comments for prain and nevapr to indicate that they are totals of
rain + snow.
models/atm/cam/src/physics/cam/stratiform.F90
. remove the fields TOTEVAPR (nevapr+evapsnow) and TOTPRECP
(prain+prodsnow) from the physics buffer. These fields were double
counting the contribution from snow since it is already included in the
prain and nevapr fields returned by subroutines pcond and mmicro_pcond.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
044 bl374 TBL.sh fm1.9dh outfrq3s 9s ..............................FAIL! rc= 7
047 bl355 TBL.sh fmo2dh off2x2.5 9s ...............................FAIL! rc= 7
051 bl381 TBL.sh fma1.9h outfrq3s+1870_prog_aero 9s ...............FAIL! rc= 7
These baselines fail because answers for both full chemistry and prognostic
aerosol codes have changed as a result of the wet deposition change.
bangkok/lf95: All PASS.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB except for runs with full chemistry
(-chem trop_mozart) or prognostic aerosols (-chem trop_mozart_aero).
===============================================================
===============================================================
Tag name: cam3_6_10
Originator(s): eaton
Date: Fri Sep 12 12:34:59 MDT 2008
One-line Summary: build-namelist fix and update driver
Purpose of changes:
. build-namelist was modified to properly produce the list of input
datasets when the root directory has been specified as an environment
variable reference, e.g., "-csmdata \$DIN_LOC_ROOT". This is a
requirement for how the CCSM scripts invoke build-namelist.
. Update to new driver code. The new driver contains a workaround for the
performance problem being encountered on bluefire when running more than
1 thread per task. Driver updated to drvseq2_0_27.
. pathf90 section of Makefile modified to turn off optimization for the 2
RRTMG modules that take excess amounts of time to compile (or won't
compile at all).
Bugs fixed (include bugzilla ID):
. build-namelist (described above)
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none known
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
SVN_EXTERNAL_DIRECTORIES
. update driver external to drvseq2_0_27
models/atm/cam/bld/config_definition.xml
. change valid value of lnd from clm2 to clm
models/atm/cam/bld/build-namelist
. modify so that the namelist files for the driver, CLM, and CICE are only
written out for CAM standalone scripts.
. modify so that the list of inputdata file produced has the root directory
properly stripped, even when it's in the form of an environment variable
as is the case when the CCSM scripts invoke build-namelist.
models/atm/cam/bld/Makefile.in
. modify to remove optimization from build of rrtmg_sw_k_g.f90 and
rrtmg_lw_k_g.f90 when using pathf90.
models/atm/cam/bld/namelist_files/namelist_definition.xml
. add new namelist variables to seq_infodata_inparm group.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
004 bl132 TBL.sh e32sdh ghgrmp 9s .................................FAIL! rc= 7
011 bl151 TBL.sh e64h outfrq24h 2d ................................FAIL! rc= 7
015 bl331 TBL.sh f4gdh co2rmp 9s ..................................FAIL! rc= 7
019 bl332 TBL.sh f4sdh ghgrmp 9s ..................................FAIL! rc= 7
030 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7
033 bl337 TBL.sh f4dh fv2d_8tsk 9s ................................FAIL! rc= 7
047 bl355 TBL.sh fmo2dh off2x2.5 9s ...............................FAIL! rc= 7
bangkok/lf95: All PASS except:
004 bl111 TBL.sh e8t5mdm ttrac 9s .................................FAIL! rc= 7
008 bl112 TBL.sh e8sdm ghgrmp 9s ..................................FAIL! rc= 7
014 bl114 TBL.sh e8dm co2rmp 9s ...................................FAIL! rc= 7
019 bl153 TBL.sh e64m outfrq24h 2d ................................FAIL! rc= 7
023 bl311 TBL.sh f10t5mdm ttrac 9s ................................FAIL! rc= 7
031 bl312 TBL.sh f10sdm ghgrmp 9s .................................FAIL! rc= 7
036 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 7
039 bl315 TBL.sh f10c3dm outfrq3s 9s ..............................FAIL! rc= 7
042 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7
049 bl511 TBL.sh s8t5mdm ttrac 9s .................................FAIL! rc= 7
053 bl512 TBL.sh s8sdm ghgrmp 9s ..................................FAIL! rc= 7
057 bl553 TBL.sh s64m outfrq24h 2d ................................FAIL! rc= 7
The baseline test failures are due to roundoff level differences in the
driver code. Many of the baseline tests pass either because the
configuration doesn't involve surface interactions (e.g., adiabatic, ideal
physics, aqua-planet), or due to some unknown coincidence all the FV
1.9x2.5 tests are passing.
CAM tag used for the baseline comparison tests if different than previous
tag: cam3_6_07
Summarize any changes to answers: BFB or roundoff. The roundoff
differences were verified by looking at the differences in the cprnc output
for all fields at each timestep for a short run. The normal perturbation
growth test isn't applicable because it uses aquaplanet mode which is BFB.
===============================================================
===============================================================
Tag name: cam3_6_09
Originator(s): Jim Edwards, Mark Taylor (HOMME dycore), Pat Worley (phys_grid.F90)
Date: 09-04-2008
One-line Summary: Update of HOMME dynamic core
Purpose of changes: Update of HOMME dynamic core
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: modified to allow HOMME dycore restart using PIO
Describe any changes made to the namelist: added pio_ctl namelist
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
- models/atm/cam/test/system/config_files/h16x4dm
- models/atm/cam/test/system/nl_files/aqua
These were added in support of new homme aquaplanet test 731
A + - models/atm/cam/src/utils/pio_utils.F90
Added to support HOMME dycore restart, also added a new external
utils/pio which is currently only compiled for the HOMME dycore.
List all existing files that have been modified, and describe the changes:
models/ocn/dom/ocn_comp_mct.F90
models/ice/csim4/ice_comp_mct.F90
models/atm/cam/src/control/atm_comp_mct.F90
models/atm/cam/src/physics/cam/phys_gmean.F90
I made the phys_grid variable nlcols private and provided a method to copy it.
models/atm/cam/test/system/tests_pretag_bluefire
Removed the 711 tests and added the 731 tests
models/atm/cam/test/system/config_files/h5x8adm
Added path to pnetcdf on bluefire
models/atm/cam/test/system/TER.sh
Revised pattern for cprnc tests so that netcdf restart files are not included.
Improved error message for cice namelist
models/atm/cam/test/system/input_tests_master
Added tests for HOMME dycore aquaplanet configuration
models/atm/cam/test/system/TSM.sh
Improved error message for cice namelist
models/atm/cam/bld/config_defaults_homme.xml
Added support for ne=16 np=4 (~2 degree) HOMME dycore resolution
models/atm/cam/bld/configure
models/atm/cam/bld/Makefile.in
models/atm/cam/bld/config_horiz_grid.xml
models/atm/cam/bld/config_definition.xml
Added support for building PIO and pnetcdf
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
models/atm/cam/bld/namelist_files/namelist_definition.xml
models/atm/cam/bld/build-namelist
Added support for pio_ctl namelist, changed some HOMME dycore defaults
models/atm/cam/src/control/cam_restart.F90
models/atm/cam/src/control/cam_comp.F90
Added support for PIO based restart in HOMME dycore
models/atm/cam/src/utils/spmd_utils.F90
Added method to export mpicom from this module
models/atm/cam/src/physics/cam/boundarydata.F90
Bug fix, this bug was exposed by the HOMME dycore only.
models/atm/cam/src/physics/cam/physics_types.F90
Changed a test to add homme to the list of dycores who do not need this subroutine
models/atm/cam/src/physics/cam/phys_grid.F90
Made nlcols private and provided a method to retreve it's value.
Changed init method to make it more efficient for 1d horizontal grids.
models/atm/cam/src/dynamics/homme/pmgrid.F90
models/atm/cam/src/dynamics/homme/dycore.F90
models/atm/cam/src/dynamics/homme/external/filter_mod.F90
models/atm/cam/src/dynamics/homme/external/prim_advance_mod.F90
models/atm/cam/src/dynamics/homme/external/cg_mod.F90
models/atm/cam/src/dynamics/homme/external/parallel_mod.F90
models/atm/cam/src/dynamics/homme/external/derivative_mod.F90
models/atm/cam/src/dynamics/homme/external/prim_si_mod.F90
models/atm/cam/src/dynamics/homme/external/prim_state_mod.F90
models/atm/cam/src/dynamics/homme/external/dimensions_mod.F90
models/atm/cam/src/dynamics/homme/external/quadrature_mod.F90
models/atm/cam/src/dynamics/homme/external/cube_mod.F90
models/atm/cam/src/dynamics/homme/external/global_norms_mod.F90
models/atm/cam/src/dynamics/homme/external/control_mod.F90
models/atm/cam/src/dynamics/homme/external/prim_driver_mod.F90
models/atm/cam/src/dynamics/homme/external/interpolate_mod.F90
models/atm/cam/src/dynamics/homme/external/linear_algebra_mod.F90
models/atm/cam/src/dynamics/homme/external/element_mod.F90
models/atm/cam/src/dynamics/homme/external/viscosity_mod.F90
models/atm/cam/src/dynamics/homme/external/prim_si_ref_mod.F90
models/atm/cam/src/dynamics/homme/external/time_mod.F90
models/atm/cam/src/dynamics/homme/external/prim_advection_mod.F90
models/atm/cam/src/dynamics/homme/external/kinds.F90
models/atm/cam/src/dynamics/homme/external/physical_constants.F90
models/atm/cam/src/dynamics/homme/external/solver_mod.F90
models/atm/cam/src/dynamics/homme/external/namelist_mod.F90
models/atm/cam/src/dynamics/homme/dyn_grid.F90
models/atm/cam/src/dynamics/homme/initcom.F90
models/atm/cam/src/dynamics/homme/dp_coupling.F90
models/atm/cam/src/dynamics/homme/dyn_comp.F90
models/atm/cam/src/dynamics/homme/inidat.F90
models/atm/cam/src/dynamics/homme/stepon.F90
models/atm/cam/src/dynamics/homme/restart_dynamics.F90
models/atm/cam/src/dynamics/homme/commap.F90
models/atm/cam/src/dynamics/homme/spmd_dyn.F90
models/atm/cam/src/dynamics/homme/inital.F90
Updated from development on homme_cam3_5_29 branch.
SVN_EXTERNAL_DIRECTORIES
Added PIO external.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: all pass except er731, new test with no baseline to compare to.
bangkok/lf95: all pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: bfb
===============================================================
Tag name: cam3_6_08
Originator(s): eaton, jenkay, mvr
Date: Thur Sep 4 2008
One-line Summary: build-namelist fix; new radiation and microphysics
diagnostic fields
Purpose of changes:
. fix build-namelist to recognize the input datasets that occur in the
rad_climate and rad_diag_N variables. This is only relevent
for the -test option which checks that all required input datasets exist
on a local disk, and for the -inputdata option which writes a list of the
required input datasets to a file.
. Add precip fluxes calculations and output fields (MGFLXPRC,
MGFLXSNW) to cldwat2m.F90 (from Jen Kay).
. Add additional radiation fields (FLDS, FSUTOA) to radiation.F90 and
radsw.F90 (from Jen Kay).
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist:
- fix to build-namelist to recognize input datasets associated
with -test option
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/test_driver.sh
- increased wall clock for bluefire testing
M models/atm/cam/bld/build-namelist
- fix to recognize input datasets associated with -test option
M models/atm/cam/src/physics/cam/cldwat2m.F90
- add precip fluxes calculations and output fields (MGFLXPRC, MGFLXSNW)
M models/atm/cam/src/physics/cam/radiation.F90
M models/atm/cam/src/physics/cam/radsw.F90
M models/atm/cam/src/physics/rrtmg/radiation.F90
M models/atm/cam/src/physics/rrtmg/radsw.F90
- Add additional radiation fields (FLDS, FSUTOA)
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS
bangkok/lf95: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: b4b
===============================================================
Tag name: cam3_6_07
Originator(s): eaton, aconley, mvr
Date: Fri Aug 29 18:24:14 MDT 2008
One-line Summary: new physprop (optics + microphysics) files; fixes for
RRTMG aerosol optics; build/run fixes for CICE
Purpose of changes:
. Update code to read new physprop datasets which contain updated aerosol
optics and microphysical properties for both the CAMRT and RRTMG codes.
These are the latest data from Steve Ghan that have been converted to CAM
naming conventions by Andrew Conley. These physprop datasets are the
ones currently being used by the radiation team in the evaluation of the
RRTMG code.
. Update the relative humidity calculation done in the CAMRT code to be
consistent with the rh calcs in the moist physics code and with the
value of rh written to the history output (i.e., rh=q/qsat).
. Fix bug in reading aerosol lw optical properties. This bug was the cause
of RRTMG solutions not being invariant when task and/or thread counts were
changed. This did not affect CAMRT simulations.
. Remove storage of optical depths from RRTMG code.
- Update rrtmg external definition to trunk_tags/rrtmg_080826
. The cam3_6_05 commit included new driver code that allows different
components to use different thread counts. This is controlled using
variables from the driver namelist group ccsm_pes. Update build-namelist
to write this namelist group to drv_in. By default the *_nthreads
variables are set to 1. This means that runs using cam3_6_05 and
cam3_6_06 will run with 1 thread in each task regardless of the setting
of OMP_NUM_THREADS. build-namelist will now set the value of *_nthreads
to the value of OMP_NUM_THREADS. Also, the user may explicitly set the
values of *_nthreads to achieve a heterogeneous threading configuration.
. Fixes to regression tests on jaguar addressing CICE and quad-core issues.
Bugs fixed (include bugzilla ID):
. Fix bug in reading aerosol lw optical properties.
Describe any changes made to build system: none
Describe any changes made to the namelist:
. iradae is no longer used by RRTMG
List any changes to the defaults for the boundary datasets:
. new physprops datasets.
. change the focndomain defaults for 1/2 and 1/4 degree fv to use the gx1v5
ocn grid rather than USGS.
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
models/atm/cam/src/physics/rrtmg/aer_rad_props.F90
. no longer needed since the physics/cam version of this file is identical
to it after the relative humidity expression is updated
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
SVN_EXTERNAL_DIRECTORIES
. update rrtmg external to trunk_tags/rrtmg_080826
models/atm/cam/bld/build-namelist
. set defaults and write the ccsm_pes namelist group to drv_in
models/atm/cam/bld/Makefile.in
. remove optimization from the PGI build of the RRTMG files
rrtmg_lw_k_g.f90 and rrtmg_sw_k_g.f90.
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. new physprops datasets
models/atm/cam/bld/namelist_files/namelist_definition.xml
. add driver namelist group ccsm_pes to control the task/thread layout.
models/atm/cam/src/physics/cam/aer_rad_props.F90
.
models/atm/cam/src/physics/cam/phys_prop.F90
. Update code to read new physprop datasets. Mainly this is new variable
names in the netcdf files.
. bugfix for reading lw_ext data. flw_ext was dimensioned with nswbands
instead of nlwbands.
models/atm/cam/src/physics/rrtmg/radiation.F90
models/atm/cam/src/physics/rrtmg/radlw.F90
models/atm/cam/src/physics/rrtmg/radsw.F90
. remove dotau option for controlling calculation frequency of optical
depths. This was previously set by using the iradae namelist variable
which has now been removed from RRTMG.
models/atm/cam/test/system/TCB.sh
. modify code to look for 'csim4' in config_files to flag that cice is not
the ice model
models/atm/cam/test/system/TBR.sh
models/atm/cam/test/system/TER.sh
. set OMP_NUM_THREADS to CAM_RESTART_THREADS before calling build-namelist
. put conditional around invocation of the cice build-namelist so that it's
not invoked when csim4 is being used.
models/atm/cam/test/system/TSM.sh
. set OMP_NUM_THREADS to CAM_THREADS before calling build-namelist
models/atm/cam/test/system/CAM_runcmnd.sh
. generalized the call to aprun to assign the number of tasks per node
based on the requested tasks and threads, assuming the use of quad-core
nodes
models/atm/cam/test/system/test_driver.sh
. modified task/thread request for hybrid testing on jaguar to work
with changes to test scripts for using the new cice model; fix to default
workspace used for test output
. version of showproj - the utility to grab a charge account for a user -
is now consistent with what ccsm is using...
models/atm/cam/test/system/config_files/f10c3dm
models/atm/cam/test/system/config_files/h5x8adm
models/atm/cam/test/system/config_files/h5x8dm
models/atm/cam/test/system/config_files/h5x8idm
. set -ice csim4. The f10c3dm config is designed to test the cam3.5
configuration. The homme tests aren't currently using a sea ice model,
but when they do the csim4 model will probably be the easiest to work
with initially.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
. The regression testing was carried out in two stages. First, all changes
except using the new optics/microphysics data for camrt and switching the
expression for rh used by camrt were tested and shown to be bit-for-bit.
Then those changes were made and the testing redone. The second set of
regression tests which include failed baseline tests is what is reported
below.
bluefire: All PASS except
004 bl132 TBL.sh e32sdh ghgrmp 9s .................................FAIL! rc= 7
006 bl133 TBL.sh e32pdh aqpgro 3s .................................FAIL! rc= 7
011 bl151 TBL.sh e64h outfrq24h 2d ................................FAIL! rc= 7
015 bl331 TBL.sh f4gdh co2rmp 9s ..................................FAIL! rc= 7
019 bl332 TBL.sh f4sdh ghgrmp 9s ..................................FAIL! rc= 7
021 bl333 TBL.sh f4pdh aqpgro 3s ..................................FAIL! rc= 7
030 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7
033 bl337 TBL.sh f4dh fv2d_8tsk 9s ................................FAIL! rc= 7
037 bl371 TBL.sh f1.9dm fvvp_lb2 9s ...............................FAIL! rc= 7
041 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7
044 bl374 TBL.sh fm1.9dh outfrq3s 9s ..............................FAIL! rc= 7
047 bl355 TBL.sh fmo2dh off2x2.5 9s ...............................FAIL! rc= 7
051 bl381 TBL.sh fma1.9h outfrq3s+1870_prog_aero 9s ...............FAIL! rc= 7
055 bl382 TBL.sh fmgpa1.9dh outfrq3s+1870_control 9s ..............FAIL! rc= 7
bangkok/lf95: All PASS except
004 bl111 TBL.sh e8t5mdm ttrac 9s .................................FAIL! rc= 7
008 bl112 TBL.sh e8sdm ghgrmp 9s ..................................FAIL! rc= 7
010 bl113 TBL.sh e8pdm aqpgro 3s ..................................FAIL! rc= 7
014 bl114 TBL.sh e8dm co2rmp 9s ...................................FAIL! rc= 7
019 bl153 TBL.sh e64m outfrq24h 2d ................................FAIL! rc= 7
023 bl311 TBL.sh f10t5mdm ttrac 9s ................................FAIL! rc= 7
031 bl312 TBL.sh f10sdm ghgrmp 9s .................................FAIL! rc= 7
033 bl313 TBL.sh f10pdm aqpgro 3s .................................FAIL! rc= 7
036 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 7
039 bl315 TBL.sh f10c3dm outfrq3s 9s ..............................FAIL! rc= 7
042 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7
045 bl377 TBL.sh f1.9m outfrq24h 2d ...............................FAIL! rc= 4
049 bl511 TBL.sh s8t5mdm ttrac 9s .................................FAIL! rc= 7
053 bl512 TBL.sh s8sdm ghgrmp 9s ..................................FAIL! rc= 7
055 bl513 TBL.sh s8pdm aqpgro 3s ..................................FAIL! rc= 7
057 bl553 TBL.sh s64m outfrq24h 2d ................................FAIL! rc= 7
All baseline tests except the adiabatic and ideal physics tests fail due to
the changes in the camrt optics and the microphysics data.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: new climate -- validation runs are being
done by the radiation team.
===============================================================
===============================================================
Tag name: cam3_6_06
Originator(s): mvertens
Date: Wed Aug 27 10:45:00 MDT 2008
One-line Summary: removal of cpl6 refs and script mods for trigrid support
Purpose of changes: removal of all cpl6 references and cam.cpl7.template
changes needed for upcoming trigrid support
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none expected
Code reviewed by: eaton
List all subroutines eliminated:
D models/atm/cam/bld/cam.cpl6.template
D models/atm/cam/src/control/ccsm_msg.F90
D models/atm/cam/src/control/con_cam.F90
- removal of cpl6 references
D models/atm/cam/bld/camsom.cpl7.template
D models/atm/cam/bld/camcsim.cpl7.template
- compsets for camsom and camcsim will no longer be there in cpl7 scripts
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/configure
M models/atm/cam/bld/build-namelist
M models/atm/cam/bld/config_definition.xml
- removed cpl6 references
M models/atm/cam/bld/cam.cpl7.template
- put in support needed for trigrid upcoming changes
M models/atm/cam/src/control/mpishorthand.F
M models/atm/cam/src/control/atm_comp_mct.F90
M models/atm/cam/src/control/cam_comp.F90
M models/atm/cam/src/utils/abortutils.F90
X models/atm/cam/src/physics/rrtmg/ext
M models/atm/cam/src/dynamics/sld/dynpkg.F90
M models/atm/cam/src/dynamics/fv/trac2d.F90
M models/atm/cam/src/dynamics/fv/cd_core.F90
- removed references to COUPCSM_
M models/atm/cam/src/control/runtime_opts.F90
- removed references to COUPCSM_
M SVN_EXTERNAL_DIRECTORIES
- updated scripts to https://svn-ccsm-models.cgd.ucar.edu/scripts/trunk_tags/scripts4_080820b
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: all tests passed
bangkok/lf95: all tests passed
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: answers are bfb
===============================================================
===============================================================
Tag name: cam3_6_05
Originator(s): John T., Mariana V.
Date: 082208
One-line Summary:Put new CICE code on the trunk as default ice model
Purpose of changes:new cice code default, scam update
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:none
Describe any changes made to the namelist:none
List any changes to the defaults for the boundary datasets:added new
cami_0000-01-01_64x128_T42_L26_c080822.nc
Describe any substantial timing or memory changes: Mariana reports a
2%-3% hit to timing not sure of memory changes.
Code reviewed by:Mariana
List all subroutines eliminated:
List all subroutines added and what they do:
These are all new configure scripts and namelists for testing scam
A models/atm/cam/test/system/config_files/scmarmiop
A models/atm/cam/test/system/config_files/scm64bfbiop_35
A models/atm/cam/test/system/config_files/e64bfbiop_35
A models/atm/cam/test/system/nl_files/scm_prep_35
A models/atm/cam/test/system/nl_files/scmarm
List all existing files that have been modified, and describe the changes:
+----
| models/atm/cam/test/system/config_files/f10sdm
| models/atm/cam/test/system/config_files/e64sm
| models/atm/cam/test/system/config_files/f4sdh
| models/atm/cam/test/system/config_files/e8sdm
| models/atm/cam/test/system/config_files/e32sdh
| models/atm/cam/test/system/config_files/s8sdm
| models/atm/cam/test/system/config_files/s32sdh
+---> Got rid of old som tests
models/atm/cam/test/system/config_files/h5x8dm
models/atm/cam/test/system/config_files/h5x8adm
models/atm/cam/test/system/config_files/h5x8idm
+----
| models/atm/cam/test/system/config_files/scm64bfbiop_35
| models/atm/cam/test/system/config_files/e64bfbiop_35
| models/atm/cam/test/system/nl_files/scm_prep_35
+---> new scam tests
+----
| models/atm/cam/test/system/tests_pretag_bangkok
| models/atm/cam/test/system/tests_posttag_bluefire
| models/atm/cam/test/system/tests_posttag_bluevista
| models/atm/cam/test/system/TCB.sh
| models/atm/cam/test/system/TBR.sh
| models/atm/cam/test/system/TER.sh
| models/atm/cam/test/system/TSM.sh
| models/atm/cam/bld/configure
| models/atm/cam/bld/build-namelist
| models/atm/cam/bld/config_definition.xml
+---> configuration/testing changes for new cice
models/atm/cam/bld/Makefile.in
changed -qautoinit flag for ibm from 7FF7FFFF7FF7FFFF
to 7FF7FFFF. Fixed compilation warnings for
uninitialized data
models/atm/cam/src/control/history_defaults.F90
Added new outfields for scam with MG microphysics
models/atm/cam/src/control/scamMod.F90
Added new variables for scam with MG microphysics
models/atm/cam/src/chemistry/trop_mozart/mo_regrider.F90
fix scam for running with MG microphysics
models/atm/cam/src/chemistry/trop_mozart/tracer_data.F90
fix scam for running with new chem
models/atm/cam/src/dynamics/eul/iop.F90
fix scam for running with MG microphysics
models/atm/cam/src/dynamics/eul/tfilt_massfix.F90
fix scam for running with MG microphysics
SVN_EXTERNAL_DIRECTORIES
updated to latest drv and cice code
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:None
bangkok/lf95:None
CAM tag used for the baseline comparison tests if different than previous
tag:No baseline since new cice changes answers
Summarize any changes to answers, i.e.,
- what code configurations:
New CICE code
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
new climate
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
Atmospheric Diagnostics at
- source tag
cice4_0_20080822
- platform/compilers:
bluefire
bangkok
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
/DBAILEY/csm/f40.033
URL for AMWG diagnostics output used to validate new climate:
http://www.cgd.ucar.edu/cms/mvr/f40.033/f40.033-f40.017/
===============================================================
===============================================================
Tag name: cam3_6_04
Originator(s): eaton
Date: Wed Aug 20 11:00:34 MDT 2008
One-line Summary: Makefile mods for AIX; mods for RRTMG
Purpose of changes:
. Change the default optimization for AIX from -O3 to -O2. This is to
address an irreproducibility problem on bluefire resulting from compiling
CLM at -O3.
*** NOTE *** The change from -O3 to -O2 has triggered a new xlf11
compiler problem. CAM encounters a seg fault when built *without*
threading enabled, i.e., either serial or pure mpi builds. The
workaround is to build for hybrid parallelization, which is the
default way to build for the IBM power platforms.
. Other Makefile changes for AIX debug mode:
- change setting of -qinitauto to a double precision signalling NAN
- turn off the optimization that is implied by -qsmp by setting
-qsmp=omp:noopt for threaded configs.
. Various fixes for RRTMG code:
- bugfix in aerosol optical properties passed to rrtmg
- change index of the visible band from 9 to 10
. Note that we are not yet ready to start doing tests of RRTMG w/ aerosols
due to issues with the latest physprop datasets. Will resolve this in an
upcoming commit.
. Change the cam*.cpl7.template files to force CCSM builds of CAM to enable
threading.
Bugs fixed (include bugzilla ID):
. aerosol optics for rrtmg
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do:
models/atm/cam/src/physics/rrtmg/aer_rad_props.F90
. version of aer_rad_props for rrtmg. currently the only difference is
that the physics/cam/aer_rad_props.F90 still contains the incorrect
expression for relative humidity that was used in cam3.
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/cam.cpl7.template
models/atm/cam/bld/camcsim.cpl7.template
models/atm/cam/bld/camdom.cpl7.template
models/atm/cam/bld/camsom.cpl7.template
. always set THREAD = TRUE. This is to deal with a problem that has
appeared on both bluevista and bluefire when trying to run in serial or
pure mpi modes after we reduced the optimization from -O3 to -O2.
models/atm/cam/bld/Makefile.in
. Change default optimization for AIX to -O2 (was -O3)
. change value of -qinitauto (AIX) to a double precision signalling NAN.
. modify AIX section to avoid having -O2 turned on implicitly by -qsmp in
debug mode
models/atm/cam/src/physics/cam/aer_rad_props.F90
. bugfix for diagnostic calculation of visible optical depth
models/atm/cam/src/physics/cam/cldwat2m.F90
. add some log output for the aerosol physical properties
. modify smcritfactor calculation to allow for hygro_aer=0
models/atm/cam/src/physics/rrtmg/radconstants.F90
. change idxVIS to 10 (from 9)
models/atm/cam/src/physics/rrtmg/radiation.F90
. remove unused calculation of relative humidity
models/atm/cam/src/physics/rrtmg/radsw.F90
. bugfix for aerosol properties that are passed to rrtmg_sw.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
004 bl132 TBL.sh e32sdh ghgrmp 9s .................................FAIL! rc= 7
006 bl133 TBL.sh e32pdh aqpgro 3s .................................FAIL! rc= 7
011 bl151 TBL.sh e64h outfrq24h 2d ................................FAIL! rc= 7
015 bl331 TBL.sh f4gdh co2rmp 9s ..................................FAIL! rc= 7
019 bl332 TBL.sh f4sdh ghgrmp 9s ..................................FAIL! rc= 7
021 bl333 TBL.sh f4pdh aqpgro 3s ..................................FAIL! rc= 7
030 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7
033 bl337 TBL.sh f4dh fv2d_8tsk 9s ................................FAIL! rc= 7
041 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7
044 bl374 TBL.sh fm1.9dh outfrq3s 9s ..............................FAIL! rc= 7
047 bl355 TBL.sh fmo2dh off2x2.5 9s ...............................FAIL! rc= 7
051 bl381 TBL.sh fma1.9h outfrq3s+1870_prog_aero 9s ...............FAIL! rc= 7
055 bl382 TBL.sh fmgpa1.9dh outfrq3s+1870_control 9s ..............FAIL! rc= 7
The baseline tests fail due the change in compiler optimization. This is
true for tests run in both debug and non-debug modes due to implicit
optimization that is turned on by -qsmp
bangkok/lf95: All PASS.
CAM tag used for the baseline comparison tests if different than previous
tag: cam3_6_01
Summarize any changes to answers: roundoff introduced due to change from
-O3 to -O2.
===============================================================
===============================================================
Tag name: cam3_6_03
Originator(s): eaton
Date: Tue Aug 12 18:51:26 MDT 2008
One-line Summary: allow setting fcrit2 via namelist; add aerosol optical
depth diagnostics to history output
Purpose of changes:
. Provide a new namelist variable fcrit2 to allow setting the value in
gw_drag. The default is still fcrit2=1.0 which is the change made in
cam3_5_50.
. Add diagnostic output of the column integrated optical depth in the
visible band for individual aerosol constituents that affect the climate
simulation. The names in the history file are of the form ODV_xxx where
xxx is the constituent name as set by the parameterization that
determines its mass distribution. xxx is also the name that is used in
the rad_climate namelist variable.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist:
. add namelist variable fcrit2 to namelist group gw_drag_nl
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/namelist_files/namelist_definition.xml
. add variable fcrit2 to namelist group gw_drag_nl
models/atm/cam/src/control/runtime_opts.F90
. add call to new gw_drag_readnl routine
models/atm/cam/src/physics/cam/aer_rad_props.F90
. add diagnostic output of optical depth in the visible band for individual
aerosol constituents in the climate list. The diagnostics lists are not
yet available.
models/atm/cam/src/physics/cam/gw_drag.F90
. add gw_drag_readnl routine to read namelist group gw_drag_nl which
contains the variable fcrit2. Remove hardcoded setting of fcrit2 from
the gw_inti routine.
models/atm/cam/src/physics/cam/rad_constituents.F90
. add optional arg aernames to the rad_cnst_get_clim_info subroutine to
give access to the aerosol names in either a climate or diagnostic list.
models/atm/cam/src/physics/waccm/gw_drag.F90
. changes to waccm version of gw_drag are necessary so that the modules
have the same interfaces and can be swapped at build time
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS
bangkok/lf95: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag: cam3_6_01
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam3_6_02
Originator(s): eaton
Date: Thu Aug 7 08:29:12 MDT 2008
One-line Summary: Enable building standalone CAM with CICE
Purpose of changes:
. In the near future the "prescribed" version of CICE will replace
CSIM4 as the default sea ice model for the standalone CAM configuration.
This commit is a step along that path; CICE has not yet been made the
default. It is currently enabled by giving the argument "-ice cice" to
configure.
. Unlike CSIM4, CICE uses its own grid decomposition which is independent
of CAM's. This decomposition is specified by parameters that are set at
build time, hence they are specified to configure. There are two ways to
do this:
- Specify the number of tasks and/or threads for hybrid, pure mpi, or
pure omp parallelism using the new configure options -ntasks, and
-nthreads. This will allow a default decomposition to be determined.
- Specify the CICE decomposition explicitly via the new configure options
-cice_bsizex, -cice_bsizey, and -cice_maxblocks.
. Update the CICE external to cice4_0_20080718
. Modify the Makefile to allow faster compilation of the RRTMG modules
rrtmg_lw_k_g.f90 and rrtmg_sw_k_g.f90. Longer term solution is to read the
data that's stored in these modules from a file.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system:
. changes to configure to support building w/ CICE
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
SVN_EXTERNAL_DIRECTORIES
. update CICE to cice4_0_20080718
. add external for stub sice
models/atm/cam/bld/configure
. add arguments -cice_bsizex, -cice_bsizey, -cice_maxblocks.
These are used to explicitly set the CICE decomposition. Note that they
must be consistent with the number of tasks and threads used to run the
executable.
. add arguments -ntasks and -nthreads. If these are both specified then a
script provided by CICE will be used to generate a default CICE
decomposition.
Setting -ntasks > 0 implies -spmd, and setting -nthreads > 0 implies
-smp. Serial mode is still set using "-nospmd -nosmp".
. add cice_decomptype to the config definition. build-namelist needs access to
it to call the CICE build-namelist
models/atm/cam/bld/config_definition.xml
. add cice_decomptype, ntasks, and nthreads
models/atm/cam/bld/build-namelist
. use the CICE build-namelist
. add -cice_nl argument as a way to pass values to the CICE build-namelist
models/atm/cam/bld/Makefile.in
. add special rules to AIX section to remove optimization from the
compilation of rrtmg_lw_k_g.f90 and rrtmg_sw_k_g.f90
models/atm/cam/src/physics/cam/cldwat2m.F90
. change names in outfld calls for number concentrations of aerosols to
avoid a name clash with the mixing ratios output by chemistry.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS
bangkok/lf95: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam3_6_01
Originator(s): mvr
Date: Wed Aug 6 2008
One-line Summary: fix to runtime bug with pgi compilers; began replacing
unit 6 flush statements with those using a variable unit number; fix to
ensure code dependencies are regenerated with each make
Purpose of changes:
any configuration other than adiabatic or ideal physics would die at runtime
with a segfault in the radiation code, when using the pgi compilers on jaguar
(version 7.1-6)...this was introduced in cam3_5_54
occasionally, messages from cam were not making it to log files when the coupled
model would die at runtime...this was due to flush statements improperly using unit 6
bug 787 describes a problem where code could change between makes where dependencies
may be altered, but make does not regenerate the Depends file
Bugs fixed (include bugzilla ID): 793 (partial), 787 (partial)
Describe any changes made to build system:
mod to Makefile.in to ensure Depends file is regenerated with each make
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/run-ibm.csh
- small bug fix to setenv statement with syntax error
M models/atm/cam/src/control/cam_comp.F90
- flush calls now using iulog rather than unit 6
M models/atm/cam/src/physics/cam/rad_constituents.F90
- fix to runtime bug (segfault) caught with pgi compilers
M models/atm/cam/bld/Makefile.in
- fix to ensure Depends file is recreated with each make
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS:
bangkok/lf95: All PASS:
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: b4b
===============================================================
===============================================================
Tag name: cam3_6
Originator(s): eaton, andrew
Date: Fri Aug 1 11:27:59 MDT 2008
One-line Summary: make MG microphysics the default
Purpose of changes:
. MG microphysics is now the default. This is controlled by the new
"-microphys" argument to configure. The valid values are "rk" (Rasch and
Kristjansson), and "mg" (Morrison and Gettelman two moment scheme). "mg"
is the default.
** Note 1 ** This default needs to implemented in configure rather than
build-namelist because it affects the total number of
advected constituents which is a build time setting.
** Note 2 ** Since configure now knows what microphysics scheme is being
used, build-namelist is able to automatically set the
correct value of the namelist variable microp_scheme. The
user no longer needs to set that variable.
. Add bugfix to MG microphysics code (from Andrew Gettelman).
. A new argument, -nadv_tt, has been added to configure to specify the
number of advected test tracers. Previously this was done by using the
-nadv argument which specifies the total number of advected tracers.
That way of doing things resulted in having to redefine all the
regression tests that include test tracers because the MG microphysics
adds 2 advected constituents. Deconvolving the specification of test
tracers and total advected constituents makes the test specifications
more robust, and provides an easier to use interface.
. The regression tests have been updated to use the new -nadv_tt argument
to configure for specifying the number of advected test tracers. Also,
the special test for MG microphysics has been replaced by a test for the
cam3.5 physics.
Bugs fixed (include bugzilla ID):
. Add bugfix to MG microphysics code (from Andrew Gettelman).
Describe any changes made to build system:
. configure has 2 new arguments: -microphys and -nadv_tt
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes:
. MG microphysics is more expensive than RK. Don't have benchmarks yet.
Code reviewed by: eaton
List all subroutines eliminated:
models/atm/cam/test/system/config_files/e8c8mdm
. renamed e8t5mdm
models/atm/cam/test/system/config_files/f10c8mdm
. renamed f10t5mdm
models/atm/cam/test/system/config_files/s8c8mdm
. renamed s8t5mdm
models/atm/cam/test/system/config_files/f10c5dm
. renamed f10c3dm
models/atm/cam/test/system/nl_files/microp_mg
. special test for MG microphysics no longer needed
List all subroutines added and what they do:
models/atm/cam/test/system/config_files/e8t5mdm
models/atm/cam/test/system/config_files/f10t5mdm
models/atm/cam/test/system/config_files/s8t5mdm
. these files are for tests using test tracers. The -nadv settings have
all been replaced by settings for -nadv_tt
models/atm/cam/test/system/config_files/f10c3dm
. this file is used to test the cam3.5 physics, which is turned on by the
setting "-chem none -microphys rk"
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/build-namelist
. add default for microp_scheme based on value from configure
. set tracers_flag=.true. if test tracers requested via configure
models/atm/cam/bld/config_defaults_trop_mozart.xml
models/atm/cam/bld/config_defaults_trop_mozart_aero.xml
models/atm/cam/bld/config_defaults_trop_mozart_ghg_paero.xml
models/atm/cam/bld/config_defaults_trop_mozart_prescribed_aero.xml
models/atm/cam/bld/config_defaults_waccm_ghg.xml
models/atm/cam/bld/config_defaults_waccm_mozart.xml
. remove default setting for nadv -- responsibility for setting this
belongs in configure.
models/atm/cam/bld/config_definition.xml
. add parameter "microphys" with valid values "rk" or "mg", and set the
default value to "mg"
. remove default value for nadv -- move all responsibility for setting
this into configure
. add parameter "nadv_tt" to specify number of advected test tracers
models/atm/cam/bld/configure
. modify how nadv values are set depending on the specific chemistry and
microphysics packages
. add commandline args for the new microphys and nadv_tt parameters
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. add values for microp_scheme
models/atm/cam/doc/ChangeLog_template
. blueice changed to bluefire
models/atm/cam/src/physics/cam/cldwat2m.F90
. one line bugfix from Andrew Gettelman
models/atm/cam/src/physics/cam/phys_control.F90
. Remove the default setting for microp_scheme_default. It is now the
responsibility of configure to set this default.
models/atm/cam/src/physics/cam/stratiform.F90
. Add check that valid value of microp_scheme has been specified.
models/atm/cam/test/system/config_files/e64bfbiop
models/atm/cam/test/system/config_files/scm64bfbiop
. add "-microphys rk" for scam tests which are currently failing when the
MG microphysics is enabled.
models/atm/cam/test/system/input_tests_master
. update test definitions to account for changed config file names
. replace special MG microphysics test with a test for the cam3.5 physics.
currently this sets "-chem none -microphysics rk"
models/atm/cam/test/system/nl_files/ttrac
models/atm/cam/test/system/nl_files/ttrac_lb1
models/atm/cam/test/system/nl_files/ttrac_lb2
models/atm/cam/test/system/nl_files/ttrac_lb3
. remove tracers_flag -- now set by build-namelist
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except:
004 bl132 TBL.sh e32sdh ghgrmp 9s .................................FAIL! rc= 7
006 bl133 TBL.sh e32pdh aqpgro 3s .................................FAIL! rc= 7
011 bl151 TBL.sh e64h outfrq24h 2d ................................FAIL! rc= 7
015 bl331 TBL.sh f4gdh co2rmp 9s ..................................FAIL! rc= 7
019 bl332 TBL.sh f4sdh ghgrmp 9s ..................................FAIL! rc= 7
021 bl333 TBL.sh f4pdh aqpgro 3s ..................................FAIL! rc= 7
030 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7
033 bl337 TBL.sh f4dh fv2d_8tsk 9s ................................FAIL! rc= 7
037 bl371 TBL.sh f1.9dm fvvp_lb2 9s ...............................FAIL! rc= 7
041 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7
044 bl374 TBL.sh fm1.9dh outfrq3s 9s ..............................FAIL! rc= 7
047 bl355 TBL.sh fmo2dh off2x2.5 9s ...............................FAIL! rc= 7
051 bl381 TBL.sh fma1.9h outfrq3s+1870_prog_aero 9s ...............FAIL! rc= 7
055 bl382 TBL.sh fmgpa1.9dh outfrq3s+1870_control 9s ..............FAIL! rc= 7
bangkok/lf95: All PASS except:
004 bl111 TBL.sh e8t5mdm ttrac 9s .................................FAIL! rc= 5
008 bl112 TBL.sh e8sdm ghgrmp 9s ..................................FAIL! rc= 7
010 bl113 TBL.sh e8pdm aqpgro 3s ..................................FAIL! rc= 7
014 bl114 TBL.sh e8dm co2rmp 9s ...................................FAIL! rc= 7
019 bl153 TBL.sh e64m outfrq24h 2d ................................FAIL! rc= 7
023 bl311 TBL.sh f10t5mdm ttrac 9s ................................FAIL! rc= 5
031 bl312 TBL.sh f10sdm ghgrmp 9s .................................FAIL! rc= 7
033 bl313 TBL.sh f10pdm aqpgro 3s .................................FAIL! rc= 7
036 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 7
039 bl315 TBL.sh f10c3dm outfrq3s 9s ..............................FAIL! rc= 5
042 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7
045 bl377 TBL.sh f1.9m outfrq24h 2d ...............................FAIL! rc= 7
049 bl511 TBL.sh s8t5mdm ttrac 9s .................................FAIL! rc= 5
053 bl512 TBL.sh s8sdm ghgrmp 9s ..................................FAIL! rc= 7
055 bl513 TBL.sh s8pdm aqpgro 3s ..................................FAIL! rc= 7
057 bl553 TBL.sh s64m outfrq24h 2d ................................FAIL! rc= 7
All baseline tests fail except for the adiabatic and ideal physics tests.
Some of the failures above are due to changing the names of config files.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: new climate in default model due to
changing default microphysics to MG. The answers for MG also change from
the baseline due to the bug fix.
===============================================================
===============================================================
Tag name: cam3_5_55
Originator(s): eaton, santos, rneale
Date: Wed Jul 30 11:22:12 MDT 2008
One-line Summary: fix compilation problems; new default files for full
chem; mods to generate input dataset lists
Purpose of changes:
. Fixes for compilation problems (with PGI) encountered during post-tag
testing of previous tag.
. Add new default files so that build-namelist produces working namelists
for 1 and 2 degree full trop chemistry runs.
. Mods to allow CCSM scripts to automatically retrieve missing input
datasets from svn archive (from Sean Santos).
. Add log output to zm_conv to help diagnose ientropy errors (from Rich
Neale).
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets:
. new input datasets for 1 and 2 degree full trop chemistry configs.
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/build-namelist
models/atm/cam/bld/cam.cpl7.template
. Mods to allow CCSM scripts to automatically retrieve missing input
datasets from svn archive.
models/atm/cam/bld/namelist_files/namelist_defaults.xml
. add files needed for 1 and 2 deg full trop chem runs
models/atm/cam/src/chemistry/trop_mozart/tracer_data.F90
. Change dummy arg "order" in both read_2d_trc and read_3d_trc from assumed
shape to the correct explicit shape.
models/atm/cam/src/chemistry/trop_mozart/mo_gas_phase_chemdr.F90
. replace new "use share_const_mod" with constants from existing "use mo_constants"
models/atm/cam/src/physics/cam/zm_conv.F90
. add diagnostic output for ientropy errors
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except
044 bl374 TBL.sh fm1.9dh outfrq3s 9s ..............................FAIL! rc= 7
This baseline test fails because the test code is using a new initial file
that contains spun up chemistry.
bangkok/lf95: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam3_5_54
Originator(s): eaton, aconley, fvitt
Date: Thu Jul 24 10:48:30 MDT 2008
One-line Summary: Add new radiative constituent interfaces. Add RRTMG as
an option. CAM3.5 configuration is now the default.
Purpose of changes:
. The new radiative constituent interface has been added. This provides
for an explicit specification of which constituents are radiatively
active for both climate and diagnostic calculations. The namelist
variables rad_climate, rad_diag_1, rad_diag_2, ... rad_diag_10 are used
to specify the constituents used for the climate and up to ten diagnostic
calculations. Each of the namelist variables specifies which
constituents are used for the gas species, aerosol species, and a file
that contains the optical and physical properties for each aerosol species.
build-namelist constructs default values for rad_climate for each of the
chemistry options. There are no diagnostic calculations specified by
default. In this initial commit the diagnostic calculations for the CAM3
radiation code are not implemented. There is a partial implementation of
the diagnostics for the RRTMG radiation.
The functionality for controlling which aerosols were radiatively active
was previously located in the aerosol_intr module. The namelist
variables previously used for this purpose have been removed.
. The RRTMG radiation code is present as a build time option. It is
enabled by specifying the flag -rrtmg to configure. Implementing aerosol
and cloud optics for RRTMG is not yet complete.
The RRTMG code is accessed by CAM via an external definition of the
source tree directory models/atm/cam/src/physics/rrtmg/ext/
. MG microphysics now gets microphysical data for aerosols from the same
files that provide the optics data. This data was previously hardcoded
in the prescribed_aerosols module. That module has been removed.
. The 1990_control use case has been made the default configuration on the
trunk. This is the configuration used for CCSM3.5 1990 control
runs. The differences from the previous default configuration are:
- The CAM-Chem prescribed aerosols replace the CAM3 aerosol climatology.
- The GHG values (co2vmr, ch4vmr, n2ovmr, f11vmr, f12vmr) and scon are
from the 1990_control use case.
- The CAM-Chem prescribed ozone replaces the CAM3 ozone (from pcmdi).
- The CAM-Chem options are enabled by making the configure option
"-chem trop_mozart_prescribed_aero" the default chemistry mode.
. A new chemistry mode "-chem none" has been added to configure.
Specifying this option turns on the CAM3 prescribed aerosols and ozone
and makes them radiatively active. The functionality to support the CAM3
prescribed aerosols and ozone is implemented by the new cam3_aero_data
and cam3_ozone_data modules.
. Modifications to CAM-Chem code:
- The prescribed aerosol and ozone modules were modified to put the
interpolated values (in mass mixing ratio units) into the physics
buffer.
- The namelist variable input has been redistributed so that it's not
handled entirely by the chemistry module.
- The names of the prescribed aerosols were changed.
- Default value of xactive_drydep changed from .false. to .true.
- More flexible tracer_data module. Allow dimensions of the input data
to occur in any order rather than restrict to (lon,lat,lev,tim).
. SVN_EXTERNAL_DIRECTORIES:
- update to esmf_wrf_timemgr_080717
- add external rrtmg_080606
Bugs fixed (include bugzilla ID):
. The CAM-Chem code that rebins the 4-bin sea salt into 2 bins for the
CAM-RT optics was missing an initializer.
. The problem with the trop_mozart_ghg_paero mode running in debug mode
under xlf11 has been fixed. The problem was with date calculation code
in tracer_data which has been updated to use time_manager methods (and
required a mod to the esmf_wrf_timemgr code.
Describe any changes made to build system:
. add -rrtmg flag to configure to build CAM with RRTMG radiation
. modifications to make "trop_mozart_prescribed_aero" the default value of
the -chem argument to configure
Describe any changes made to the namelist:
. namelist variables removed
- rampyear_prescribed_sulfur
- scenario_prescribed_sulfur
- tauback
- aeroptics
- aero_feedback_carbon
- aero_feedback_dust
- aero_feedback_progsslt
- aero_feedback_sea_salt
- bndtvcarbonscale
- carscl, dustscl, ssltscl, sulscl, volcscl
- bgscl_rf, carscl_rf, dustscl_rf, ssltscl_rf, sulscl_rf, volcscl_rf
- prescribed_sulfur
- radforce
- scenario_carbon_scale
- use_cam_ghg_data, use_cam_ozone_data
- use_data_co2, use_data_o3
. namelist variables added
- cam3_ozone_data_on: enables the CAM3 prescribed ozone
- cam3_aero_data_on: enables the CAM3 prescribed aerosols
- rad_climate
- rad_diag_1, rad_diag_2, ..., rad_diag_10
- prescribed_aero_datapath
- prescribed_aero_file
- prescribed_aero_filelist
- prescribed_aero_rmfile
- prescribed_aero_specifier
- prescribed_aero_tod
- prescribed_aero_type
- prescribed_aero_ymd
- prescribed_ghg_datapath
- prescribed_ghg_file
- prescribed_ghg_filelist
- prescribed_ghg_rmfile
- prescribed_ghg_specifier
- prescribed_ghg_tod
- prescribed_ghg_type
- prescribed_ghg_ymd
. namelist variables modified
- bndtvaer: moved from group cam_inparm to cam3_aero_data_nl
- bndtvo: moved from group cam_inparm to cam3_ozone_data_nl
- ozncyc: moved from group cam_inparm to cam3_ozone_data_nl
List any changes to the defaults for the boundary datasets:
. the namelist_defaults.xml file has been modified to remove all the
resolution dependent start_ymd values. This was origianlly implemented
to ensure that a default initial file would be found without the user
needing to know the corresponding date. This feature was mainly for the
regression testing. But we now have the capability to specify
-ignore_ic_date or -ignore_ic_year to build-namelist which is a more
explicit and flexible way for the user to indicate that the exact date of
the initial file isn't important. The TBL.sh script uses the
-ignore_ic_date arg in its build-namelist call.
The result of this change is that some of our regression tests which
previously got a resolution dependent start_ymd that was different from
901 will now get the resolution independent default which has been
changed to 101 for consistency with the CCSM scripts.
Describe any substantial timing or memory changes: not checked
Code reviewed by: eaton
List all subroutines eliminated:
models/atm/cam/src/chemistry/trop_mozart/rad_cnst_data_interface.F90
. not needed. prescribed aerosols now put in physics buffer and
rad_constituents is responsible for knowing what to use for climate and
diagnostic calcs.
models/atm/cam/src/physics/cam/aer_optics.F90
models/atm/cam/src/physics/cam/aerosol_index.F90
models/atm/cam/src/physics/cam/aerosol_radiation_interface.F90
. aerosol optical calcs moved to new aer_rad_props module
models/atm/cam/src/physics/cam/carbonscales.F90
. functionality replaced by providing time series data for carbon aerosol
distributions.
models/atm/cam/src/physics/cam/ozone_data.F90
. This module has been renamed cam3_ozone_data.
models/atm/cam/src/physics/cam/prescribed_aerosols.F90
. This module has been renamed cam3_aero_data. All the functionality
related to determining the aerosol mass for both the climate and
diagnostic calculations now belongs in the rad_constituents module.
List all subroutines added and what they do:
models/atm/cam/src/chemistry/trop_mozart/prescribed_aero.F90
models/atm/cam/src/chemistry/trop_mozart/prescribed_ghg.F90
. New CAM-Chem code for prescribed aerosols and ozone. The prescribed
constituents are maintained in the physics buffer.
models/atm/cam/src/physics/cam/aer_rad_props.F90
. New module for computing aerosol radiative properties for both climate
and diagnostic calculations.
models/atm/cam/src/physics/cam/cam3_aero_data.F90
. Responsible for putting the CAM3 aerosol mixing ratios into the physics
buffer.
models/atm/cam/src/physics/cam/cam3_ozone_data.F90
. Responsible for putting the CAM3 ozone mixing ratios into the physics
buffer.
models/atm/cam/src/physics/cam/hirsbt.f90
models/atm/cam/src/physics/cam/hirsbtpar.f90
. Diagnostic calculations of TOA brightness temperatures for 7 TOVS/HIRS
channels and 4 TOVS/MSU channels.
models/atm/cam/src/physics/cam/phys_prop.F90
. responsible for maintaining the physical property data (includes optics)
for the aerosols.
models/atm/cam/src/physics/cam/radconstants.F90
. constants for CAM-RT
models/atm/cam/src/physics/cam/sslt_rebin.F90
. rebins 4 sea salt bins into 2 bins.
models/atm/cam/src/physics/rrtmg/ext/rrtmg_lw/README
models/atm/cam/src/physics/rrtmg/ext/rrtmg_lw/parrrtm.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_lw/rrlw_cld.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_lw/rrlw_con.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_lw/rrlw_kg01.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_lw/rrlw_kg02.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_lw/rrlw_kg03.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_lw/rrlw_kg04.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_lw/rrlw_kg05.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_lw/rrlw_kg06.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_lw/rrlw_kg07.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_lw/rrlw_kg08.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_lw/rrlw_kg09.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_lw/rrlw_kg10.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_lw/rrlw_kg11.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_lw/rrlw_kg12.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_lw/rrlw_kg13.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_lw/rrlw_kg14.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_lw/rrlw_kg15.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_lw/rrlw_kg16.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_lw/rrlw_ref.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_lw/rrlw_tbl.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_lw/rrlw_vsn.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_lw/rrlw_wvn.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_lw/rrtmg_lw_cldprop.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_lw/rrtmg_lw_init.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_lw/rrtmg_lw_k_g.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_lw/rrtmg_lw_rad.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_lw/rrtmg_lw_rtrn.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_lw/rrtmg_lw_rtrnmr.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_lw/rrtmg_lw_setcoef.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_lw/rrtmg_lw_taumol.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_mcica/mcica_random_numbers.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_mcica/mcica_subcol_gen_lw.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_mcica/mcica_subcol_gen_sw.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_mcica/rrtmg_lw_cldprmc.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_mcica/rrtmg_lw_rad.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_mcica/rrtmg_lw_rtrnmc.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_mcica/rrtmg_sw_cldprmc.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_mcica/rrtmg_sw_rad.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_mcica/rrtmg_sw_spcvmc.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_sw/README
models/atm/cam/src/physics/rrtmg/ext/rrtmg_sw/parrrsw.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_sw/rrsw_aer.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_sw/rrsw_cld.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_sw/rrsw_con.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_sw/rrsw_kg16.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_sw/rrsw_kg17.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_sw/rrsw_kg18.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_sw/rrsw_kg19.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_sw/rrsw_kg20.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_sw/rrsw_kg21.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_sw/rrsw_kg22.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_sw/rrsw_kg23.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_sw/rrsw_kg24.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_sw/rrsw_kg25.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_sw/rrsw_kg26.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_sw/rrsw_kg27.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_sw/rrsw_kg28.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_sw/rrsw_kg29.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_sw/rrsw_ref.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_sw/rrsw_tbl.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_sw/rrsw_vsn.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_sw/rrsw_wvn.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_sw/rrtmg_sw_cldprop.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_sw/rrtmg_sw_init.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_sw/rrtmg_sw_k_g.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_sw/rrtmg_sw_rad.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_sw/rrtmg_sw_reftra.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_sw/rrtmg_sw_setcoef.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_sw/rrtmg_sw_spcvrt.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_sw/rrtmg_sw_taumol.f90
models/atm/cam/src/physics/rrtmg/ext/rrtmg_sw/rrtmg_sw_vrtqdr.f90
. new RRTMG code. This code is pulled in via an external definition.
models/atm/cam/src/physics/rrtmg/radconstants.F90
models/atm/cam/src/physics/rrtmg/radiation.F90
models/atm/cam/src/physics/rrtmg/radlw.F90
models/atm/cam/src/physics/rrtmg/radsw.F90
. CAM specific interface code to the RRTMG radiation parameterization.
List all existing files that have been modified, and describe the changes:
SVN_EXTERNAL_DIRECTORIES
. update to esmf_wrf_timemgr_080717
. add rrtmg_080606
models/atm/cam/bld/Makefile.in
. add suffix recognition and rules for compiling .f90 files
models/atm/cam/bld/build-namelist
. add defaults for the scon and GHG values
. only put absems_data into namelist when CAM-RT is active
. turn on the cam3 prescribed ozone and aerosols for configurations that
don't yet work with the CAM-Chem versions (i.e., waccm)
. add appropriate defaults for the prescribed aerosols and ozone depending
on whether cam3 or CAM-Chem versions being used (or both).
. construct new rad_climate specifier depending on the chem option
. remove old specifiers of aerosol feedback
. update namelist variables used by CAM-Chem options
models/atm/cam/bld/config_defaults_trop_mozart.xml
models/atm/cam/bld/config_defaults_trop_mozart_aero.xml
models/atm/cam/bld/config_defaults_trop_mozart_ghg_paero.xml
models/atm/cam/bld/config_defaults_trop_mozart_prescribed_aero.xml
. remove specification of cppdefs="-DTROPCHEM"
models/atm/cam/bld/config_definition.xml
. add parameter rrtmg
. modify chem parameter to accept 'none' as a valid value, and to make
trop_mozart_prescribed_aero the default value.
models/atm/cam/bld/configure
. add -rrtmg option
. modify so that when the user specifies -nadv the value is not modified by
configure
. mod to supply the -DTROPCHEM cpp def when any trop_mozart* chemistry
option is specified. Then that specification may be removed from all the
config_defaults_trop_mozart*.xml files
models/atm/cam/bld/mkSrcfiles
. update to recognize .f90 suffix for files that need to be built. This
was needed for building the rrtmg code.
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
. add Jan 1 IC file for T5
. add new defaults for aerosol optics
. add defaults for new prescribed ozone (same as 1990_control use case)
. add defaults for new prescribed aerosol (same as 1990_control use case)
. add defaults for GHGs and scon (same as 1990_control use case)
models/atm/cam/bld/namelist_files/namelist_definition.xml
. changes to namelist as described above
models/atm/cam/bld/namelist_files/use_cases/1870_prog_aero.xml
. add GHG and scon settings consistent with 1870_control values used in
CCSM3.5 simulations
. update for namelist changes in the CAM-Chem code
models/atm/cam/bld/namelist_files/use_cases/1870-2000_control.xml
models/atm/cam/bld/namelist_files/use_cases/1870_control.xml
models/atm/cam/bld/namelist_files/use_cases/1990_control.xml
. update for namelist changes in the CAM-Chem code
models/atm/cam/src/chemistry/trop_mozart/chemistry.F90
. remove a bunch of namelist vars that have been moved to other modules
models/atm/cam/src/chemistry/trop_mozart/mo_chemini.F90
. remove namelist variables from arg list that are now managed by module
specific *_{default,set}opts methods. Call new *_init methods.
models/atm/cam/src/chemistry/trop_mozart/mo_drydep.F90
. Change default value of xactive_drydep from .false. to .true.
models/atm/cam/src/chemistry/trop_mozart/mo_airplane.F90
models/atm/cam/src/chemistry/trop_mozart/mo_jshort.F90
models/atm/cam/src/chemistry/trop_mozart/mo_lightning.F90
. access parameter rearth from physconst rather than dycore specific
dynconst module
models/atm/cam/src/chemistry/trop_mozart/mz_aerosols_intr.F90
. remove set_aerosol_from_mz_prescribed routine
models/atm/cam/src/chemistry/trop_mozart/prescribed_ozone.F90
. add *_{default,set}opts methods
. put ozone in physics buffer in mass mixing ratio units
models/atm/cam/src/chemistry/trop_mozart/tracer_data.F90
. allocate memory if field not already in pbuf
. allow dimensions of input data to occur in any order
. use date calculation routines from time_manager
models/atm/cam/src/chemistry/trop_mozart/tracer_cnst.F90
models/atm/cam/src/chemistry/trop_mozart/tracer_srcs.F90
. add *_{default,set}opts methods
models/atm/cam/src/chemistry/trop_mozart_ghg_paero/mo_sim_dat.F90
models/atm/cam/src/chemistry/trop_mozart_prescribed_aero/mo_sim_dat.F90
. change names of prescribed aerosols so that they can co-exist with the
prognostic aerosols.
models/atm/cam/src/chemistry/waccm_mozart/chemistry.F90
. move namelist variables for prescribed_ozone, tracer_cnst, tracer_srcs
modules into the *_{default,set}opts methods of those modules
models/atm/cam/src/chemistry/waccm_mozart/mo_sad.F90
. comment out sad_set_aer_mass routine which needs to be updated to use the
physics buffer
models/atm/cam/src/control/cam_comp.F90
. add pbuf as actual arg in phys_init call
models/atm/cam/src/control/filenames.F90
. remove public module data bndtvaer, bndtvcarbonscale, aeroptics
models/atm/cam/src/control/runtime_opts.F90
. move bndtvaer, ozncyc to cam3_aero_data
. move bndtvo to cam3_ozone_data
. remove aeroptics, bndtvcarbonscale
scenario_carbon_scale, scenario_prescribed_sulfur, rampyear_prescribed_sulfur,
prescribed_sulfur, tauback, radforce
aero_feedback_carbon, aero_feedback_sea_salt, aero_feedback_progsslt,
aero_feedback_dust
sulscl, carscl, ssltscl, dustscl, volcscl,
sulscl_rf, carscl_rf, ssltscl_rf, dustscl_rf, volcscl_rf
use_data_o3, use_data_co2, use_cam_ozone_data, use_cam_ghg_data
. added prescribed_ghg_specifier, prescribed_ghg_file, prescribed_ghg_filelist,
prescribed_ghg_datapath, prescribed_ghg_type, prescribed_ghg_rmfile,
prescribed_ghg_ymd, prescribed_ghg_tod,
prescribed_aero_specifier, prescribed_aero_file,
prescribed_aero_filelist, prescribed_aero_datapath, prescribed_aero_type,
prescribed_aero_rmfile, prescribed_aero_ymd, prescribed_aero_tod
. added *_{default,set}opts methods for tracer_cnst, tracer_srcs,
prescribed_ozone, prescribed_ghg, prescribed_aero modules
. removed *_{default,set}opts methods for ozone_data, rad_constituents,
aerosol_radiation_interface modules
. add methods to read module specific namelists for cam3_aero_data,
cam3_ozone_data, rad_constituents modules
models/atm/cam/src/physics/cam/advnce.F90
. update interfaces for cam3 prescribed aerosols and ozone
models/atm/cam/src/physics/cam/aerosol_intr.F90
. remove variables that control whether aerosols are radiatively active
. remove set_aerosol_from_prognostics
models/atm/cam/src/physics/cam/cldwat2m.F90
. add addfld/add_default calls for existing outfld calls
. update rad_constituents interfaces
. comment output of aerosol number concentrations
models/atm/cam/src/physics/cam/phys_gmean.F90
models/atm/cam/src/dynamics/fv/mean_module.F90
models/atm/cam/src/dynamics/fv/d2a3dikj.F90
models/atm/cam/src/dynamics/fv/par_xsum.F90
models/atm/cam/src/dynamics/fv/p_d_adjust.F90
. comment out log messages about "fixed and floating point sums differ"
models/atm/cam/src/physics/cam/initindx.F90
. new register routines for cam3 aerosols and ozone
. new register routines for CAM-Chem prescribed ozone, aerosols, ghgs
. add log output after all pbuf fields registered
models/atm/cam/src/physics/cam/phys_buffer.F90
. new print method for log file info
. new query method pbuf_get_fld_name
models/atm/cam/src/physics/cam/physpkg.F90
. add pbuf to phys_init dummy args
. update init calls for cam3 aerosols and ozone
. new init calls for rad_constituents and aer_rad_props
. new init for the sea salt rebinning module
models/atm/cam/src/physics/cam/param_cldoptics.F90
models/atm/cam/src/physics/cam/pkg_cldoptics.F90
. add cloud optical depth calc
models/atm/cam/src/physics/cam/rad_constituents.F90
. implement new interface design that treats both gases and aerosols
models/atm/cam/src/physics/cam/radae.F90
. use radconstants
models/atm/cam/src/physics/cam/radiation.F90
. use radconstants
. remove calc of sat sp hum
. update rad_constituents interfaces
. new aer_rad_props interfaces
. temporarily remove diagnostic call and indirect call to radcswmx
models/atm/cam/src/physics/cam/radlw.F90
. use radconstants
. add aerosol optical depth arg to radclwmx
. add subroutine aer_trans_from_od
models/atm/cam/src/physics/cam/radsw.F90
. use radconstants
. remove rh arg from radcswmx
. add aerosol optical properties as args to radcswmx
models/atm/cam/src/physics/cam/stratiform.F90
. move addfld calls for MG microphysics quantities to cldwat2m.F90
. update the rad_constituent interfaces. they are only used by MG.
models/atm/cam/src/physics/cam/tphysbc.F90
. Add a call to sslt_rebin_adv to rebin the seasalt from 4 into 2 bins.
This is only need for the CAM-RT radiation code for which there are only
optical characterizations of 2 sea salt bins.
models/atm/cam/src/physics/cam/volcanicmass.F90
. register the volcanic aerosol in pbuf and use pbuf to store the
interpolated data
** this functionality is currently broken **
models/atm/cam/src/physics/waccm/nlte_lw.F90
. update to new rad_constituent interfaces
models/atm/cam/src/utils/repro_sum_mod.F90
. bug fix for non-spmd mode
models/atm/cam/src/utils/time_manager.F90
. add methods set_time_float_from_date and set_date_from_time_float
models/atm/cam/test/system/config_files/e64bfbiop
models/atm/cam/test/system/config_files/scm64bfbiop
. change configurations used to SCAM test to use "-chem none"
models/utils/esmf_wrf_timemgr/ESMF_TimeIntervalMod.F90
. enabled the d_ optional argument to ESMF_TimeIntervalSet
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire: All PASS except
004 bl132 TBL.sh e32sdh ghgrmp 9s .................................FAIL! rc= 7
006 bl133 TBL.sh e32pdh aqpgro 3s .................................FAIL! rc= 7
009 bl134 TBL.sh e32adh adia 9s ...................................FAIL! rc= 7
011 bl151 TBL.sh e64h outfrq24h 2d ................................FAIL! rc= 7
015 bl331 TBL.sh f4gdh co2rmp 9s ..................................FAIL! rc= 7
019 bl332 TBL.sh f4sdh ghgrmp 9s ..................................FAIL! rc= 7
021 bl333 TBL.sh f4pdh aqpgro 3s ..................................FAIL! rc= 7
024 bl334 TBL.sh f4adh adia 9s ....................................FAIL! rc= 7
027 bl335 TBL.sh f4idh idphys 9s ..................................FAIL! rc= 7
030 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7
033 bl337 TBL.sh f4dh fv2d_8tsk 9s ................................FAIL! rc= 7
037 bl371 TBL.sh f1.9dm fvvp_lb2 9s ...............................FAIL! rc= 7
041 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7
044 bl374 TBL.sh fm1.9dh outfrq3s 9s ..............................FAIL! rc= 7
047 bl355 TBL.sh fmo2dh off2x2.5 9s ...............................FAIL! rc= 7
051 bl381 TBL.sh fma1.9h outfrq3s+1870_prog_aero 9s ...............FAIL! rc= 7
055 bl382 TBL.sh fmgpa1.9dh outfrq3s+1870_control 9s ..............FAIL! rc= 5
bangkok/lf95: All PASS except
004 bl111 TBL.sh e8c8mdm ttrac 9s .................................FAIL! rc= 7
008 bl112 TBL.sh e8sdm ghgrmp 9s ..................................FAIL! rc= 7
010 bl113 TBL.sh e8pdm aqpgro 3s ..................................FAIL! rc= 7
014 bl114 TBL.sh e8dm co2rmp 9s ...................................FAIL! rc= 7
017 bl115 TBL.sh e8idm idphys 9s ..................................FAIL! rc= 7
019 bl153 TBL.sh e64m outfrq24h 2d ................................FAIL! rc= 7
023 bl311 TBL.sh f10c8mdm ttrac 9s ................................FAIL! rc= 7
031 bl312 TBL.sh f10sdm ghgrmp 9s .................................FAIL! rc= 7
033 bl313 TBL.sh f10pdm aqpgro 3s .................................FAIL! rc= 7
036 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 7
039 bl315 TBL.sh f10c5dm microp_mg 9s .............................FAIL! rc= 7
042 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7
045 bl377 TBL.sh f1.9m outfrq24h 2d ...............................FAIL! rc= 7
049 bl511 TBL.sh s8c8mdm ttrac 9s .................................FAIL! rc= 7
053 bl512 TBL.sh s8sdm ghgrmp 9s ..................................FAIL! rc= 7
055 bl513 TBL.sh s8pdm aqpgro 3s ..................................FAIL! rc= 7
057 bl553 TBL.sh s64m outfrq24h 2d ................................FAIL! rc= 7
These baseline tests are expected to fail due to changed answers. The
ideal physics and adiabatic configurations in tests bl115, bl334, and bl335
are actually BFB, but the test is failing due to the changes to the default
start_ymd values.
CAM tag used for the baseline comparison tests if different than previous
tag: cam3_5_52
Summarize any changes to answers:
Model configurations that are BFB with the baseline:
====================================================
. adiabatic and ideal physics
. full physics configurations in which the only radiatively active species
are the gases. If prescribed ozone is active, it must be the CAM3
ozone. The CAM3 ozone is enabled by setting cam3_ozone_data_on=.true. in
the namelist. The aerosols are turned off in the baseline code by
setting sulscl=dustscl=carscl=ssltscl=0.0 in the namelist.
This was verified for the -chem options: trop_mozart_prescribed_aero,
trop_mozart_ghg_paero, trop_mozart_aero, trop_mozart, waccm_ghg, and
waccm_mozart.
Model configurations that have only roundoff differences with the baseline:
===========================================================================
. full physics configurations in which the only radiatively active species
are the gases *and* the active ozone is the CAM-Chem prescribed ozone.
This version of ozone is the now the default on the trunk. The aerosols
are turned off in the baseline code by setting
sulscl=dustscl=carscl=ssltscl=0.0 in the namelist.
This was verified via a pergro test for the -chem option
trop_mozart_prescribed_aero.
. full physics configurations with CAM-Chem version of aerosols (prescribed
or prognostic).
In order to validate the roundoff diffs it is necessary to fix the bug in
the baseline code that rebins the 4-bin sea salt into 2 bins.
Configurations that were verified by pergro test are the -chem options:
trop_mozart_prescribed_aero and trop_mozart_ghg_paero.
Model configurations with larger than roundoff differences from the baseline:
=============================================================================
. full physics configurations using the CAM3 prescribed aerosols. The
issue here is that these aerosol mixing ratios were incorrectly treated
as moist mmr in CAM3 code. This has been fixed. The result is larger
than roundoff differences (but presumably the same climate) have been
introduced into these configurations. That includes the waccm_ghg and
waccm_mozart modes which aren't yet configured to use the new CAM-Chem
prescribed aerosols.
Validation with MG microphysics enabled:
========================================
. The validation of the default configuration + MG microphysics is
complicated by very large error growth when MG is enabled. Our standard
pergro test is not valid in this configuration. So to attempt to validate
MG I looked just at a 6 step run using the pergro/aqua_planet
configuration, and looked at the output from cprnc for all fields written
to the default history file. The roundoff size differences are introduced
at nstep=0, and it can be seen that all fields with differences show 12-16
digits of agreement on average. This isn't a stong test, but it did point
out problems that resulted in larger than roundoff size differences at
nstep=0. In particular, to get this test to pass it was necessary to make
a bugfix to the baseline code. In stratiform_tend the conversion of aerosol
mass to mmr made before calling the MG microphysics was using moist instead
of dry air mass.
===============================================================
===============================================================
Tag name: cam3_5_53
Originator(s): mvr, jedwards, dbailey, pworley
Date: 17 July 2008
One-line Summary: bug fix for omp-only mode; code workaround for pathscale
compiler problem; backed out timing library introduced in cam3_5_51; new
cice tag with prescribed ice similar to csim
Purpose of changes:
cam3_5_51 introduced a few things needing attention...omp-only mode needed
a bug fix and posttag testing with pathscale was disabled due to new code
that triggered a compiler problem...also, running with the ccsm scripts
showed a problem with the timing library introduced
a new cice tag became available where the prescribed ice looks similar to
csim...this is the next step on the path to replacing csim with cice
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/test_driver.sh
M models/atm/cam/bld/run-ibm.csh
- revised the list of ibm env vars (from jedwards)
M models/atm/cam/bld/cam.cpl7.template
- cleanup, removed sourcing of ccsm env files; use_case now set from ccsm env var
M models/atm/cam/bld/camsom.cpl7.template
M models/atm/cam/bld/camcsim.cpl7.template
M models/atm/cam/bld/camdom.cpl7.template
- cleanup, removed sourcing of ccsm env files
M models/atm/cam/src/utils/repro_sum_mod.F90
- code workaround for pathscale compiler problem which surfaced with cam3_5_51
M models/atm/cam/src/dynamics/fv/mean_module.F90
- bug fix for running in omp-only mode
M .
M SVN_EXTERNAL_DIRECTORIES
- new clm, cice, timing, and scripts tag
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
052 sm382 TSM.sh fmgpa1.9dh outfrq3s+1870_control 9s ..............FAIL! rc= 8
053 er382 TER.sh fmgpa1.9dh outfrq3s+1870_control 4+5s ............FAIL! rc= 5
054 br382 TBR.sh fmgpa1.9dh outfrq3s+1870_control 6+3s ............FAIL! rc= 5
055 bl382 TBL.sh fmgpa1.9dh outfrq3s+1870_control 9s ..............FAIL! rc= 4
- "382" tests are existing failures which first occurred with introduction of XLF
11.1 compilers on the IBM's, which is the default on bluevista and bluefire
bangkok/lf95:
all pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate): bfb
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
Tag name: cam3_5_52
Originator(s): pworley
Date: 11 July 2008
One-line Summary: use scalable reproducible sum logic in FV distributed sums
Purpose of changes: improve performance and memory scalability
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes: Substantial performance
improvement for runs with large numbers of tracers; modest improvements in
current FV production configurations.
Code reviewed by: Eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
M models/atm/cam/src/dynamics/fv/d2a3dikj.F90
M models/atm/cam/src/dynamics/fv/fv_prints.F90
M models/atm/cam/src/dynamics/fv/mean_module.F90
M models/atm/cam/src/dynamics/fv/p_d_adjust.F90
M models/atm/cam/src/dynamics/fv/par_xsum.F90
- Replaced existing logic based on "master computes" with the scalable,
reproducible algorithm implemented in repro_sum_module. If the relative
difference between the sum produced by the reproducible fixed-point
algorithm and by the nonreproducible floating-point algorithm is greater
than repro_sum_rel_diff_max (default 1.0e-13) and if repro_sum_recompute
is set equal to true (default .false.), then the sum is recomputed with
the original (nonscalable) algorithm. Since this is done on a field-by-field
basis, the memory requirements are less than in the original implementation,
but the cost is potentially much greater.
M models/atm/cam/src/dynamics/fv/FVperf_module.F90
- defined the grid argument in FVstartclock, FVstopclock, and
FVbarrierclock to be intent(in) instead of intent(inout) when
GEOS_MODE is NOT defined. This removes an incompatibility introduced by
using FVstart/stopclock in par_xsum.F90.
M models/atm/cam/src/dynamics/fv/te_map.F90
- in previous versions, par_xsum was not called when SPMD was not defined
or when nprxy_x = 1. To preserve reproducibility when using the repro_sum
routine, te_map was modified to call par_xsum in all instances.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
blueice:
015 bl331 TBL.sh f4gdh co2rmp 9s ..................................FAIL! rc= 7
019 bl332 TBL.sh f4sdh ghgrmp 9s ..................................FAIL! rc= 7
021 bl333 TBL.sh f4pdh aqpgro 3s ..................................FAIL! rc= 7
024 bl334 TBL.sh f4adh adia 9s ....................................FAIL! rc= 7
027 bl335 TBL.sh f4idh idphys 9s ..................................FAIL! rc= 7
030 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7
033 bl337 TBL.sh f4dh fv2d_8tsk 9s ................................FAIL! rc= 7
037 bl371 TBL.sh f1.9dm fvvp_lb2 9s ...............................FAIL! rc= 7
041 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7
044 bl374 TBL.sh fm1.9dh outfrq3s 9s ..............................FAIL! rc= 7
047 bl355 TBL.sh fmo2dh off2x2.5 9s ...............................FAIL! rc= 7
Use of repro_sum algorithm in FV distributed sums introduces a roundoff level
difference in the numerics. All FV comparisons with the baseline code fail.
051 bl381 TBL.sh fma1.9h outfrq3s+1870_prog_aero 9s ...............FAIL! rc= 7
052 sm382 TSM.sh fmgpa1.9dh outfrq3s+1870_control 9s ..............FAIL! rc= 8
053 er382 TER.sh fmgpa1.9dh outfrq3s+1870_control 4+5s ............FAIL! rc= 5
054 br382 TBR.sh fmgpa1.9dh outfrq3s+1870_control 6+3s ............FAIL! rc= 5
055 bl382 TBL.sh fmgpa1.9dh outfrq3s+1870_control 9s ..............FAIL! rc= 4
- "382" tests are existing failures which first occurred with introduction of XLF
11.1 compilers on the IBM's, which is the default on bluevista and bluefire
bangkok/lf95:
023 bl311 TBL.sh f10c8mdm ttrac 9s ................................FAIL! rc= 7
031 bl312 TBL.sh f10sdm ghgrmp 9s .................................FAIL! rc= 7
033 bl313 TBL.sh f10pdm aqpgro 3s .................................FAIL! rc= 7
036 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 7
039 bl315 TBL.sh f10c5dm microp_mg 9s .............................FAIL! rc= 7
042 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7
045 bl377 TBL.sh f1.9m outfrq24h 2d ...............................FAIL! rc= 7
Due to roundoff level difference in numerics, FV comparisons with the baseline
code fail.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: FV dycore
- what platforms/compilers: all
- nature of change (roundoff; larger than roundoff but same climate; new
climate): roundoff
If bitwise differences were observed, how did you show they were no worse
than roundoff? standard pergro test
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
Tag name: cam3_5_51
Originator(s): pworley
Date: 10 July 2008
One-line Summary: BGP support; new timing library; new
reproducible sum routine; new performance optimization options
in cd_core; ALTGATHER support; clean-up and bug fixes
Purpose of changes:
Bugs fixed (include bugzilla ID): omega diagnostic calculation
is now correct when geopktrans==1 (and 2)
Describe any changes made to build system: added support for BGP in
Makefile and in configure. Adding ALTGATHER to CPPFLAGS may
eliminate the gather failure on the Cray XT without requiring setting
MPI environment variables.
Describe any changes made to the namelist: eliminated
phys_float_repro_gmean, phys_fixed_repro_gmean, and
phys_nonrepro_gmean; replaced phys_rdiff_warning_gmean with
repro_sum_rel_diff_max; added repro_sum_recompute,
geopkblocks, and cd_core_swapm; added option 2 to geopktrans.
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes: performance improved
for certain problems and platforms by setting geopktrans=2 and
geopkblocks appropriately, and/or by setting cd_core_swapm to .true.
Code reviewed by: Eaton
List all subroutines eliminated:
in models/atm/cam/src/physics/cam/phys_gmean.F90
1) phys_gmean_defaultopts
2) phys_gmean_setopts
3) gmean_nonrepro
List all subroutines added and what they do:
in new file models/atm/cam/src/utils/repro_sum_defaultopts.F90
1) repro_sum_defaultopts
- get default runtime options (replacement for phys_gmean_defaultopts)
2) repro_sum_setopts
- set runtime options (replacement for phys_gmean_setopts)
3) repro_sum
- calculate both fixed-point scalable, reproducible distributed sums
and floating-point scalable, nonreproducible sums, returning
fixed-point results and relative difference between two algorithms.
in models/utils/pilgrim/mod_comm.F90
1) mp_sendirr_fc
- version of mp_sendirr with additional flow control
2) mp_sendirr_fc_r4
- version of mp_sendirr_r4 with additional flow control
3) mp_sendirr_fc_i4
- version of mp_sendirr_i4 with additional flow control
4) mp_swapmirr
- replacement for mp_sendirr/mp_recvirr that gives up
support for one-sided and MPI datatype options and communication/
computation overlap, but adds swap_comm protocol
options, improved communication ordering, and the ability to
transpose multiple fields simultaneously.
in models/atm/cam/control/wrap_mpi.F90
1) mpigathervr4
- version of mpigatherv for r4 data
2) mpigathervint
- version of mpigatherv for int data
in models/atm/cam/dynamics/fv/geopk.F90
1) geopk_d
- implementation of same calculation as geopk without requiring
transpose. Calculation is reproducible, and numerics are identical
to geopk. Performance is somewhat similar to geopk16.
List all existing files that have been modified, and describe the changes:
Timing library
Updated to include a more recent version of Jim Rosinski's
GPTL library, supporting global statistics, and improved control of
output
M SVN_EXTERNAL_DIRECTORIES
- updated to timing/trunk_tags/timing080629
M models/atm/cam/bld/Makefile.in
- added OpenMP-enabling flags to CFLAGS when SMP is defined
for all platforms where missing. Note that have tested only
on Bluefire, Bangkok, Cray XT4 (Jaguar), and IBM BG/P
ALTGATHER
Defining ALTGATHER replaces mpi_gather in mpigatherv with a point-to-point
implementation with additional flow control. It also calls new routines
mp_sendirr_fc, mp_sendirr_fc_r4, and mp_sendirr_fc_i4 with additional
flow control for gathers in FV I/O routines. This avoids many of the gather
failures on the Cray XT. These may become unecessary when PIO becomes
operational, but this approach is more robust than using MPI environment
variables to avoid the gather error.
M models/utils/pilgrim/mod_comm.F90
- added mp_sendirr_fc, mp_sendirr_fc_r4, mp_sendirr_fc_i4
M models/atm/cam/src/control/binary_io.F90
M models/atm/cam/src/physics/cam/phys_grid.F90
M models/atm/cam/src/control/cam_history.F90
- call mpigathervr4 and mpigathervint instead of mpigatherv when
gathering r4 and int data, respectively
M models/atm/cam/src/dynamics/fv/io_dist.F90
- if ALTGATHER defined, call mp_sendirr_fc,
mp_sendirr_fc_r4 and mp_sendirr_fc_i4 instead of
mp_sendirr, mp_sendirr_fc and mp_sendirr_fc, respectively
M models/atm/cam/src/control/wrap_mpi.F90
- added new routines mpigathervr4 and mpigathervint; if ALTGATHER
defined, use point-to-point implementations of the gather
with flow control in mpigatherv, mpigathervr4, and migathervint
instead of MPI_GATHERV
BG/P support:
M models/atm/cam/bld/Makefile.in
- added BGP section that works on ORNL BGP system; have not yet tried
on ANL BGP system
M models/atm/cam/bld/configure
- added bgp as -target_os option
M models/atm/cam/src/utils/abortutils.F90
- added BGP to existing BGL ifdef
M models/atm/cam/src/physics/cam/zm_conv_intr.F90
- BGP compiler bug workaround (having to do with an
optional parameter). Enabled only if BGP defined.
Reproducible distributed sum
Extracted fixed-precision algorithm from phys_gmean and put into
a new module. New algorithm can be used with any distributed sum, not
just a global mean. Both fixed-precision and floating-point algorithms
are now always computed, and the relative difference is
returned in an optional parameter. The error tolerance
repro_sum_rel_diff_max is used in the calling routine (phys_gmean here)
to decide whether to just write out a warning message or to also
recompute using the original "master computes" nonscalable algorithm.
The recalculation occurs in phys_gmean if the namelist parameter
repro_sum_recompute is set to .true.
M models/atm/cam/src/control/runtime_opts.F90
- eliminated phys_float_repro_gmean, phys_fixed_repro_gmean,
phys_nonrepro_gmean, and phys_rdiff_warning_gmean ; added
repro_sum_rel_diff_max and repro_sum_recompute
M models/atm/cam/src/physics/cam/phys_gmean.F90
- replaced in-place algorithm with call to repro_sum
- deleted public routines phys_gmean_defaultopts, phys_gmean_setopts,
and private routine gmean_nonrepro
M models/atm/cam/src/physics/cam/phys_grid.F90
- made nlcols public for use in phys_gmean
Alternative GEOPK routine
The calculation of geopotential to the kappa (geopk) is one of the
more expensive operators in the FV dycore when using a 2D domain
decomposition as it requires transposing from a YZ to an
XY decompostion and back again for each call. An existing
non-transpose-based alternative algorithm (geopk16) is not
reproducible unless quad precision is used, which is not
available on a number of platforms, such as the Cray XT.
A new optional algorithm, geopk_d, has been introduced that does
not require a transpose and is reproducible without requiring
quad precision. Performance using geopk_d is sometimes better
(and sometimes worse) than the other options. geopk_d
is enabled by setting the namelist parameter geopktrans
to 2. (Setting it to 0, the default, enbles geopk; setting it to 1
enables geopk16.) A tuning parameter geopkblocks is used with geopk_d
to define the amount of overlap to try to exploit in this
distributed algorithm. The default is 1 (no overlap), while the
optimal choice appears to be approximately 3 when using large
latitude decompositions. However, the optimum is dependent on problem
size, processor count, and platform. In the process of introducing
this option, it was discovered that the non-transpose-based options
were no longer working due to the reworking of the omega calculation.
This has been fixed.
M models/atm/cam/src/dynamics/fv/geopk.F90
- added geopk_d
M models/atm/cam/src/dynamics/fv/cd_core.F90
- added call to geopk_d and bug fix
M models/atm/cam/src/dynamics/fv/dyn_comp.F90
- distributed algorithm bug fix
M models/atm/cam/src/dynamics/fv/dynamics_vars.F90
M models/atm/cam/src/dynamics/fv/spmd_dyn.F90
- added support for geopk_d
M models/atm/cam/src/control/runtime_opts.F90
- added geopkblocks namelist parameter
M models/atm/cam/src/dynamics/eul/spmd_dyn.F90
M models/atm/cam/src/dynamics/homme/spmd_dyn.F90
M models/atm/cam/src/dynamics/sld/spmd_dyn.F90
- added geopkblocks to spmd_dyn_defaultopts and
spmd_dyn_setopts
Alternative transpose algorithm
A major source of communication overhead in FV when using
the 2D decomposition is the transposes between the YZ and XY
decompositions. The current Pilgrim/mod_comm implementation
supports both one-sided and two-sided MPI communication, and
both explicit buffer packing and MPI datatypes. Every effort
is made to overlap communication with computation by separating
the commands mp_sendirr and mp_recvirr. As an evaluation tool,
an alternative mechanism called mp_swapmirr has been introduced
that uses the swapm routine. mp_swapmirr gives up
the one-sided and MPI data type options and does not try to
overlap the communication. but it does enable
trying many different two-sided communication protocols, uses
a more efficient communication ordering, and has the ability
to transpose multiple fields simultaneously. This capability is
introduced initially into cd_core only, and is enabled by
setting the namelist parameter cd_core_swapm to .true.
Initial experiments indicate that mp_swapmirr is sometimes
more efficient. If this holds upon further investigation,
the relevant functionality will either be introduced
into the standard mod_comm routines or the new routine
will be used outside of cd_core as well.
M models/atm/cam/src/dynamics/fv/cd_core.F90
- added call to mp_swapmirr
M models/utils/pilgrim/mod_comm.F90
- added mp_swapmirr
M models/atm/cam/src/control/runtime_opts.F90
- added cd_core_swapm namelist parameter
M models/atm/cam/src/dynamics/fv/dyn_comp.F90
M models/atm/cam/src/dynamics/fv/spmd_dyn.F90
M models/atm/cam/src/dynamics/fv/dynamics_vars.F90
- added support for mp_swapmirr (cd_core_swapm)
M models/atm/cam/src/utils/spmd_utils.F90
- added communicator argument to swapm (so that could
be used in FV); added optional parameters to override
communication protocol defaults
- changed default protocol to mpi_isend/mpi_irevc
(from mpi_sendrecv) as a general performance optimization
M models/atm/cam/src/control/wrap_mpi.F90
- added communicator argument to swapm
M models/atm/cam/src/dynamics/eul/spmd_dyn.F90
M models/atm/cam/src/dynamics/homme/spmd_dyn.F90
M models/atm/cam/src/dynamics/sld/spmd_dyn.F90
- added cd_core_swapm to spmd_dyn_defaultopts and
spmd_dyn_setopts
MISCELLANEOUS
M models/atm/cam/src/utils/spmd_utils.F90
- changed default swapm MPI protocol to mpi_isend/mpi_irevc
(from mpi_sendrecv) as a general performance optimization
M models/atm/cam/src/physics/cam/phys_grid.F90
- changed message tags to differentiate from dycore message tags
M models/atm/cam/src/control/cam_comp.F90
- disabled output of spmdstats files unless SPMDSTATS defined
M models/atm/cam/src/dynamics/fv/stepon.F90
M models/atm/cam/src/dynamics/fv/io_dist.F90
M models/atm/cam/src/dynamics/fv/inidat.F90
M models/atm/cam/src/dynamics/fv/dynamics_vars.F90
M models/atm/cam/src/dynamics/fv/te_map.F90
M models/atm/cam/src/dynamics/fv/dryairm.F90
- deleted unnecessary use statements or unused elements
in use statements
- (for some) changed CCP_PRT_PREFIX definition
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
052 sm382 TSM.sh fmgpa1.9dh outfrq3s+1870_control 9s ..............FAIL! rc= 8
053 er382 TER.sh fmgpa1.9dh outfrq3s+1870_control 4+5s ............FAIL! rc= 5
054 br382 TBR.sh fmgpa1.9dh outfrq3s+1870_control 6+3s ............FAIL! rc= 5
055 bl382 TBL.sh fmgpa1.9dh outfrq3s+1870_control 9s ..............FAIL! rc= 4
- "382" tests are existing failures which first occurred with introduction of XLF
11.1 compilers on the IBM's, which is the default on bluevista and bluefire
bangkok/lf95: all pass
jaguar: all pass (using posttag tests)
** ran tests both with and without cd_core_swapm = .true and
geoptrans = 2, with identical results.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
Tag name: cam3_5_50
Originator(s): mvr,jwolfe,pel,jrichter,jedwards
Date: 24 June 2008
One-line Summary: change to critical froude number; fix to fv polar
filter; fix for memory bug in ccsm bgc runs; processor binding included
with mpirun calls for bluefire in test and template scripts
Purpose of changes:
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: none
Describe any changes made to the namelist:
-Implement fft_flt flag to switch between the FFT filter always and a
combination FFT/algebraic filter. Make default FFT always.
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
D models/atm/cam/test/system/tests_posttag_blueice
D models/atm/cam/test/system/tests_pretag_blueice
-removed default test lists for decommissioned machine
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/TCT.ccsm.sh
-test script modified to use ccsm scripts from cam sandbox rather than
external ccsm tag
M models/atm/cam/test/system/test_driver.sh
-top level test script modified to remove code for decommissioned machine,
blueice and updated aix environment variables for bluevista,bluefire
M models/atm/cam/test/system/tests_posttag_bluefire
M models/atm/cam/test/system/tests_posttag_bluevista
-modified sets of default pretag/posttag tests
M models/atm/cam/test/system/input_tests_master
-added test of cam using cice as ice model (via ccsm scripts)
M models/atm/cam/test/system/CAM_runcmnd.sh
-processor binding in mpirun.lsf calls; blueice cleanup
M models/atm/cam/bld/run-ibm.csh
-processor binding in mpirun.lsf call; modified list of aix env vars
M models/atm/cam/bld/namelist_files/namelist_definition.xml
-add definition for fft_flt. read from cam_inparm for now
M models/atm/cam/src/control/runtime_opts.F90
-read fft_flt from cam_inparm namelist
M models/atm/cam/src/control/atm_comp_mct.F90
-fix to bug which caused memory leak in bgc runs of ccsm
M models/atm/cam/src/control/fv_control_mod.F90
-Add fft_flt flag (namelist) to switch between FFT filter always and a
combination FFT/algebraic filter
M models/atm/cam/src/physics/cam/gw_drag.F90
-change fcrit2 from 0.5 to 1.0
M models/atm/cam/src/dynamics/fv/pft_module.F90
-Implement fft_flt flag to switch between the FFT filter always and a
combination FFT/algebraic filter. Make default FFT always.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
004 bl132 TBL.sh e32sdh ghgrmp 9s .................................FAIL! rc= 7
011 bl151 TBL.sh e64h outfrq24h 2d ................................FAIL! rc= 7
015 bl331 TBL.sh f4gdh co2rmp 9s ..................................FAIL! rc= 7
019 bl332 TBL.sh f4sdh ghgrmp 9s ..................................FAIL! rc= 7
021 bl333 TBL.sh f4pdh aqpgro 3s ..................................FAIL! rc= 7
024 bl334 TBL.sh f4adh adia 9s ....................................FAIL! rc= 7
027 bl335 TBL.sh f4idh idphys 9s ..................................FAIL! rc= 7
030 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7
033 bl337 TBL.sh f4dh fv2d_8tsk 9s ................................FAIL! rc= 7
037 bl371 TBL.sh f1.9dm fvvp_lb2 9s ...............................FAIL! rc= 7
041 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7
044 bl374 TBL.sh fm1.9dh outfrq3s 9s ..............................FAIL! rc= 7
047 bl355 TBL.sh fmo2dh off2x2.5 9s ...............................FAIL! rc= 7
051 bl381 TBL.sh fma1.9h outfrq3s+1870_prog_aero 9s ...............FAIL! rc= 7
052 sm382 TSM.sh fmgpa1.9dh outfrq3s+1870_control 9s ..............FAIL! rc= 8
053 er382 TER.sh fmgpa1.9dh outfrq3s+1870_control 4+5s ............FAIL! rc= 5
054 br382 TBR.sh fmgpa1.9dh outfrq3s+1870_control 6+3s ............FAIL! rc= 5
055 bl382 TBL.sh fmgpa1.9dh outfrq3s+1870_control 9s ..............FAIL! rc= 4
057 bl711 TBL.sh h5x8adm adia 9s ..................................FAIL! rc= 5
- baseline tests failed due to change to gravity wave constant and mods to the
fv polar filter
- "382" tests are existing failures which first occurred with introduction of XLF
11.1 compilers on the IBM's, which is the default on bluevista and bluefire
bangkok/lf95:
004 bl111 TBL.sh e8c8mdm ttrac 9s .................................FAIL! rc= 7
008 bl112 TBL.sh e8sdm ghgrmp 9s ..................................FAIL! rc= 7
014 bl114 TBL.sh e8dm co2rmp 9s ...................................FAIL! rc= 7
019 bl153 TBL.sh e64m outfrq24h 2d ................................FAIL! rc= 7
023 bl311 TBL.sh f10c8mdm ttrac 9s ................................FAIL! rc= 7
031 bl312 TBL.sh f10sdm ghgrmp 9s .................................FAIL! rc= 7
033 bl313 TBL.sh f10pdm aqpgro 3s .................................FAIL! rc= 7
036 bl314 TBL.sh wg10dm outfrq3s 9s ...............................FAIL! rc= 7
039 bl315 TBL.sh f10c5dm microp_mg 9s .............................FAIL! rc= 7
042 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7
045 bl377 TBL.sh f1.9m outfrq24h 2d ...............................FAIL! rc= 7
049 bl511 TBL.sh s8c8mdm ttrac 9s .................................FAIL! rc= 7
053 bl512 TBL.sh s8sdm ghgrmp 9s ..................................FAIL! rc= 7
057 bl553 TBL.sh s64m outfrq24h 2d ................................FAIL! rc= 7
- baseline tests failed due to change to gravity wave constant and mods to the
fv polar filter
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate): larger than roundoff
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
evaluations were assumed to have been done by the associated scientists
providing the source mods: Jaga Richter for the gravity wave change
and Peter Lauritzen for the fv polar filter change
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
Tag name: cam3_5_49
Originator(s): mvr,mvertens
Date: 12 June 2008
One-line Summary: added support for new machine bluefire; enabled
create_newcase (of the ccsm scripts) to be called from within cam;
added code work-around for older pgi compilers
Purpose of changes:
bluefire (the new IBM clustered system at NCAR based on the Power6 chip)
came online june 9th and blueice will be taken offline around june 16
we wanted the ability to do model runs from cam using the ccsm
scripts in order to benefit from the features these scripts offer, like:
swapping in other component models (like replacing csim for cice),
use of the ccsm run database to document model runs, etc.
fix that went into cam3_5_48 corrected problems for pgi and pathscale
compilers as well as ifort, but not older versions (6.1-6) of pgi used
on cgd machines bangkok and calgary; a code work-around was added to
enable posttag testing on those machines while we wait for a compiler
upgrade
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
List all subroutines added and what they do:
A models/atm/cam/test/system/tests_pretag_bluefire
A models/atm/cam/test/system/tests_posttag_bluefire
- default sets of tests for new machine bluefire
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/CAM_runcmnd.sh
M models/atm/cam/bld/run-ibm.csh
M models/atm/cam/test/system/test_driver.sh
- added support for new machine bluefire
M models/atm/cam/bld/cam.cpl7.template
- minor bug fixes
M models/atm/cam/src/physics/cam/tidal_diag.F90
- code work-around for older versions of pgi compiler
M .
M SVN_EXTERNAL_DIRECTORIES
- new clm, scripts, cice tags; now grabbing clm bld directory with checkouts
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluefire:
052 sm382 TSM.sh fmgpa1.9dh outfrq3s+1870_control 9s ..............FAIL! rc= 8
053 er382 TER.sh fmgpa1.9dh outfrq3s+1870_control 4+5s ............FAIL! rc= 5
054 br382 TBR.sh fmgpa1.9dh outfrq3s+1870_control 6+3s ............FAIL! rc= 5
055 bl382 TBL.sh fmgpa1.9dh outfrq3s+1870_control 9s ..............FAIL! rc= 4
- existing failures which first occurred with introduction of XLF 11.1 compilers
on the IBM's, which is the default on bluevista and bluefire
bangkok/lf95: all pass
calgary/pgi: just baselines failed (due to compiler bug in baseline tag)
**additionally ran the ccsm test ERS_D.f19_f19.F.blueice (passed)
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate): b4b
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
Tag name: cam3_5_48
Originator(s): Francis Vitt
Date: 3 June 2008
One-line Summary: changes for ifort compiler
Purpose of changes: to compile with ifort compiler
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: Brian Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/src/physics/cam/tidal_diag.F90
M models/atm/cam/src/chemistry/waccm_mozart/mo_apex.F90
M models/atm/cam/src/chemistry/waccm_mozart/apex_subs.F90
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
blueice: all pass
bangkok/lf95: all pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
cam3_5_47
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_5_47
Originator(s): Francis Vitt
Date: 29 May 2008
One-line Summary: WACCM bug fixes and enhancements.
Purpose of changes:
Enhance science in WACCM chemistry and fix a few bugs. The enhancements
and bug fixes include:
- Updated chemistry to JPL-2006.
- Added ClOy and BrOy family tracers (CLY and BRY) to improve
conservation of Cl and Br during advection.
- Made photolysis more flexible by using label tags in the
preprocessor input file.
- Don't allow wet removal above 300 mb in polar region.
- Added tidal diagnostics.
- Updates in aurora code.
- Added the ability to input surface fields that are lat, and time
dependent only in addition to fields that are lon, lat, and time
dependent.
- Added F107, F107a, Kp, Ap to history tapes.
- Added offline waccm driver capability.
- Added capability to use the Judith Lean solar irradiance data.
- Added the radiative feed back option for stratospheric sulfur aerosols
Bugs fixed (include bugzilla ID):
Fixed bug in O2 data used in radsw for waccm
Fixed inithist bug (bugzilla ID 715)
Fixed a bug in mo_jlong which was producing jagged photolysis
rate profiles. There was in error in the calculation of the
ozone ratios used in the interpolations of the radiative source
functions.
Fixed negative del_p bug in mo_jlong
Describe any changes made to build system:
Increased the number of advected tracers by two for waccm_mozart chemistry
Added build-namelist use_cases for waccm:
- waccm_1950_smax
- waccm_1950_smin
- waccm_1995_smax
- waccm_1995_smin
- waccm_1950_ramped
- waccm_1953_ramped_qbo
Describe any changes made to the namelist:
Added new namelist options for offline driver
- met_rlx_top
- met_rlx_bot
- met_max_rlx
- met_fix_mass
- met_shflx_name
- met_qflx_name
- met_shflx_factor
- met_qflx_factor
- solar_photons_file
- strat_aero_feedback
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Brian Eaton, Francis Vitt
List all subroutines eliminated:
D models/atm/cam/src/chemistry/waccm_mozart/mo_read_chm_sim.F90
List all subroutines added and what they do:
A models/atm/cam/bld/namelist_files/use_cases/waccm_1950_ramped.xml
A models/atm/cam/bld/namelist_files/use_cases/waccm_1950_smax.xml
A models/atm/cam/bld/namelist_files/use_cases/waccm_1953_ramped_qbo.xml
A models/atm/cam/bld/namelist_files/use_cases/waccm_1995_smin.xml
A models/atm/cam/bld/namelist_files/use_cases/waccm_1995_smax.xml
A models/atm/cam/bld/namelist_files/use_cases/waccm_1950_smin.xml
* new waccm use cases
A models/atm/cam/src/chemistry/waccm_mozart/clybry_fam.F90
* module added to manage setting the family tracer before advection
and to adjust the individual component species after advection
A models/atm/cam/src/physics/cam/tidal_diag.F90
* added tidal diagnostics for waccm
A models/atm/cam/src/physics/cam/solvar_interface.F90
A models/atm/cam/src/chemistry/waccm_mozart/solar_photons.F90
A models/atm/cam/src/chemistry/waccm_mozart/solvar_woods.F90
A models/atm/cam/src/chemistry/waccm_mozart/solvar_data.F90
* Files added for the use of the Lean solar irradiance data
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
* changed default namelist files
- xs_short_file
- xs_long_file
- rsf_file
M models/atm/cam/src/control/runtime_opts.F90
M models/atm/cam/bld/namelist_files/namelist_definition.xml
* added new namelist options
- met_rlx_top
- met_rlx_bot
- met_max_rlx
- met_fix_mass
- met_shflx_name
- met_qflx_name
- met_shflx_factor
- met_qflx_factor
- solar_photons_file
- strat_aero_feedback
M models/atm/cam/bld/namelist_files/use_cases/waccm_1995_climo.xml
* added default start_ymd setting
M models/atm/cam/bld/config_defaults_waccm_mozart.xml
* increased nadv to 65 for waccm (BRY and CLY added)
M models/atm/cam/src/control/cam_history.F90
* fixed init hist bug
* added time-dimensional only fields 'f107', 'f107a', 'ap', 'kp' to the
history files in the same manor as 'co2vmr'
M models/atm/cam/src/physics/cam/tphysbc.F90
* invocation added to adjust the individual BRY and CLY family
tracers after advection
M models/atm/cam/src/physics/cam/tphysac.F90
* invocation added to set CLY and BRY family tracers before advection
M models/atm/cam/src/physics/cam/cam_diagnostics.F90
* added tidal diagnostics for waccm
M models/atm/cam/src/chemistry/waccm_mozart/mo_jlong.F90
M models/atm/cam/src/chemistry/waccm_mozart/mo_jshort.F90
M models/atm/cam/src/chemistry/waccm_mozart/mo_waccm_photo.F90
* changes for tagged photolysis rates scheme
* changed max solar zenith angle to 97.01 degrees
* In subroutines jlong_hrates and jlong_photo corrected the
calculation of the O3 ratios used in the interpolations of
the radiative source functions
* Files changed for the use of the Lean irradiance data
M models/atm/cam/src/chemistry/waccm_mozart/mo_chemini.F90
M models/atm/cam/src/physics/cam/aerosol_radiation_interface.F90
M models/atm/cam/src/physics/cam/radsw.F90
M models/atm/cam/src/chemistry/waccm_mozart/woods.F90
M models/atm/cam/src/chemistry/waccm_mozart/solvar_interface.F90
* Files changed for the use of the Lean irradiance data
M models/atm/cam/src/chemistry/waccm_mozart/chem_surfvals.F90
* set O2 mmr data to 0.23143
* Added the ability to input surface fields that are lat, and time
dependent only in addition to fields that are lon, lat, and time
dependent. This is backwards compatible so the the "old" surface
data file can still be used.
M models/atm/cam/src/chemistry/waccm_mozart/chemistry.F90
* history addfld calls added
* call to advance Lean irradiance data added
M models/atm/cam/src/chemistry/waccm_mozart/mo_gas_phase_chemdr.F90
* history outfld calls added
* wetdep interface change
M models/atm/cam/src/chemistry/waccm_mozart/mo_wetdep.F90
* changed min pressure where wet removal is applied to
300 mbar poleward of 60 degrees.
M models/atm/cam/src/chemistry/waccm_mozart/efield.F90
* write to iulog
M models/atm/cam/src/chemistry/waccm_mozart/mo_aurora.F90
M models/atm/cam/src/chemistry/waccm_mozart/mag_parms.F90
* Changes in aurora code supplied by Liying Qian
M models/atm/cam/src/chemistry/trop_mozart/mo_lightning.F90
M models/atm/cam/src/chemistry/waccm_mozart/mo_strato_rates.F90
* change for new JPL 2006 chemistry mechanism
M models/atm/cam/src/chemistry/waccm_mozart/chem_mods.F90
M models/atm/cam/src/chemistry/waccm_mozart/mo_lu_solve.F90
M models/atm/cam/src/chemistry/waccm_mozart/mo_phtadj.F90
M models/atm/cam/src/chemistry/waccm_mozart/mo_nln_matrix.F90
M models/atm/cam/src/chemistry/waccm_mozart/mo_lu_factor.F90
M models/atm/cam/src/chemistry/waccm_mozart/mo_sim_dat.F90
M models/atm/cam/src/chemistry/waccm_mozart/mo_indprd.F90
M models/atm/cam/src/chemistry/waccm_mozart/mo_adjrxt.F90
M models/atm/cam/src/chemistry/waccm_mozart/mo_lin_matrix.F90
M models/atm/cam/src/chemistry/waccm_mozart/mo_setrxt.F90
M models/atm/cam/src/chemistry/waccm_mozart/mo_prod_loss.F90
M models/atm/cam/src/chemistry/waccm_mozart/mo_usrrxt.F90
M models/atm/cam/src/chemistry/waccm_mozart/m_het_id.F90
M models/atm/cam/src/chemistry/waccm_mozart/m_spc_id.F90
M models/atm/cam/src/chemistry/waccm_mozart/m_rxt_id.F90
* new code generated by the chemistry preprocessor
M models/atm/cam/src/chemistry/waccm_mozart/mo_sad.F90
* JF's change on lightning factor implementation
M models/atm/cam/src/dynamics/fv/inidat.F90
M models/atm/cam/src/dynamics/fv/dyn_comp.F90
M models/atm/cam/src/dynamics/fv/cd_core.F90
M models/atm/cam/src/dynamics/fv/sw_core.F90
M models/atm/cam/src/dynamics/fv/uv3s_update.F90
M models/atm/cam/src/dynamics/fv/metdata.F90
* changes for offline waccm driver
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
blueice:
030 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7
047 bl355 TBL.sh fmo2dh off2x2.5 9s ...............................FAIL! rc= 7
These are expected to fail due to changes in waccm chemistry and the offline driver.
bangkok/lf95:
all pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_5_46
Originator(s): mvr,pworley,robj,eaton,mvertens
Date: Tue May 27 MDT 2008
One-line Summary: test scripts updated for new jaguar hardware and compilers;
updated to new clm,mct and scripts externals to address bugs; other bug fixes
Purpose of changes:
jaguar returned to production with new quad-core processors - the test
scripts needed mods to work with the new hardware and pgi compilers (7.1-6)
a new mct tag was released that addressed several cam bugs - one with fv
decomp and phys_loadbalance=0 on jaguar and the other with excessive writes
from processors other than the master
a new clm tag addressed a bug in logic for do-albedo calculation
numerous ccsm tests were failing due to bugs in the scripts used by the coupled
system, including a mismatch in the grids specified for the eul dycore
new default aquaplanet initial files added to help in running pergro tests
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
- mods to reflect dir name change from ~models/lnd/clm2 to ~models/lnd/clm
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
- added default initial files for running in aquaplanet mode
Describe any substantial timing or memory changes:
Code reviewed by: myself, eaton
List all subroutines eliminated:
D models/atm/cam/bld/run-sgi.csh
- run template script for obsolete machine 'tempest' removed
List all subroutines added and what they do:
none
List all existing files that have been modified, and describe the changes:
M models/ocn/dom/ocn_comp.F90
M models/ocn/som/ocn_comp.F90
M models/ice/csim4/ice_comp.F90
M models/atm/cam/src/control/ncdio_atm.F90
- mods needed for fix to bug in fv decomp with phys_loadbalance=0 on jaguar
M models/atm/cam/test/system/test_driver.sh
- top level test script enabled for jaguar quad-core and 7.1-6 pgi compilers
M models/atm/cam/bld/configure
- mods to reflect dir name change from ~models/lnd/clm2 to ~models/lnd/clm
M models/atm/cam/bld/run-ibm.csh
M models/atm/cam/bld/run-pc.csh
M models/atm/cam/bld/run-lightning.csh
- run template scripts now default to validating archives to mss with file
comparison
M models/atm/cam/bld/camsom.cpl7.template
M models/atm/cam/bld/camcsim.cpl7.template
M models/atm/cam/bld/cam.cpl7.template
M models/atm/cam/bld/camdom.cpl7.template
- fixes for bugs when running as part of ccsm - including grid name mismatch
for eul dycore, initial file usage for start dates other than 1/1 and 9/1, and
other cleanup
M models/atm/cam/bld/build-namelist
- only check that user has specified start_ymd when ncdata has been
specified when building namelists for standalone cam scripts.
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- added default initial files for running in aquaplanet mode
M .
M SVN_EXTERNAL_DIRECTORIES
- new external code for clm,mct, and scripts
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
blueice:
all pass
bangkok/lf95:
004 bl511 TBL.sh s8c8mdm ttrac 9s .................................FAIL! rc= 7
008 bl512 TBL.sh s8sdm ghgrmp 9s ..................................FAIL! rc= 7
012 bl553 TBL.sh s64m outfrq24h 2d ................................FAIL! rc= 7
- sld baseline tests fail due to a doalb bug fix introduced in clm3_6_03
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate): bit-for-bit
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_5_45
Originator(s): mvr,dennis,jedwards
Date: Thurs May 8 MDT 2008
One-line Summary: code work-arounds for bluevista compiler problems;
new default initial conditions file for fv 0.47x0.63 and jan 1 start date
Purpose of changes:
we're still waiting for ibm to provide a fix for problems with the new
xlf compiler (version 11.1)...in the meantime, jedwards has provided
code work-arounds for two cam files to enable running with two of
the configurations known to trigger problems...
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
added default initial conditions file at fv 0.47x0.63 w/ start date of jan 1
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
M models/ocn/dom/ocn_comp.F90
M models/atm/cam/src/physics/cam/cam_diagnostics.F90
above files modified with code work-arounds for bluevista compiler problems
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
added default initial conditions file at fv 0.47x0.63 w/ start date of jan 1
M models/atm/cam/bld/cam.cpl7.template
mod to make use of new default initial conditions file for fv 0.47x0.63, jan 1
M models/atm/cam/test/system/test_driver.sh
minor fixes to top-level test script
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
blueice: ALL PASS
bangkok/lf95: ALL PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam3_5_44
Originator(s): eaton
Date: Wed May 7 14:44:38 MDT 2008
One-line Summary: build-namelist mods
Purpose of changes:
. fix bug in waccm_1995_climo use case
. update the Build::NamelistDefinition module with methods for creating
documentation from the definition xml file
. extend build-namelist to write the /modelio/ namelists used by components
to redirect their log output to a named file.
Bugs fixed (include bugzilla ID):
. fixed bug in waccm_1995_climo use case
Describe any changes made to build system: none
Describe any changes made to the namelist:
. add namelist variables used by components to redirect their log output to
named files. The new variables are
atm_logfile - Name of file that the atmosphere component log messages will
be written to. Default: none
atm_logfile_diro - Absolute pathname of directory that the file specified
by atm_logfile will be written to. Default: "./"
cpl_logfile - Name of file that the driver component log messages will
be written to. Default: none
cpl_logfile_diro - Absolute pathname of directory that the file specified
by cpl_logfile will be written to. Default: "./"
lnd_logfile - Name of file that the land component log messages will
be written to. Default: none
lnd_logfile_diro - Absolute pathname of directory that the file specified
by lnd_logfile will be written to. Default: "./"
Having each component write to it's own named log file is the default
when running with the ccsm4_alpha scripts. These new variables allow the
same flexibility when running with the standalone CAM scripts. The
default for the standalone CAM scripts remains that all components write
their log messages to stdout.
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/build-namelist
. extend to write /modelio/ namelists to component specific files
models/atm/cam/bld/namelist_files/namelist_definition.xml
. add new namelist variables for named logfiles
models/atm/cam/bld/namelist_files/use_cases/waccm_1995_climo.xml
. fix broken root element (terminating tag missing leading slash)
models/atm/cam/bld/perl5lib/Build/Namelist.pm
. convert embedded newlines to spaces before _parse_next method is called
by _split_namelist_value
models/atm/cam/bld/perl5lib/Build/NamelistDefaults.pm
. add check that the defaults xml file has a root element
models/atm/cam/bld/perl5lib/Build/NamelistDefinition.pm
. add methods for creating documentation from the definition xml file
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
blueice: All PASS
bangkok/lf95: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam3_5_43
Originator(s): mvertens,tcraig,mvr
Date: 080424
One-line Summary: update to externals list of ccsm4_0_alpha28;
removed all code related to permutation of grid points;
minor modifications to test scripts
Purpose of changes: permutation of grid points now handled within
mct code, so update to latest mct tag allowed removal of this code
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: mvertens,tcraig
List all subroutines eliminated:
D models/atm/cam/test/system/config_files/wg10dh
- configuration options file replaced with one for mpi-only
D models/atm/cam/test/system/tests_pretag_bluevista
- pretag testing will now be required on blueice
List all subroutines added and what they do:
A models/atm/cam/test/system/config_files/wg10dm
- new configuration options file for running test in mpi-only mode
A models/atm/cam/test/system/tests_posttag_bluevista
- bluevista testing will now be done posttag
A models/atm/cam/test/system/tests_pretag_blueice
- blueice will now be the required pretag platform for testing
List all existing files that have been modified, and describe the changes:
M models/ocn/dom/ocn_comp_mct.F90
M models/ocn/som/ocn_comp_mct.F90
M models/ice/csim4/ice_comp_mct.F90
M models/atm/cam/src/control/atm_comp_mct.F90
- removed code related to permutation of grid points
M models/atm/cam/test/system/test_driver.sh
- testing on jaguar will now use charge account specific to user; other
minor fixes
M models/atm/cam/test/system/TER.sh
- minor fix to restart test script when using fv 2d decomposition
M models/atm/cam/test/system/input_tests_master
- mod needed to have test run in mpi-only mode on bangkok/calgary/lightning
M models/atm/cam/doc/ChangeLog_template
- mod to reflect pretag testing now required on blueice rather than bluevista
M .
M SVN_EXTERNAL_DIRECTORIES
- updated to externals of ccsm4_0_alpha28
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
blueice:
ALL PASS
bangkok/lf95:
ALL PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change: b4b
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_5_42
Originator(s): eaton, mvertens
Date: Thu, 17 Apr 2008 14:42:08
One-line Summary: misc mods for build, trop_mozart_prescribed_aero, cam-som
Purpose of changes:
. build-namelist mods:
- when user specifies ncdata require that start_ymd also be specified
- don't write dom namelist when ocn=none. similarly for csim when
ice=none.
. configure mods:
- don't put utils/timing in filepath for build using ccsm scripts (they
build the timing lib separately)
. Modify behavior of "-chem trop_mozart_prescribed_aero". This option has
been changed to allow one to use the prescribed mozart ozone via the
"prescribed_ozone_file" namelist option. If "prescribed_ozone_file" is
not set then the default cam ozone data is used for the radiation
calculation, just as before.
. Modify DOM so that it writes TSOCN to CAM's initial file. This is needed
for CAM-SOM.
. Add ENDOFRUN option to the inithist namelist variable. This allows an
initial file to be written only at the end of the run.
:N.B. this option won't be fully functional until mods are made to
CLM and to the driver code.
Bugs fixed (include bugzilla ID):
. bugID 739 - don't write dom and csim namelists unless needed
Describe any changes made to build system: none
Describe any changes made to the namelist:
. add ENDOFRUN as possible value for inithist
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/build-namelist
. add check to require that the user specify start_ymd whenever ncdata has
been specified.
. don't write dom_inparm group to atm_in file if configuration ocn=none
. don't write csim_inparm group to atm_in file if configuration ice=none
models/atm/cam/bld/configure
. remove timing lib from filepath for ccsm build
models/atm/cam/bld/namelist_files/namelist_definition.xml
. make definition file consistent with new seq_timemgr_inparm and
seq_infodata_inparm namelist groups.
. changes to quite a few "catagory" attributes for documentation purposes
. add ENDOFRUN as valid value of inithist
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
models/atm/cam/src/chemistry/trop_mozart_prescribed_aero/chem_mods.F90
models/atm/cam/src/chemistry/trop_mozart_prescribed_aero/mo_sim_dat.F90
. allow use of CAM-Chem prescribed ozone in trop_mozart_prescribed_aero
configuration.
models/atm/cam/src/control/cam_history.F90
models/atm/cam/src/control/runtime_opts.F90
. add ENDOFRUN option to inithist
models/ocn/dom/ocn_comp.F90
. write TSOCN to cam initial file.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
blueice: All PASS except
015 bl331 TBL.sh f4gdh co2rmp 9s ..................................FAIL! rc= 7
calgary/lf95:
004 bl111 TBL.sh e8c8mdm ttrac 9s .................................FAIL! rc= 7
014 bl114 TBL.sh e8dm co2rmp 9s ...................................FAIL! rc= 7
023 bl311 TBL.sh f10c8mdm ttrac 9s ................................FAIL! rc= 7
The failed tests are all baseline comparisons that fail due to the
appearance of a new field (TSOCN) in the initial file.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: none
===============================================================
===============================================================
Tag name: cam3_5_41
Originator(s): mvr,tcraig
Date: 080401
One-line Summary:
mods to test scripts in preparation of the decommission of machine
tempest; enhanced post-tag testing on jaguar; fix to cleanup problems in
cpl history file
Purpose of changes:
machine tempest is being decommissioned on 4/5. the default tests
performed there as a requirement for committing to the cam developement
trunk had to be farmed out to other platforms. going forward, pretag
testing will be required on just two platforms, bangkok/lahey and
bluevista.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: myself,eaton
List all subroutines eliminated:
D models/atm/cam/test/system/config_files/fmgpa1.9h
D models/atm/cam/test/system/config_files/fmgpa1.9m
- removed obsolete configure options files
D models/atm/cam/test/system/tests_pretag_tempest
- machine tempest being decommissioned
D models/atm/cam/test/system/tests_posttag_bluevista
- post-tag tests on bluevista folded into post-tag blueice suite
List all subroutines added and what they do:
A models/atm/cam/test/system/config_files/e8dm
A models/atm/cam/test/system/config_files/e8idm
A + models/atm/cam/test/system/config_files/fmgpa1.9dh
A + models/atm/cam/test/system/config_files/fmgpa1.9dm
- new configure options files required for test scripts
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/tests_pretag_bangkok
M models/atm/cam/test/system/tests_pretag_bluevista
M models/atm/cam/test/system/tests_posttag_blueice
M models/atm/cam/test/system/tests_posttag_calgary
M models/atm/cam/test/system/tests_posttag_lightning
- new default test lists reflecting tests inherited from tempest
M models/atm/cam/test/system/tests_posttag_jaguar_cb
M models/atm/cam/test/system/tests_posttag_jaguar
- enhanced list of tests for jaguar including performance testing, testing
of omp threading, pgi debug mode, and fv 2d decomposition
M models/atm/cam/test/system/test_driver.sh
- removed references to machine tempest; minor fixes to jaguar directives
M models/atm/cam/test/system/input_tests_master
- mods to master test list reflecting tweaks to tests farmed out of tempest
M models/atm/cam/test/system/CAM_runcmnd.sh
- removed references to machine tempest; mod to jaguar run command enabling
ability to test in hybrid mode (spmd and smp)
M models/atm/cam/doc/ChangeLog_template
- removed references to machine tempest
M models/atm/cam/src/control/atm_comp_mct.F90
- minor fix from tonyc where uninitialized data was causing problems
in the cpl history file
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista:
bl382 TBL.sh fmgpa1.9dh outfrq3s+1870_control 9s ..............FAIL! rc= 5
bangkok/lf95:
bl114 TBL.sh e8dm co2rmp 9s ...................................FAIL! rc= 5
bl115 TBL.sh e8idm idphys 9s ..................................FAIL! rc= 5
- these baseline tests failed because their configurations were not
recognized by the scripts in the baseline code base
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_5_40
Originator(s): Dani Coleman (bundy@ucar.edu)
Date: 28 Mar 2008
One-line Summary: Fix pdeldry bug and remove cnst_need_pdeldry option
Purpose of changes:
The bug caused the EUL dycore to fail when run with phys_loadbalance=2 and
cnst_need_pdeldry. Now the model always saves pdeldry, regardless of the constituents.
Bugs fixed (include bugzilla ID): 388 (spmdbuf_siz needs to be increased in spectral Eulerian dycore)
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
Removed cnst_need_pdeldry logic in all modified subroutines (now model
runs as if cnst_need_pdeldry = .true. )
M models/atm/cam/src/physics/cam/cam_diagnostics.F90
M models/atm/cam/src/physics/cam/phys_gmean.F90
M models/atm/cam/src/physics/cam/constituents.F90
M models/atm/cam/src/physics/cam/physics_types.F90
M models/atm/cam/src/physics/cam/physpkg.F90
M models/atm/cam/src/physics/cam/zm_conv_intr.F90
M models/atm/cam/src/dynamics/eul/scanslt.F90
M models/atm/cam/src/dynamics/eul/inidat.F90
M models/atm/cam/src/dynamics/eul/prognostics.F90
M models/atm/cam/src/dynamics/eul/scandyn.F90
M models/atm/cam/src/dynamics/eul/scan2.F90
M models/atm/cam/src/dynamics/eul/restart_dynamics.F90
M models/atm/cam/src/dynamics/eul/tfilt_massfix.F90
Also changed buffer increment from 4 to 5.
M models/atm/cam/src/dynamics/eul/dp_coupling.F90 line 146
M models/atm/cam/src/dynamics/eul/spmd_dyn.F90 line 1065
Also calculate pdeldry in call to set_state_pdry. It either has to be
calculated here (like in the FV & SLD dycores) or passed back from the
dynamics (like in the EUL dycore)
M models/atm/cam/src/dynamics/homme/dp_coupling.F90
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: all pass
tempest: all pass
bangkok/lf95: all pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate): n/a
If bitwise differences were observed, how did you show they were no worse
than roundoff? n/a
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
n/a
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate: n/a
URL for AMWG diagnostics output used to validate new climate: n/a
==============================================================
Tag name: cam3_5_39
Originator(s): mirin
Date: 3/21/08
One-line Summary: Advect tracers simultaneously and allow finer decomp. with FV
Purpose of changes: Several issues are addressed:
A. Ability to advect multiple tracers simultaneously with FV
Tracers may be divided into "trac_decomp" groups, with trac2d invoked
concurrently for the various groups. The additional groups are solved
using auxiliary computational processes. In particular, the kth group of
tracers is advected on the kth set of npes_yz processs. The relevant
namelist variable is trac_decomp, whose default is 1. The implementation
of this option is very similar to that of overlapping trac2d and
cd_core subcycles; however, these options are presently mutually exclusive.
B. Ability to use larger FV domain decomposition
The condition that a subdomain contain 3 vertical lines has been relaxed.
Subdomains need contain at least one vertical line.
C. Bug fix and minor modification to mod_comm
A bug was discovered that affects scenarios with more than 1024 processes
ONLY WHEN at least one of the modcomm_?? namelist variables equals 1.
This was remedied by changing some static arrays to allocatable.
Also, the quantity "alloc_slack_factor" was changed to 8, to accommodate
certain fine-domain-decomposition scenarios.
D. Allow new 64-bit netCDF format
This is necessary for some high-resolution scenarios. The relevant
namelist variable is use_64bit_nc, which defaults as .false.
E. Incorporate minor modification and bug fix from MVR
Bug fix involves use with PGI.
Bugs fixed (include bugzilla ID):
Mod_comm bug (see item C above) and Makefile.in bug (see item E above).
Describe any changes made to build system:
Describe any changes made to the namelist: Namelist variables trac_decomp and use_64bit_nc
added. See items A and D above.
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Sawyer
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
models/atm/cam/dynamics/fv:
cd_core.F90 - post sends for simultaneous tracer advection; merge logic
with that of overlapping cd_core and trac2d subcycles; calling
arguments affected.
dyn_comp.F90 - insert main logic for simultaneous tracer advection; merge
logic with that of overlapping cd_core and trac2d subcycles; these
two options are mutually exclusive.
dynamics_vars.F90 - initialize necessary quantities for simultaneous
tracer advection.
spmd_dyn.F90 - add communicator logic for simultaneous tracer advection;
relax requirement of 3 verticals per subdomain; fix a couple of minor
allocation bugs with auxiliary processes.
trac2d.F90 - allow advection of a subset of tracers.
models/atm/cam/control:
cam_control_mod.F90 - add use_64bit_nc namelist variable for new 64-bit
netCDF format.
cam_history.F90 - optionally invoke new 64-bit netCDF format when creating
netCDF history file.
fv_control_mod.F90 - add trac_decomp namelist variable for concurrent
tracer advection.
runtime_opts.F90 - add namelist variables use_64bit_nc and trac_decomp.
models/utils/pilgrim:
mod_comm.F90 - change InHandle and OutHandle from static to allocatable
arrays; change alloc_slack_factor to 5; add comments.
models/atm/cam/bld:
Makefile.in - incorporate Mat's corrections for PGI.
cam.cpl7.template - allow one vertical per subdomain for FV
camsom.cpl7.template - incorporate Mat's corrections for ccsm4.
models/atm/cam/bld/namelist_files:
namelist_definition.xml - add use_64bit_namelist, trac_decomp, and
ct_overlap (previously omitted).
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista:
tempest:
bangkok/lf95:
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff? Answers are BFB.
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_5_38
Originator(s): mvr, mvertens
Date: 080314
One-line Summary: brought cam up to date with ccsm4_0_alpha25 driver
Purpose of changes: To bring the ccsm4_0_alpha25 cam branch and other
ccsm4_0_alpha25 externals (primarily the drvseq2_0_10 driver) onto
the cam trunk. The cam "stand-alone" configuration (for both
cam/dom and cam/som) will always be run with the new driver
"ocean_tight_coupling" namelist option set to .true. The ocean
tight coupling option treats ocean model like lnd/ice in
coupling. When this is true, the updated state from the ocean is
merged with the updated state from the land and ice to create the
input atmospheric state. The updated state from the ocean is also
used in the atm/ocn flux calculation. When this is not true
(i.e. loose coupling) the previous state of the ocean is used
along with the updated state from the land and ice to be merged
into the input atmosphere. In addition, the previous state of the
ocean is used in the atm/ocn flux calculation too. Loose coupling
is the way pop2 is always coupled into the system and corresponds
to the way a concurrent ocean would be run.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: modified configure to
add new flexibility to bring in cice (which is
not the default yet) and added COUP_DOM as a CPP variable.
Describe any changes made to the namelist: driver namelist modifications
List any changes to the defaults for the boundary datasets: the following
sst datasets are now the defaults (since they have the correct
time and calendar attributes to run with the cice in prescribed mode)
atm/cam/sst/sst_HadOIBl_bc_128x256_clim_c050526.nc
atm/cam/sst/sst_HadOIBl_bc_64x128_clim_c050526.nc
atm/cam/sst/sst_HadOIBl_bc_48x96_clim_c050526.nc
atm/cam/sst/sst_HadOIBl_bc_8x16_clim_c050526.nc
atm/cam/sst/sst_HadOIBl_bc_1.9x2.5_clim_c061031.nc
atm/cam/sst/sst_HadOIBl_bc_4x5_clim_c061031.nc
atm/cam/sst/sst_HadOIBl_bc_10x15_clim_c050526.nc
Describe any substantial timing or memory changes:
Code reviewed by: mvertens
List all subroutines eliminated:
D models/atm/cam/bld/run-pc-cice-docn7.csh
D models/atm/cam/bld/run-ibm-cice-docn7.csh
List all subroutines added and what they do:
A + models/atm/cam/bld/cam.cpl7.template
A + models/atm/cam/bld/camdom.cpl7.template
A + models/atm/cam/bld/camsom.cpl7.template
A + models/atm/cam/bld/camcsim.cpl7.template
- the above were modified to enable cam stand-alone configurations
to run out of the ccsm4_0_alpha25 (and later) scripts
List all existing files that have been modified, and describe the changes:
M models/ocn/dom/ocn_comp.F90
M models/ocn/dom/ocn_comp_mct.F90
M models/ocn/som/ocn_comp.F90
M models/ocn/som/ocn_comp_mct.F90
- interface changes needed to run with drvseq2_0_10
M models/ice/csim4/ice_comp.F90
M models/ice/csim4/ice_comp_mct.F90
- interface changes needed to run with drvseq2_0_10
M models/atm/cam/test/system/TCB.sh
M models/atm/cam/test/system/TBR.sh
M models/atm/cam/test/system/TPF.sh
M models/atm/cam/test/system/test_driver.sh
M models/atm/cam/test/system/TER.sh
M models/atm/cam/test/system/TSM.sh
- changes needed to run with drvseq2_0_10 utilizing the
ocean_tight_coupling driver namelist option (see above)
M models/atm/cam/bld/configure
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
M models/atm/cam/bld/namelist_files/namelist_definition.xml
M models/atm/cam/bld/build-namelist
M models/atm/cam/bld/config_definition.xml
- added new flags for ocn and ice configurations
- put in flexibility to run with cice in ccsm4_0_alpha25
- modified driver namelist definition to be consistent with drvseq2_0_10
- changed defaults sst datasets (see above)
- set ocean_tight_coupling to .true. (see above)
M models/atm/cam/src/control/runtime_opts.F90
- removed shr_orb_mod from use
- now broadcast obliqr,mvelpp,lambm0 instead of obliq,mvelp, iyear_ad
M models/atm/cam/src/control/ccsm_msg.F90
- removed psl as an allocatable array - now obtaining psl as in atm_comp_mct.F90
via cam_out%psl
M models/atm/cam/src/control/camsrfexch_types.F90
- added psl to surface_state data type
M models/atm/cam/src/control/cam_control_mod.F90
- orbital variables modified to be consistent with drvseq2_0_10
M models/atm/cam/src/control/atm_comp_mct.F90
M models/atm/cam/src/control/cam_comp.F90
- modifications made necessary for interacting with drvseq2_0_10
M models/atm/cam/src/utils/time_manager.F90
M models/atm/cam/src/physics/cam/tphysbc.F90
- surface_state2d (or in ccsm_msg, cam_out%psl) is now filled in via the
call to diag_phys_writeout
M models/atm/cam/src/physics/cam/cam_diagnostics.F90
- removed references to COUP_CSM - added psl as an optional argument to
diag_phys_writeout instead of utilizing the COUP_CSM directive
M models/atm/cam/src/physics/cam/co2_cycle.F90
- removed COUP_CSM if-def
M .
M SVN_EXTERNAL_DIRECTORIES
models/drv/seq_mct https://svn-ccsm-models.cgd.ucar.edu/drv/seq_mct/trunk_tags/drvseq2_0_10
models/lnd/clm2/src https://svn-ccsm-models.cgd.ucar.edu/clm2/branch_tags/ccsm4a_tags/ccsm4a03_clm3_5_18/src
models/ice/cice https://svn-ccsm-models.cgd.ucar.edu/cice/trunk_tags/cice4_0_20080227
models/ocn/docn7 https://svn-ccsm-models.cgd.ucar.edu/docn7/branch_tags/drva_docn7_070824_tags/drva11_docn7_071129
models/csm_share https://svn-ccsm-models.cgd.ucar.edu/csm_share/branch_tags/loga_share3_070903_tags/loga25_share3_071107
models/utils/esmf_wrf_timemgr https://svn-ccsm-models.cgd.ucar.edu/esmf_wrf_timemgr/trunk_tags/esmf_wrf_timemgr_060616
models/utils/timing https://svn-ccsm-models.cgd.ucar.edu/timing/trunk_tags/timing_070525
models/utils/mct https://svn-ccsm-models.cgd.ucar.edu/mct/branch_tags/seqa_MCT2_3_0_070524_tags/seqa07_MCT2_4_2_071026
scripts https://svn-ccsm-models.cgd.ucar.edu/scripts/branch_tags/seqmct_scripts_070823_tags/seqmct44_scripts_080108
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
blueice:
004 bl132 TBL.sh e32sdh ghgrmp 9s .................................FAIL! rc= 7
011 bl151 TBL.sh e64h outfrq24h 2d ................................FAIL! rc= 7
015 bl332 TBL.sh f4sdh ghgrmp 9s ..................................FAIL! rc= 7
023 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7
026 bl337 TBL.sh f4dh fv2d_8tsk 9s ................................FAIL! rc= 7
030 bl371 TBL.sh f1.9dm fvvp_lb2 9s ...............................FAIL! rc= 7
034 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7
037 bl374 TBL.sh fm1.9dh outfrq3s 9s ..............................FAIL! rc= 7
040 bl355 TBL.sh fmo2dh off2x2.5 9s ...............................FAIL! rc= 7
044 bl381 TBL.sh fma1.9h outfrq3s+1870_prog_aero 9s ...............FAIL! rc= 7
048 bl382 TBL.sh fmgpa1.9h outfrq3s+1870_control 9s ...............FAIL! rc= 7
052 bl532 TBL.sh s32sdh ghgrmp 9s .................................FAIL! rc= 7
059 bl551 TBL.sh s64h outfrq24h 2d ................................FAIL! rc= 7
tempest:
004 bl131 TBL.sh e32dh co2rmp 9s ..................................FAIL! rc= 7
007 bl132 TBL.sh e32sdh ghgrmp 9s .................................FAIL! rc= 7
014 bl152 TBL.sh e64o outfrq24h 2d ................................FAIL! rc= 7
017 bl314 TBL.sh wg10dh outfrq3s 9s ...............................FAIL! rc= 7
021 bl331 TBL.sh f4gdh co2rmp 9s ..................................FAIL! rc= 7
024 bl332 TBL.sh f4sdh ghgrmp 9s ..................................FAIL! rc= 7
031 bl376 TBL.sh f1.9o outfrq24h 2d ...............................FAIL! rc= 7
035 bl531 TBL.sh s32dh co2rmp 9s ..................................FAIL! rc= 7
038 bl532 TBL.sh s32sdh ghgrmp 9s .................................FAIL! rc= 7
045 bl552 TBL.sh s64o outfrq24h 2d ................................FAIL! rc= 7
bangkok/lf95:
004 bl111 TBL.sh e8c8mdm ttrac 9s .................................FAIL! rc= 7
008 bl112 TBL.sh e8sdm ghgrmp 9s ..................................FAIL! rc= 7
012 bl153 TBL.sh e64m outfrq24h 2d ................................FAIL! rc= 7
016 bl311 TBL.sh f10c8mdm ttrac 9s ................................FAIL! rc= 7
024 bl312 TBL.sh f10sdm ghgrmp 9s .................................FAIL! rc= 7
029 bl315 TBL.sh f10c5dm microp_mg 9s .............................FAIL! rc= 7
032 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7
035 bl377 TBL.sh f1.9m outfrq24h 2d ...............................FAIL! rc= 7
039 bl511 TBL.sh s8c8mdm ttrac 9s .................................FAIL! rc= 7
043 bl512 TBL.sh s8sdm ghgrmp 9s ..................................FAIL! rc= 7
047 bl553 TBL.sh s64m outfrq24h 2d ................................FAIL! rc= 7
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: all
- what platforms/compilers: all
- nature of change (roundoff; larger than roundoff but same climate; new
climate): larger than roundoff but same climate
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository): ccsm4_0_alpha25
- platform/compilers: blueice
- configure commandline:
som:
create_newcase -mach blueice -compset F_CAMSOM_CSIM -res f19_f19 -case /ptmp/mvr/f40.021
dom:
create_newcase -mach blueice -compset F_CAMDOM_CSIM -res f19_f19 -case /ptmp/mvr/f40.017
- build-namelist command (or complete namelist):
see the entries for f40.017 and f40.021 in run database, http://ccsm-rundb.cgd.ucar.edu
- MSS location of output:
som: mss:/MVR/csm/f40.021
dom: mss:/MVR/csm/f40.017
- MSS location of control simulations used to validate new climate:
som: mss:/MVR/csm/seqnoflxave06_cam3_5_24_som4
dom: mss:/MVR/csm/seqnoflxave06_cam3_5_24
- URL for AMWG diagnostics output used to validate new climate:
som: mss:/MVR/csm/f40.021/f40.021-seqnoflxave06_cam3_5_24_som4.tar
http://www.cgd.ucar.edu/cms/mvr/f40.021
dom: mss:/MVR/csm/f40.017/f40.017-seqnoflxave06_cam3_5_24.tar
http://www.cgd.ucar.edu/cms/mvr/f40.017
===============================================================
===============================================================
Tag name: cam3_5_37
Originator(s): andrew, eaton
Date: Thu Mar 13 12:27:27 MDT 2008
One-line Summary: Mods and bug fixes in MG microphysics and CLDTOP/CLDBOT indexes.
Purpose of changes:
. Mods and bug fixes for MG microphysics -- see details in changes to
prescribed_aerosols and cldwat2m modules below.
. Fix bug in UW shallow convection scheme that was setting cloud top and
cloud bottom indexes incorrectly in cloud free columns. The chemistry
is the only parameterization affected by this bug.
. Change the default averaging flag for radiative forcing and aerosol
indirect diagnostic output fields from 'I' to 'A'
Bugs fixed (include bugzilla ID):
. bug in cldtop/cldbot indexes (bugz ID 716)
. bug fixes for MG microphysics
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: not checked
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/src/physics/cam/prescribed_aerosols.F90
. modify dry_radaer for OCPHO,OCPHI,BCPHO,BCPHI,VOLC to 0.05e-6 (same as sulfate).
[this changes answers slightly, and makes the microphysical distributions
better in the model]
models/atm/cam/src/physics/cam/cldwat2m.F90
. several bug fixes for diagnostic CCN output only [does not change answers]
. fix prodprec (prain) variable [changes answers only with prognostic aerosols]
. additional mod: put prodsnow and evapsnow into prain, evaprain
[changes answers only with prognostic aerosols: this fixes the
wet-deposition]
models/atm/cam/src/physics/cam/mcshallow.F90
. In compute_mcshallow change the initializers for cnt_out to 1 and for
cnb_out to pver. Only do it in the last of the 4 places where the
cnt_out/cnb_out values are set. This appears to be the section of code
that re-initializes values for columns with no convection.
models/atm/cam/src/physics/cam/radiation.F90
. change default averaging of radiative forcing diagnostics from 'I' to 'A'
. change default averaging of aerosol indirect forcing diagnostics from 'I' to 'A'
models/atm/cam/src/physics/waccm/gw_drag.F90
models/atm/cam/src/physics/waccm/mo_msis_ubc.F90
. get pi from CAM's physconst module rather than from shr_const_mod.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
blueice: All PASS
tempest: All PASS
calgary/lf95: All PASS except
029 bl315 TBL.sh f10c5dm microp_mg 9s .............................FAIL! rc= 7
bl315 is expected to fail due to changes to MG microphysics.
CAM tag used for the baseline comparison tests if different than previous
tag: cam3_5_34
Summarize any changes to answers: BFB except when MG microphysics is on
due to bug fixes in that code.
Andrew has verified this code is reasonable both with our microphysical
diagnostics and with the standard diagnostics.
Diagnostics at:
http://www.cgd.ucar.edu/cms/andrew/diag/nmicro2.65_cam3_5_36-obs/
Diagnostics for a pair of 2 year runs with the
fixed version (nmicro2.68) and the current version on the trunk
(cam3_5_36_MG) are at:
http://www.cgd.ucar.edu/cms/andrew/diag/cam3_5_36_MG-nmicro2.68_cam3_5_36
Note that both of these are with the Morrison-Gettelman microphysics.
===============================================================
===============================================================
Tag name: cam3_5_36
Originator(s): eaton
Date: Mon Mar 3 10:11:54 MST 2008
One-line Summary: bug fixes for build-namelist
Purpose of changes:
. Bug fixes for build-namelist -- see below.
. Reimplement the -test option to build-namelist. This checks whether the
input datasets exist on the local disk. It's also an easy way to
generate a list of the input datasets required for the run.
. Allow the phys_debug_util module to read its own namelist. The
implementation is meant to serve as an example for other modules for
which it's appropriate to read namelist input. The general design that
we've been moving towards is for modules to read their own namelist
rather than have all namelist variables live in one big namelist. The
new build-namelist utility makes it easy to define a new namelist group
and have that namelist group be written to the atm_in file simply by
adding an appropriate definition of the variables in the
namelist_definition.xml file.
Bugs fixed (include bugzilla ID):
. build-namelist wasn't validating namelist array variables that were specified
in the input as array elements.
. build-namelist was not recognizing values in the -namelist argument when
the string contained embedded newlines. This happens when using the C
shell and splitting the -namelist arg into multiple lines.
. build-namelist wasn't checking the date attributes in the default values
for finidat.
Describe any changes made to build system: none
Describe any changes made to the namelist:
. phys_debug_lat and phys_debug_lon have moved to the new namelist group
phys_debug_nl.
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/build-namelist
. modify so that when a new namelist group is defined for the CAM
component that group will automatically be written to CAM's namelist
file.
. fix setting the default value of finidat to recognize the date
attribute.
models/atm/cam/bld/namelist_files/namelist_definition.xml
. move phys_debug_lat and phys_debug_lon into namelist group phys_debug_nl
models/atm/cam/bld/perl5lib/Build/NamelistDefinition.pm
. modify the _validate_pair method so that an index specifier is stripped
off of the variable name before checking whether it's in the definition
file.
models/atm/cam/bld/perl5lib/Build/Namelist.pm
. allow strings used to initialize Namelist objects to contain embedded
newlines.
models/atm/cam/src/physics/cam/phys_debug_util.F90
. add subroutine phys_debug_readnl to read the namelist.
models/atm/cam/src/control/runtime_opts.F90
. remove phys_debug_lat, phys_debug_lon from cam_inparm
. add call to phys_debug_readnl
models/atm/cam/src/physics/cam/convect_shallow.F90
. add CLDTOP and CLDBOT to master field list for debugging.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
blueice: All PASS.
tempest: All PASS.
calgary/lf95: All PASS.
CAM tag used for the baseline comparison tests if different than previous
tag: cam3_5_34
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam3_5_35
Originator(s): sungsu, pjr, pworley, mvr, eaton
Date: Fri Feb 15 17:15:40 MST 2008
One-line Summary: Mods to PBL code; add physics debug utility; fix archiving script
Purpose of changes:
. Modifications to UW PBL scheme. (sungsu)
. Add HBR option to eddy_scheme. (pjr)
. Add flux_avg module which contains code to smooth the surface fluxes
to reduce instabilities in the surface layer. (pjr)
. Add physics debugging utility code that allows easy location of the
closest column in the physics grid to a specified location. (pjr, pworley)
. Some debugging code to help analyze how PBL schemes are behaving near the
surface has been left on the trunk until the CAM4 PBL scheme has been
verified. The debugging code will be removed once the verification has
taken place. (pjr, eaton)
. Update external for scripts directory. *** N.B. *** This fixes a bug in
the archiving scripts introduced at cam3_5_31. (mvr)
Bugs fixed (include bugzilla ID): Fix archiving script bug.
Describe any changes made to build system: none
Describe any changes made to the namelist: added variables
. logical rhminl_31 - to allow use of special value of rhminl in 31 layer model
(default: .false.)
. logical freeze_dry - to allow use of Vavrus "freeze dry" mod (default: .true.)
. integer srf_flux_avg - on=1/off=0 switch for code to adjust surface fluxes to
stabilize near surface layers (default: 0 if
eddy_scheme='HB', 1 otherwise )
. real(r8) phys_debug_lat - requested latitude for debugging physics (degrees)
. real(r8) phys_debug_lon - requested longitude for debugging physics (degrees)
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: not tested
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do:
models/atm/cam/src/physics/cam/flux_avg.F90
. surface flux adjustments
models/atm/cam/src/physics/cam/phys_debug_util.F90
. physics debugging utility
models/atm/cam/src/physics/cam/phys_debug.F90
. user code for specific debugging output
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/namelist_files/namelist_definition.xml
models/atm/cam/src/control/runtime_opts.F90
. add new namelist variables freeze_dry, rhminl_31, do_flux_averaging,
phys_debug_lat, phys_debug_lon
models/atm/cam/src/physics/cam/cam_diagnostics.F90
. don't make outfld call for PSDRY if atm is dry
models/atm/cam/src/physics/cam/check_energy.F90
. add message to log about status of printing conservation errors
models/atm/cam/src/physics/cam/cloud_fraction.F90
. add namelist variables
- rhminl_31 to allow use of special value of rhminl in 31 layer model
(default is .false.)
- freeze_dry to allow use of Vavrus "freeze dry" mod (default is .true.)
. change rhminh from .9 to .85 for UW shallow scheme without MG
microphysics, but leave it at .9 when UW is run w/ MG.
models/atm/cam/src/physics/cam/eddy_diff.F90
. change default averaging of diagnostic output from instantaneous to
averaged
. mods from sungsu
models/atm/cam/src/physics/cam/hb_diff.F90
. mods for HBR option
. add temporary debug code
models/atm/cam/src/physics/cam/initindx.F90
. add register routine for flux averaging
models/atm/cam/src/physics/cam/mcshallow.F90
. change default averaging of diagnostic output from instantaneous to
averaged
. set rdrag=1.0 (was 2.0)
. modify evprain expression
. change some hardcoded constants in qvten expression
models/atm/cam/src/physics/cam/phys_control.F90
. add check for eddy_scheme='HBR'
. add namelist variable srf_flux_avg to control whether the flux averaging
code is turned on.
models/atm/cam/src/physics/cam/phys_grid.F90
. add subroutine phys_grid_find_col
models/atm/cam/src/physics/cam/physconst.F90
. add pi
models/atm/cam/src/physics/cam/physics_types.F90
. add temporary debug code
models/atm/cam/src/physics/cam/physpkg.F90
. add init call for physics_debug
. move tracer_init call to before exit for adiabatic/ideal physics
models/atm/cam/src/physics/cam/qneg4.F90
. modify warning message
models/atm/cam/src/physics/cam/rayleigh_friction.F90
. fix initialization of ptend
models/atm/cam/src/physics/cam/tphysac.F90
. add optional call for surface flux averaging
models/atm/cam/src/physics/cam/tphysbc.F90
. add temporary debug code
models/atm/cam/src/physics/cam/vertical_diffusion.F90
. mods to UW code by Sungsu
. add HBR option to eddy_scheme
. add eddy_scheme as actual arg to init_hb_diff
. change default averaging from I to A for UW PBL diagnostic fields
. add temporary debug code
models/atm/cam/src/physics/cam/zm_conv_intr.F90
. move assignment of nstep to top of zm_conv_tend
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: (used blueice due to XLF11 problem on bluevista): All PASS.
tempest: All PASS.
bangkok/lf95: All PASS.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB.
===============================================================
===============================================================
Tag name: cam3_5_34
Originator(s): mvertens
Date: Tue Feb 5 23:19:00 MST 2008
One-line Summary: minor bug fixes for cpl6 mode compatibilty
Purpose of changes: fix a compiler problem on jaguar and put fix back
in (that was taken out of cam3_5_33) for coupling every time step
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: mvertens
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
M src/control/runtime_opts.F90
- finished cleaning up uses of eshr_timemgr_mod and shr_inputinfo_mod
that are no longer needed
M src/control/ccsm_msg.F90
- a fix was needed in ccsm_msg.F90 to match a corresponding correction
in the flux coupler to correctly specify the albedo shift and enable
communication every cam timestep - this fix was inadvertently taken
out of cam3_5_33 - and has been put back in
M src/control/cam_control_mod.F90
- removed reference to flxave which was causing a compiler problem on
jaguar
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista:
all pass
tempest:
all pass
bangkok/lf95:
all pass
CAM tag used for the baseline comparison tests if different than previous
tag: cam3_5_32
Summarize any changes to answers: none, answers are bfb
===============================================================
===============================================================
Tag name: cam3_5_33
Originator(s): mvertens
Date: Tue Feb 5 08:39:49 MST 2008
One-line Summary: migrated all cpl7 framework references to xxx_comp_mct.F90
Purpose of changes: enable cam-cpl6 to run without reference to any cpl7
data structures and move move all cpl7 framework references to xxx_comp_mct.F90
(e.g. remove all references from cam_comp.F90, etc).
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: new namelist, camcpl6_inparm, was introduced
that is independent of any cpl7 utilities.
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: mvertens
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
All framework specific data structures were moved into xxx_comp_mct.F90
routies.
M models/ocn/dom/ocn_comp.F90
M models/ocn/dom/ocn_comp_mct.F90
M models/ocn/som/ocn_comp.F90
M models/ocn/som/ocn_comp_mct.F90
- moved calls to routimemgr_init into ocn_comp_mct from ocn_comp.F90
M models/atm/cam/src/control/runtime_opts.F90
- removed all references to shr_input_info or eshr_timemgr module routines
and variables. Obtaining data from sync clock, setting nsrest and getting
data from ccsminit object have been moved to atm_comp_mct.F90.
- removed routine runtime_options, now call read_namelist directly in
both cpl6 and cpl7 cam interfaces (call preset directly from read_namelist)
- removed flxave from cam_inparm and placed it in camcpl6_inparm
- removed broadcast for single column variables
M models/atm/cam/src/control/ccsm_msg.F90
M models/atm/cam/src/control/con_cam.F90
- eliminated shr_input_info and eshr_timemgr references
- explicit use now made of camcpl6_inparm namelist input
M models/atm/cam/src/control/atm_comp_mct.F90
M models/atm/cam/src/control/cam_comp.F90
M models/atm/cam/bld/cam.cpl6.template
- modifications to get branch and hybrid tests working
M models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
- removed reference to flxave
M models/atm/cam/bld/namelist_files/namelist_definition.xml
- removed reference to flxave
M models/atm/cam/bld/build-namelist
- removed reference to flxave
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista:
all pass
tempest:
all pass
bangkok/lf95:
all pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: none, answers are bfb
===============================================================
===============================================================
Tag name: cam3_5_32
Originator(s): mvr,mvertens
Date: 080131
One-line Summary: mods to enable cam to work with cpl7
Purpose of changes:
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
namelist 'timemgr_inparm' changed name to 'camcpl6_inparm'
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: mvertens,eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/cam.cpl6.template
M models/atm/cam/bld/build-namelist
M models/atm/cam/src/control/runtime_opts.F90
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: not tested
tempest: not tested
bangkok/lf95: all pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_5_31
Originator(s): mvr
Date: 080130
One-line Summary: enable test_driver for jaguar (cray xt); now pulling in
entire ccsm scripts directory as external; mod to allow communication with
flux coupler every cam timestep
Purpose of changes:
there was a need to perform regular cam testing on jaguar as reliance
on this machine for production use increases
a bug fix was added to the archiving scripts, so the external for
its scripts tag needed updating...we eventually will need the ccsm
scripts available as the two systems merge, so we decided to pull
in the entire ccsm scripts directory as the external
a fix was needed in ccsm_msg.F90 to match a corresponding correction
in the flux coupler to correctly specify the albedo shift and enable
communication every cam timestep
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
pgi compiler flags now include -Kieee; non-debug pgi compiler settings
now use '-fast -mvect=nosse' instead of -O1
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: mvr,eaton
List all subroutines eliminated:
D models/atm/cam/test/system/config_files/fm2dh
- config options file with fv 2x2.5 replaced with one at fv 1.9x2.5
List all subroutines added and what they do:
A models/atm/cam/test/system/config_files/wm1.9m
A models/atm/cam/test/system/config_files/e64sm
A models/atm/cam/test/system/config_files/fm1.9dh
A models/atm/cam/test/system/config_files/fm1.9m
A models/atm/cam/test/system/config_files/fma1.9m
A models/atm/cam/test/system/config_files/e64pm
A models/atm/cam/test/system/config_files/fmgpa1.9m
A models/atm/cam/test/system/config_files/e64am
- new config options files required for testing on jaguar
A models/atm/cam/test/system/tests_posttag_jaguar
- new file containing the default set of tests for jaguar
A models/atm/cam/test/system/tests_posttag_jaguar_cb
- set of tests for jaguar containing just the configure/build's
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/test_driver.sh
- added functionality for testing on jaguar (cray xt)
M models/atm/cam/test/system/tests_pretag_bluevista
M models/atm/cam/test/system/tests_posttag_blueice
- modified default test lists for testing on ibm's to reflect testing
fv 1.9x2.5 rather that fv 2x2.5
M models/atm/cam/test/system/input_tests_master
- master list of tests now include those added for jaguar
M models/atm/cam/test/system/CAM_runcmnd.sh
- code added to properly build the run command on jaguar
M models/atm/cam/bld/Makefile.in
- new default compiler settings for pgi
M models/atm/cam/bld/run-ibm.csh
M models/atm/cam/bld/run-pc-cice-docn7.csh
M models/atm/cam/bld/run-pc.csh
M models/atm/cam/bld/run-sgi.csh
M models/atm/cam/bld/run-lightning.csh
- run template scripts modified to reflect new location of archive scripts
M models/atm/cam/src/control/ccsm_msg.F90
- fix to specify albedo shift correctly and enable communication with coupler
every cam timestep
M .
M SVN_EXTERNAL_DIRECTORIES
- now pulling in entire ccsm scripts directory as external
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista:
all pass
tempest:
all pass
bangkok/lf95:
all pass
calgary/pgf:
004 bl111 TBL.sh e8c8mdm ttrac 9s .................................FAIL! rc= 7
008 bl112 TBL.sh e8sdm ghgrmp 9s ..................................FAIL! rc= 7
010 bl113 TBL.sh e8pdm aqpgro 3s ..................................FAIL! rc= 7
012 bl153 TBL.sh e64m outfrq24h 2d ................................FAIL! rc= 7
016 bl311 TBL.sh f10c8mdm ttrac 9s ................................FAIL! rc= 7
024 bl312 TBL.sh f10sdm ghgrmp 9s .................................FAIL! rc= 7
026 bl313 TBL.sh f10pdm aqpgro 3s .................................FAIL! rc= 7
029 bl315 TBL.sh f10c5dm microp_mg 9s .............................FAIL! rc= 7
032 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7
035 bl377 TBL.sh f1.9m outfrq24h 2d ...............................FAIL! rc= 7
039 bl511 TBL.sh s8c8mdm ttrac 9s .................................FAIL! rc= 7
043 bl512 TBL.sh s8sdm ghgrmp 9s ..................................FAIL! rc= 7
045 bl513 TBL.sh s8pdm aqpgro 3s ..................................FAIL! rc= 7
047 bl553 TBL.sh s64m outfrq24h 2d ................................FAIL! rc= 7
all baselines fail due to changes to pgf compiler flags
NOTE: tests of this cam tag within ccsm4 show that future mods will be required
including a minor fix to the new build-namelist and fixes to cam.cpl6.template
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: all
- what platforms/compilers: pgf compilers only
- nature of change (roundoff; larger than roundoff but same climate; new
climate): roundoff
If bitwise differences were observed, how did you show they were no worse
than roundoff? pergro tests passed on calgary/pgf and jaguarcnl/pgf
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_5_30
Originator(s): eaton
Date: Thu Jan 24 09:32:04 MST 2008
One-line Summary: new version of build-namelist
Purpose of changes:
. A new version of build-namelist has been implemented with new features
which will make namelist generation more robust, as well as being easier
to extend and maintain.
From the user perspective the biggest differences with the old version
are:
- All valid namelist variables are known to build-namelist. So an
incorrectly specified variable from the user (supplied either by the
-infile or -namelist options) will cause build-namelist to fail with an
error message telling which namelist variable is invalid. This is a
big improvement over a runtime failure caused by an invalid variable
which typically gives no hint as to which variable caused the problem.
- In addition to knowing all valid variable names and their types,
build-namelist also knows which namelist group each variable belongs
to. This means that the user only needs to specify variable names to
build-namelist and not the group names. The -infile and -namelist
options still require valid namelist syntax as input. But the group
name(s) is ignored. So all variables can be put in a single group with
an arbitrary name, for example, "&in ... /".
From the developer perspective the main differences are:
- New namelist variables added to existing namelist groups require adding
an entry to the definition file (namelist_definition.xml). If the new
variable has default values these are provided by modifying the
build-namelist script and possibly the defaults file
(namelist_defaults_cam.xml).
- Implementing a new namelist group no longer requires writing a new perl
module. The variables are added following the same procedure used to
add a variable to an existing namelist group.
The commandline interface to build-namelist has not been changed except
to remove the -cam_cfg option. This was used to specify the directory
containing the CAM configuration files. It's not needed because we
assume the files are located in directories relative to the one that
contains the build-namelist script being executed (which is known by
looking at $0).
The -test option has not yet been implemented.
The -v (verbose) option no longer accepts a value. It's an on switch.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: nrevsn
nrevsn has been given component specific names. It's functionality is
unchanged. The new names for each component are:
&cam_inparm nrevsn --> cam_branch_file
&csim_inparm nrevsn --> csim_branch_file
&dom_inparm nrevsn --> dom_branch_file
&som_inparm nrevsn --> dom_branch_file
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
models/atm/cam/bld/CAM_config.pm
models/atm/cam/bld/DefaultCAM_INPARM_Namelist.xml
models/atm/cam/bld/DefaultCCSM_INPARM_Namelist.xml
models/atm/cam/bld/DefaultCLM_INPARM_Namelist.xml
models/atm/cam/bld/DefaultCSIM_INPARM_Namelist.xml
models/atm/cam/bld/DefaultCTL_NL_Namelist.xml
models/atm/cam/bld/DefaultDOM_INPARM_Namelist.xml
models/atm/cam/bld/DefaultFILTER_NL_Namelist.xml
models/atm/cam/bld/DefaultSOM_INPARM_Namelist.xml
models/atm/cam/bld/DefaultTIMEMGR_INPARM_Namelist.xml
models/atm/cam/bld/NamelistsDescriptions.xml
models/atm/cam/bld/SeqCCSM_namelist.pm
models/atm/cam/bld/cam_inparm.pm
models/atm/cam/bld/camexp.pm
models/atm/cam/bld/ccsm_inparm.pm
models/atm/cam/bld/clm_inparm.pm
models/atm/cam/bld/clmexp.pm
models/atm/cam/bld/compnl.pm
models/atm/cam/bld/csim_inparm.pm
models/atm/cam/bld/ctl_nl.pm
models/atm/cam/bld/dom_inparm.pm
models/atm/cam/bld/filter_nl.pm
models/atm/cam/bld/namelist.pm
models/atm/cam/bld/nl_descrips.pm
models/atm/cam/bld/prof_inparm.pm
models/atm/cam/bld/som_inparm.pm
models/atm/cam/bld/timemgr_inparm.pm
models/atm/cam/bld/use_cases/1870-2000_control.nl
models/atm/cam/bld/use_cases/1870_control.nl
models/atm/cam/bld/use_cases/1870_prog_aero.nl
models/atm/cam/bld/use_cases/1990_control.nl
models/atm/cam/bld/use_cases/waccm_1995_climo.nl
List all subroutines added and what they do:
models/atm/cam/bld/namelist_files/namelist_defaults_cam.xml
models/atm/cam/bld/namelist_files/namelist_definition.xml
models/atm/cam/bld/namelist_files/use_cases/1870-2000_control.xml
models/atm/cam/bld/namelist_files/use_cases/1870_control.xml
models/atm/cam/bld/namelist_files/use_cases/1870_prog_aero.xml
models/atm/cam/bld/namelist_files/use_cases/1990_control.xml
models/atm/cam/bld/namelist_files/use_cases/waccm_1995_climo.xml
models/atm/cam/bld/perl5lib/Build/Namelist.pm
models/atm/cam/bld/perl5lib/Build/NamelistDefaults.pm
models/atm/cam/bld/perl5lib/Build/NamelistDefinition.pm
. files to implement new build-namelist
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/build-namelist
models/atm/cam/bld/cam.cpl6.template
models/atm/cam/src/control/cam_restart.F90
models/atm/cam/src/control/runtime_opts.F90
. replace namelist variable nrevsn by cam_branch_file
models/atm/cam/test/system/TBR.sh
. change nrevsn to the component appropriate name.
models/atm/cam/test/system/TSM.sh
. remove -cam_cfg option from build-namelist command.
models/ice/csim4/ice_comp.F90
. replace namelist variable nrevsn by csim_branch_file
models/ocn/dom/ocn_comp.F90
. replace namelist variable nrevsn by dom_branch_file
models/ocn/som/ocn_comp.F90
. replace namelist variable nrevsn by som_branch_file
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: All PASS.
tempest: All PASS.
bangkok/lf95: All PASS.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: NONE
===============================================================
===============================================================
Tag name: cam3_5_29
Originator(s): andrew, edwards, worley, eaton
Date: Wed Jan 9 11:21:24 MST 2008
One-line Summary: bug fixes for MG microphysics, aerosol optics, & misc
| 24 Jan 2008, eaton, update ChangeLog for cam3_5_29
|
| Andrew Conley let me know that the indexing bug in aer_optics.F90
| shouldn't affect the default physics because in the original prescribed
| aerosols the longwave effect of seasalt and dust was zero. I verified
| this. The ChangeLog documention has been updated accordingly.
Purpose of changes:
. Bug fixes and diagnostics changes for MG microphysics. Also changed some
diagnostics from RK microphysics. Only affects simulations with MG
microphysics turned on.
. Fix indexing bug in aer_optics.F90 which affects simulations where the
new seasalt and dust aerosols are radiatively active.
. Add fix or workaround (not sure which) for a problem with the pgi
compiler on jaguarcnl. Compiler is failing on a declaration of a zero
size array which uses a scalar initializer.
. Change entropy function (in zm_conv.F90) to return a real(r8) result. It
was returning a single precision result. This results in a larger than
roundoff difference in the simulations with default physics.
Also replace the stop with an endrun call in the convergence checking
loop of subroutine ientropy (in zm_conv.F90).
Bugs fixed (include bugzilla ID):
. bugz #683 - fix for single column output (was broken when making memory
use improvements in cam3_5_26).
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/src/chemistry/trop_mozart/mo_extfrc.F90
. initialize arrays has_extfrc and spc_fnames, which can potentially be of
size zero, in the extfrc_inti method rather than with initializers
models/atm/cam/src/control/cam_history.F90
. fix for single column output
models/atm/cam/src/physics/cam/aer_optics.F90
. fix indexing bug affecting new seasalt and dust optics
models/atm/cam/src/physics/cam/cldwat2m.F90
. Snow bug fixed (morrison email 27 Nov 2007) Line 1523 changed. Top level
autoconversion bug fixed (line 1515-1516) Rain diagnostic output changed
to gridbox average (the first 2 change answers for MG microphysics only,
the third is diagnostic)
models/atm/cam/src/physics/cam/stratiform.F90
. cloud fraction diagnostic calculations AFTER second cloud fraction call
to make in-cloud values consistent with outputed cloud fraction. changed
for MG and RK microphysics. This affects diagnostic in-cloud values in
RK and MG microphysics (icimr, icwmr in RK) Also: cleaned up in-cloud
diagnostics which were using ptend_loc after it was set to zero.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: All PASS except:
004 bl132 TBL.sh e32sdh ghgrmp 9s .................................FAIL! rc= 7
006 bl133 TBL.sh e32pdh aqpgro 3s .................................FAIL! rc= 7
011 bl151 TBL.sh e64h outfrq24h 2d ................................FAIL! rc= 7
015 bl332 TBL.sh f4sdh ghgrmp 9s ..................................FAIL! rc= 7
017 bl333 TBL.sh f4pdh aqpgro 3s ..................................FAIL! rc= 7
023 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7
026 bl337 TBL.sh f4dh fv2d_8tsk 9s ................................FAIL! rc= 4
030 bl371 TBL.sh f1.9dm fvvp_lb2 9s ...............................FAIL! rc= 7
034 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7
037 bl354 TBL.sh fm2dh outfrq3s 9s ................................FAIL! rc= 7
040 bl355 TBL.sh fmo2dh off2x2.5 9s ...............................FAIL! rc= 7
044 bl381 TBL.sh fma1.9h outfrq3s+1870_prog_aero 9s ...............FAIL! rc= 7
048 bl382 TBL.sh fmgpa1.9h outfrq3s+1870_control 9s ...............FAIL! rc= 7
052 bl532 TBL.sh s32sdh ghgrmp 9s .................................FAIL! rc= 7
054 bl533 TBL.sh s32pdh aqpgro 3s .................................FAIL! rc= 7
059 bl551 TBL.sh s64h outfrq24h 2d ................................FAIL! rc= 7
tempest: All PASS except:
004 bl131 TBL.sh e32dh co2rmp 9s ..................................FAIL! rc= 7
007 bl132 TBL.sh e32sdh ghgrmp 9s .................................FAIL! rc= 7
009 bl133 TBL.sh e32pdh aqpgro 3s .................................FAIL! rc= 7
014 bl152 TBL.sh e64o outfrq24h 2d ................................FAIL! rc= 7
017 bl314 TBL.sh wg10dh outfrq3s 9s ...............................FAIL! rc= 7
021 bl331 TBL.sh f4gdh co2rmp 9s ..................................FAIL! rc= 7
024 bl332 TBL.sh f4sdh ghgrmp 9s ..................................FAIL! rc= 7
026 bl333 TBL.sh f4pdh aqpgro 3s ..................................FAIL! rc= 7
031 bl376 TBL.sh f1.9o outfrq24h 2d ...............................FAIL! rc= 7
035 bl531 TBL.sh s32dh co2rmp 9s ..................................FAIL! rc= 7
038 bl532 TBL.sh s32sdh ghgrmp 9s .................................FAIL! rc= 7
040 bl533 TBL.sh s32pdh aqpgro 3s .................................FAIL! rc= 7
045 bl552 TBL.sh s64o outfrq24h 2d ................................FAIL! rc= 7
bangkok/lf95: All PASS except:
004 bl111 TBL.sh e8c8mdm ttrac 9s .................................FAIL! rc= 7
008 bl112 TBL.sh e8sdm ghgrmp 9s ..................................FAIL! rc= 7
010 bl113 TBL.sh e8pdm aqpgro 3s ..................................FAIL! rc= 7
012 bl153 TBL.sh e64m outfrq24h 2d ................................FAIL! rc= 7
016 bl311 TBL.sh f10c8mdm ttrac 9s ................................FAIL! rc= 7
024 bl312 TBL.sh f10sdm ghgrmp 9s .................................FAIL! rc= 7
026 bl313 TBL.sh f10pdm aqpgro 3s .................................FAIL! rc= 7
029 bl315 TBL.sh f10c5dm microp_mg 9s .............................FAIL! rc= 7
032 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7
035 bl377 TBL.sh f1.9m outfrq24h 2d ...............................FAIL! rc= 7
039 bl511 TBL.sh s8c8mdm ttrac 9s .................................FAIL! rc= 7
043 bl512 TBL.sh s8sdm ghgrmp 9s ..................................FAIL! rc= 7
045 bl513 TBL.sh s8pdm aqpgro 3s ..................................FAIL! rc= 7
047 bl553 TBL.sh s64m outfrq24h 2d ................................FAIL! rc= 7
** All baseline comparisons fail except for configurations using adiabatic or
ideal physics.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers:
. There are larger than roundoff changes in the default physics
configurations due to changing the entropy function from single to double
precision.
. There are larger than roundoff changes to runs in which the new dust and
seasalt aerosols are radiatively active.
. There are additional larger than roundoff changes in the MG microphysics
due to bug fixes.
===============================================================
===============================================================
Tag name: cam3_5_28
Originator(s): eaton
Date: Fri Jan 4 16:34:15 MST 2008
One-line Summary: refactor the adiabatic and ideal physics options
Purpose of changes:
. The ideal physics mode was broken when using the FV dycore. The
process of fixing that configuration inspired a refactoring of the
implementation of both adiabatic and ideal physics modes for all
dycores. The main features of the refactoring are:
- The dry atmosphere is implemented by setting the specific humidity to
zero rather than by setting physical constants for water vapor to
those for dry air. The Q field is no longer required on the initial
file for these modes.
- The condensed water species are no longer allocated and transported
around as inert species.
- The FV dycore is not called in a special mode but is called the same
way as when full physics forcings are applied.
- The adiabatic and ideal physics forcings are returned as tendencies and
applied to the state in the identical way that full physics forcings
are applied.
- The energy conservation for the FV dycore which is done in the physics
part of the code is now done in adiabatic and ideal modes identically
to how it's done in full physics mode.
- Eliminate write/read of fields from restart files that aren't used in
ideal/adiabatic modes.
Bugs fixed (include bugzilla ID):
#677 - don't turn on co2 cycle code for CCSM_BGC==CO2A
Describe any changes made to build system:
. add "ideal" and "adiabatic" as valid values of configure's -phys option.
The reason for making adiabatic and ideal physics build time options is
that the condensed water species need not be allocated and transported
for these dry atmosphere configurations. For either of these physics
options only one advected specie is allocated (by default).
Describe any changes made to the namelist:
. No namelist variables changed. But since the adiabatic and ideal physics
modes are specified to configure, build-namelist knows about these modes
and can automatically add the appropriate namelist variable, i.e., either
setting atm_ideal_phys=.true. or atm_adiabatic=.true.. So the use does
not need to worry about setting a namelist variable for adiabatic or
ideal physics modes.
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: didn't check
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do:
models/atm/cam/test/system/config_files/e32adh
models/atm/cam/test/system/config_files/e32idh
models/atm/cam/test/system/config_files/f4adh
models/atm/cam/test/system/config_files/f4idh
models/atm/cam/test/system/config_files/h5x8adm
models/atm/cam/test/system/config_files/h5x8idm
models/atm/cam/test/system/config_files/s32adh
models/atm/cam/test/system/config_files/s32idh
. new config files which use the -phys arg to configure to build CAM
for an adiabatic or ideal physics run.
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/DefaultCAM_INPARM_Namelist.xml
. remove unneeded PHYSICS attribute from ncdata entries so that the initial
files can be used for ideal/adiabatic physics modes
models/atm/cam/bld/DefaultTIMEMGR_INPARM_Namelist.xml
. remove unneeded PHYSICS attributes.
models/atm/cam/bld/cam.cpl6.template
. don't set -co2_cycle flag to configure, and don't set namelist variable
co2_flag=.true. when CCSM_BGC==CO2A
models/atm/cam/bld/ccsm_inparm.pm
. use config_cache.xml to set atm_adiabatic/atm_ideal_phys namelist vars.
models/atm/cam/bld/config_definition.xml
. add ideal and adiabatic as valid_values for phys.
models/atm/cam/bld/configure
. add ideal and adiabatic as valid options for the -phys arg. These
options decrement the number of advected species by 2 (cldliq and
cldice).
models/atm/cam/src/control/cam_comp.F90
. remove actual arg etamid from call to phys_run1
models/atm/cam/src/control/cam_control_mod.F90
. add moist_physics
models/atm/cam/src/control/cam_history.F90
. comment out the 'outfld' event timer
models/atm/cam/src/control/hycoef.F90
. add etamid
models/atm/cam/src/control/runtime_opts.F90
. set moist_physics
models/atm/cam/src/control/startup_initialconds.F90
. allow setting bnd_topo=' ' to override default set in build-namelist
models/atm/cam/src/dynamics/eul/inidat.F90
. initialize q=0 in adiabatic or ideal physics modes
models/atm/cam/src/dynamics/fv/dp_coupling.F90
. remove full_phys dummy arg from d_p_coupling and p_d_coupling
. remove full_phys conditionals -- keep only the full_phys=.true. branches
. remove full_phys actual arg from call to p_d_adjust
. remove dummy arg adiabatic from d_p_coupling
. change timer event tag from 'DP_CPLN: ctem' to DP_CPLN_ctem
models/atm/cam/src/dynamics/fv/dyn_comp.F90
. make convt a module parameter set to .true., so the FV core is always
called in its "full physics" mode.
. remove convt as a dummy arg from dyn_run
. remove addfld calls for constituent tendencies that don't have
corresponding outfld calls.
models/atm/cam/src/dynamics/fv/inidat.F90
. initialize q=0 in adiabatic or ideal physics modes
models/atm/cam/src/dynamics/fv/p_d_adjust.F90
. remove full_phys dummy arg
. remove full_phys conditionals -- assume full_phys=.true.
models/atm/cam/src/dynamics/fv/stepon.F90
. remove full_phys logical.
. don't set zvir=0 for adiabatic/ideal physics. Set q=0 instead.
. use new moist_physics variable to control call to dryairm
(moist_physics = .not. adiabatic .and. .not. ideal_physics)
. remove full_phys as actual arg in calls to dyn_run, d_p_coupling,
p_d_coupling
. moist_phys is only used in call to fv_out to control whether or not
precip stats are output.
. remove adiabatic as actual arg in call to p_d_coupling
models/atm/cam/src/dynamics/sld/inidat.F90
. initialize q=0 in adiabatic or ideal physics modes
models/atm/cam/src/physics/cam/cam_diagnostics.F90
. add outfld call for QRS to diag_init for ideal_phys run
. exit diag_init after basic state variables are added to history in case
of adiabatic or ideal physics run
. don't output RELHUM for dry atm
models/atm/cam/src/physics/cam/check_energy.F90
. modify so that cloud liq/ice don't need to be present.
models/atm/cam/src/physics/cam/constituents.F90
. add init of names for constituent diagnostics to cnst_chk_dim
models/atm/cam/src/physics/cam/initindx.F90
. if ideal/adiabatic phys then don't call the physics register routines
. move the init of constituent diagnostic names to cnst_chk_dim
. replace cph2o by cpwv
. move setting sflxnam, apcnst, hadvnam, vadvnam, fixcnam, tendnam,
ptendnam, dmetendname, tottnam to cnst_chk_dim subroutine in constituents
module.
models/atm/cam/src/physics/cam/physconst.F90
. set rh2o, cpvir, cpwv, zvir as parameters.
. remove rwat (duplicate of rh2o), and cph2o (duplicate of cpwv).
models/atm/cam/src/physics/cam/physics_types.F90
. add conditionals so special tests for cloud liq/ice aren't done when atm
is dry.
. remove unused "use" statements in physics_tend_init
models/atm/cam/src/physics/cam/physpkg.F90
. phys_init
- remove setting special values of physical constants for adiabatic/ideal
physics modes
- return early when ideal_phys or adiabatic
- move call to diag_init to happen before early exit for
ideal_phys/adiabatic modes
- move call to check_energy_init to happen before early exit for
ideal_phys/adiabatic modes
. phys_run1
- remove dummy arg etamid
- remove actual args etamid and cam_in from call to
phys_run1_adiabatic_or_ideal
- move call to check_energy_gmean up so that it gets done whether using
ideal or full physics.
. phys_run1_adiabatic_or_ideal:
- remove call to geopotential_t since all versions of d_p_coupling make
this call.
- use physics_tend_init to initialize phys_tend
- update calling args to tphysidl
- remove dummy args cam_in, etamid
models/atm/cam/src/physics/cam/qneg3.F90
. use same criteria for printing error messages as is used by WACCM
models/atm/cam/src/physics/cam/restart_physics.F90
. don't write fields to restart files that aren't used when running
adiabatic/ideal physics.
models/atm/cam/src/physics/cam/tphysidl.F90
. Remove args for taux and tauy. These are inputs to the atm and are
initialized elsewhere. Don't know why tphysidl would be trying to set
these to zero.
. Remove etamid arg -- use hycoef module
. Have ideal physics algorithms set values in ptend, i.e., set dry static
energy tendency in ptend rather than temperature tendency in tend.
. use update_physics routine to update the output variable tend -- just like
it's done for the full physics routines.
models/atm/cam/src/physics/cam/water_tracers.F90
. replace cph2o by cpwv
models/atm/cam/src/physics/cam/zm_conv.F90
. replace rwat by rh2o
models/atm/cam/test/system/input_tests_master
. move the specification for adiabatic and ideal physics tests out of the
namelist template and into the configure arguments
models/atm/cam/test/system/nl_files/adia
. remove setting of atm_adiabatic -- done by build namelist based on
configuration.
models/atm/cam/test/system/nl_files/idphys
. remove setting of atm_ideal_phys -- done by build namelist based on
configuration.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: All PASS except:
009 bl134 TBL.sh e32adh adia 9s ...................................FAIL! rc= 5
020 bl334 TBL.sh f4adh adia 9s ....................................FAIL! rc= 5
057 bl534 TBL.sh s32adh adia 9s ...................................FAIL! rc= 5
bl134 and bl534: Diffs in fields depending on Q since it's now set to zero, but
identical otherwise.
bl334: Larger than roundoff diffs due to bug fixes.
tempest: All PASS except:
012 bl135 TBL.sh e32idh idphys 9s .................................FAIL! rc= 5
029 bl335 TBL.sh f4idh idphys 9s ..................................FAIL! rc= 5
043 bl535 TBL.sh s32idh idphys 9s .................................FAIL! rc= 5
bl135 and bl535: Q fields are different since now Q=0. Roundoff diffs in the
diagnostic field QRS.
bl335: Larger than roundoff diffs due to bug fixes.
bangkok/lf95: All PASS except:
012 bl153 TBL.sh e64m outfrq24h 2d ................................FAIL! rc= 7
047 bl553 TBL.sh s64m outfrq24h 2d ................................FAIL! rc= 7
These baselines fail due to a roundoff size diff in first appears in the
field DTH.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB except roundoff in spectral dycores
with lf95, and larger that roundoff in FV ideal and adiabatic due to bug
fixes in those configurations.
===============================================================
===============================================================
Tag name: cam3_5_27
Originator(s): mvr, mvertens
Date: 071212
One-line Summary:
CAM will now couple to the surface components every CAM time step in a way
that mirrors the coupling mechanism in cpl6; orbital parameter orb_iyear_ad
will now default to 1990 rather than 1950.
Purpose of changes:
This new coupling mechanism also solves the current problem whereby history
files were not obtained in the last month of production runs. Now when cam
and the surface components couple every cam time step, cam and clm do not
communicate with the driver at nstep=0. Backwards compatibility was maintained
to still permit the coupling of cam with the surface components on the cam
radiation time step. The cam tests now use coupling every time step as the
default approach.
Orbital parameter now matches that used in CCSM.
Bugs fixed (include bugzilla ID): 592
Describe any changes made to build system:
Describe any changes made to the namelist:
- default coupling frequency is value of dtime from atmnl
List any changes to the defaults for the boundary datasets:
- new default finidat, fatmlndfrc, focndomain files for fv1.9x2.5 at gx1v5
Describe any substantial timing or memory changes:
Code reviewed by: mvertens, eaton
List all subroutines eliminated:
D models/atm/cam/test/system/nl_files/outfrq2h
- obsolete namelist file with move to coupling every cam timestep
List all subroutines added and what they do:
A models/atm/cam/test/system/nl_files/outfrq3s
- new namelist file needed for testing when coupling every cam timestep
List all existing files that have been modified, and describe the changes:
M models/ocn/som/mixed_layer.F90
M models/ocn/som/ocn_comp.F90
M models/ocn/som/somini.F90
M models/ocn/som/ocn_comp_mct.F90
M models/atm/cam/src/control/runtime_opts.F90
M models/atm/cam/src/control/ccsm_msg.F90
M models/atm/cam/src/control/con_cam.F90
M models/atm/cam/src/control/atm_comp_mct.F90
M models/atm/cam/src/control/cam_comp.F90
- mods required for new default coupling frequency
M models/atm/cam/test/system/test_driver.sh
- will now test ccsm with latest tag from ccsm3_9_beta series
M models/atm/cam/test/system/input_tests_master
M models/atm/cam/test/system/nl_files/fvvp_lb2
M models/atm/cam/test/system/nl_files/ghgrmp
M models/atm/cam/test/system/nl_files/aqpgro
M models/atm/cam/test/system/nl_files/adia
M models/atm/cam/test/system/nl_files/fv1d_lb2
M models/atm/cam/test/system/nl_files/idphys
M models/atm/cam/test/system/nl_files/co2rmp
M models/atm/cam/test/system/nl_files/fv1d_4tsk
M models/atm/cam/test/system/nl_files/fv2d_4tsk
M models/atm/cam/test/system/nl_files/scm_prep
M models/atm/cam/test/system/nl_files/fv1d_8tsk
M models/atm/cam/test/system/nl_files/no_ttrac
M models/atm/cam/test/system/nl_files/off2x2.5
M models/atm/cam/test/system/nl_files/fv2d_8tsk
M models/atm/cam/test/system/nl_files/ttrac_lb1
M models/atm/cam/test/system/nl_files/microp_mg
M models/atm/cam/test/system/nl_files/ttrac_lb2
M models/atm/cam/test/system/nl_files/ttrac
M models/atm/cam/test/system/nl_files/ttrac_lb3
M models/atm/cam/test/system/nl_files/fvvp_lb0
- namelist and run length settings tweaked for test scripts to reflect new
coupling frequencies
M models/atm/cam/bld/cam.cpl6.template
M models/atm/cam/bld/timemgr_inparm.pm
M models/atm/cam/bld/SeqCCSM_namelist.pm
- scripts modified to include only the required namelist vars for coupled runs
M models/atm/cam/bld/DefaultCLM_INPARM_Namelist.xml
M models/atm/cam/bld/DefaultTIMEMGR_INPARM_Namelist.xml
M models/atm/cam/bld/DefaultDOM_INPARM_Namelist.xml
M models/atm/cam/bld/DefaultCSIM_INPARM_Namelist.xml
M models/atm/cam/bld/DefaultSOM_INPARM_Namelist.xml
- new default files at gx1v5 for fv1.9x2.5; orb_iyear_ad now 1990
M .
M SVN_EXTERNAL_DIRECTORIES
- updated to drv/seq_mct/trunk_tags/drvseq1_0_62 and
clm2/branch_tags/seqccsm_clm3_5_08_tags/seqccsm10_clm3_5_11
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista:
004 bl132 TBL.sh e32sdh ghgrmp 9s .................................FAIL! rc= 7
006 bl133 TBL.sh e32pdh aqpgro 3s .................................FAIL! rc= 7
009 bl134 TBL.sh e32dh adia 9s ....................................FAIL! rc= 7
011 bl151 TBL.sh e64h outfrq24h 2d ................................FAIL! rc= 7
015 bl332 TBL.sh f4sdh ghgrmp 9s ..................................FAIL! rc= 7
017 bl333 TBL.sh f4pdh aqpgro 3s ..................................FAIL! rc= 7
020 bl334 TBL.sh f4dh adia 9s .....................................FAIL! rc= 7
023 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 5
026 bl337 TBL.sh f4dh fv2d_8tsk 9s ................................FAIL! rc= 7
030 bl371 TBL.sh f1.9dm fvvp_lb2 9s ...............................FAIL! rc= 7
034 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7
037 bl354 TBL.sh fm2dh outfrq3s 9s ................................FAIL! rc= 5
040 bl355 TBL.sh fmo2dh off2x2.5 9s ...............................FAIL! rc= 7
044 bl381 TBL.sh fma1.9h outfrq3s+1870_prog_aero 9s ...............FAIL! rc= 5
048 bl382 TBL.sh fmgpa1.9h outfrq3s+1870_control 9s ...............FAIL! rc= 5
052 bl532 TBL.sh s32sdh ghgrmp 9s .................................FAIL! rc= 7
054 bl533 TBL.sh s32pdh aqpgro 3s .................................FAIL! rc= 7
057 bl534 TBL.sh s32dh adia 9s ....................................FAIL! rc= 7
059 bl551 TBL.sh s64h outfrq24h 2d ................................FAIL! rc= 7
061 bl711 TBL.sh h5x8dm adia 9s ...................................FAIL! rc= 7
tempest:
004 bl131 TBL.sh e32dh co2rmp 9s ..................................FAIL! rc= 7
007 bl132 TBL.sh e32sdh ghgrmp 9s .................................FAIL! rc= 7
009 bl133 TBL.sh e32pdh aqpgro 3s .................................FAIL! rc= 7
012 bl135 TBL.sh e32dh idphys 9s ..................................FAIL! rc= 7
014 bl152 TBL.sh e64o outfrq24h 2d ................................FAIL! rc= 7
017 bl314 TBL.sh wg10dh outfrq3s 9s ...............................FAIL! rc= 5
021 bl331 TBL.sh f4gdh co2rmp 9s ..................................FAIL! rc= 7
024 bl332 TBL.sh f4sdh ghgrmp 9s ..................................FAIL! rc= 7
026 bl333 TBL.sh f4pdh aqpgro 3s ..................................FAIL! rc= 7
029 bl335 TBL.sh f4dh idphys 9s ...................................FAIL! rc= 7
031 bl376 TBL.sh f1.9o outfrq24h 2d ...............................FAIL! rc= 7
035 bl531 TBL.sh s32dh co2rmp 9s ..................................FAIL! rc= 7
038 bl532 TBL.sh s32sdh ghgrmp 9s .................................FAIL! rc= 7
040 bl533 TBL.sh s32pdh aqpgro 3s .................................FAIL! rc= 7
043 bl535 TBL.sh s32dh idphys 9s ..................................FAIL! rc= 7
045 bl552 TBL.sh s64o outfrq24h 2d ................................FAIL! rc= 7
calgary/lf95:
004 bl111 TBL.sh e8c8mdm ttrac 9s .................................FAIL! rc= 7
008 bl112 TBL.sh e8sdm ghgrmp 9s ..................................FAIL! rc= 7
010 bl113 TBL.sh e8pdm aqpgro 3s ..................................FAIL! rc= 7
012 bl153 TBL.sh e64m outfrq24h 2d ................................FAIL! rc= 7
016 bl311 TBL.sh f10c8mdm ttrac 9s ................................FAIL! rc= 7
024 bl312 TBL.sh f10sdm ghgrmp 9s .................................FAIL! rc= 7
026 bl313 TBL.sh f10pdm aqpgro 3s .................................FAIL! rc= 7
029 bl315 TBL.sh f10c5dm microp_mg 9s .............................FAIL! rc= 7
032 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7
035 bl377 TBL.sh f1.9m outfrq24h 2d ...............................FAIL! rc= 4
039 bl511 TBL.sh s8c8mdm ttrac 9s .................................FAIL! rc= 7
043 bl512 TBL.sh s8sdm ghgrmp 9s ..................................FAIL! rc= 7
045 bl513 TBL.sh s8pdm aqpgro 3s ..................................FAIL! rc= 7
047 bl553 TBL.sh s64m outfrq24h 2d ................................FAIL! rc= 7
-all baseline tests failed due to a combination of non-b4b changes introduced
and the altering of the test scripts that were not available in the baseline code
CAM tag used for the baseline comparison tests if different than previous
tag: cam3_5_26
Summarize any changes to answers, i.e.,
- what code configurations: ALL
- what platforms/compilers: ALL
- nature of change: larger than roundoff but same climate
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository): seqnoflxave06_cam3_5_24
- platform/compilers: blueice/mpxlf90_r
- configure commandline:
som run:
$cfgdir/configure -dyn fv -res 1.9x2.5 -ocn som
dom run:
$cfgdir/configure -dyn fv -res 1.9x2.5 -gensom
- build-namelist command (or complete namelist):
som run:
$cfgdir/build-namelist -s -case $case -runtype $runtype \
-namelist "&camexp stop_option='nmonths', stop_n=$stop_n \
bndtvs='/fis/cgd/cseg/people/mvr/cam_mydata/som/seqnoflxave06_cam3_5_24_0011-0020_qflux.nc' \
focndomain='/fs/cgd/csm/inputdata/atm/cam/ocnfrac/domain.camocn.1.9x2.5_gx1v4_070807.nc' / &clmexp \
fatmlndfrc='/fs/cgd/csm/inputdata/lnd/clm2/griddata/fracdata_1.9x2.5_gx1v4_060808.nc' \
finidat='/fs/cgd/csm/inputdata/lnd/clm2/inidata_3.1/cam/clmi_0000-09-01_1.9x2.5_gx1v4_c070309.nc' /"
dom run:
$cfgdir/build-namelist -s -case $case -runtype $runtype \
-namelist "&camexp stop_option='nmonths', stop_n=$stop_n start_ymd=101 / &clmexp \
finidat='/fs/cgd/csm/inputdata/lnd/clm2/inidata_3.1/ccsm/clmi_0000-01-01_1.9x2.5_gx1v5_c070523.nc' /"
- MSS location of output:
som: mss:/MVR/csm/seqnoflxave06_cam3_5_24_som4
dom: mss:/MVR/csm/seqnoflxave06_cam3_5_24
- MSS location of control simulations used to validate new climate:
som: mss:/MVR/csm/cam3_5_24_som4
dom: mss:/MVR/csm/cam3_5_24
- URL for AMWG diagnostics output used to validate new climate:
som: mss:/MVR/csm/seqnoflxave06_cam3_5_24_som4/seqnoflxave06_cam3_5_24_som4-cam3_5_24_control_som.tar
dom: mss:/MVR/csm/seqnoflxave06_cam3_5_24/seqnoflxave06_cam3_5_24-cam3_5_24_control.tar
===============================================================
===============================================================
Tag name: cam3_5_26
Originator(s): Jim Edwards
Date: 12/04/2007
One-line Summary: Remove memory consuming arrays from cam_history
Purpose of changes: Memory usage reduction
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
The pflds array of cam_history.F90 was replaced by a linked list. This
allows the masterlist to be allocated as needed instead of statically allocated
to the largest anticipated size. The hash table lookup feature of outfld is retained
by creating an array of pointers into the linked list. The sum result is a slight reduction of
performance of the outfld subroutine with up to 2x reduction in memory high water as measured on
bluevista.
Code reviewed by: Brian Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
add_entry_to_master - add an entry to the masterlist
init_masterlinkedlist - initialize the linkedlist
get_entry_by_name - lookup a masterlist entry by name
List all existing files that have been modified, and describe the changes:
cam_history.F90 - modified to remove pflds from masterlist and from tape structures
these are now allocated structures. pflds is now only used to support
reading fincl and fexcl from the namelist and has been reduced from 10000
to 500. Changes required moving location of the variable varid in the restart
file, so there is a change in restart file format from previous versions.
cam_comp.F90 - call the init_masterlinkedlist function on startup
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: all PASS except:
051 br532 TBR.sh s32sdh ghgrmp 4+2s ...............................FAIL! rc= 11
tempest: all PASS except:
034 br531 TBR.sh s32dh co2rmp 4+2s ................................FAIL! rc= 11
bangkok/lf95: all PASS except:
038 br511 TBR.sh s8c8mdm ttrac 4+2s ...............................FAIL! rc= 11
br532, br531, and br511 are pre-existing failures
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_5_25
Originator(s): eaton
Date: Mon Dec 3 18:50:09 MST 2007
One-line Summary: misc bugfixes
Purpose of changes:
. Merge bugfixes from the ccsm35 branch (between tags ccsm35_03_cam3_5_07
and ccsm35_12_cam3_5_07) onto the trunk.
These fixes change answers for the following CAM configurations.
- All WACCM configurations
- CAM-CHEM configurations "-chem trop_mozart" and "-chem
trop_mozart_aero" with the prognostic dust radiatively passive and the
old prescribed dust radiatively active.
. Makefile.in change in AIX section to support new timing library.
. Update the 1870-2000_control use case to use the version of prescribed
aerosol datasets that contains 10-year climatologies.
. Update to the CAM-CHEM time interpolation codes to deal with input
datasets that contain gaps in a time series of monthly averages.
. Change the check on the sum of cell areas when running with cpl6 from an
absolute to a relative error check.
. Loosen check in interpolate_data module for what percent of interpolated
grid points fall outside the range of the data points. This check was
causing a 31 level model version to fail when interpolating the
noaamisc.r8.nc file for simple GHG chemistry.
. Bugfix in co2_data_flux module (missing mpibcast).
Bugs fixed (include bugzilla ID):
. Bugzilla id 667 - Fix iyear_ad bug in WACCM.
. WACCM bugfix for H2O saturation table.
. CAM-CHEM bug in old prescribed dust optics as described above.
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets:
. new default aerosol dataset for 1870-2000_control use case
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/Makefile.in
. changes in AIX section for new timing lib code
models/atm/cam/bld/SeqCCSM_namelist.pm
. add tracer_cnst_filelist to use case hack.
models/atm/cam/bld/use_cases/1870-2000_control.nl
. use aerosol dataset that contains 10 year climatologies
models/atm/cam/src/chemistry/trop_mozart/chemistry.F90
. changes to namelist variables for 'INTERP_MISSING_MONTHS' option
models/atm/cam/src/chemistry/trop_mozart/mz_aerosols_intr.F90
. add mz_prescribed_dust function for dust aerosol optics bugfix
models/atm/cam/src/chemistry/trop_mozart/tracer_data.F90
. implement 'INTERP_MISSING_MONTHS' interpolation option
models/atm/cam/src/chemistry/waccm_mozart/efield.F90
models/atm/cam/src/chemistry/waccm_mozart/exbdrift.F90
models/atm/cam/src/chemistry/waccm_mozart/mo_apex.F90
. replace iyear_ad by magfield_fix_year
models/atm/cam/src/chemistry/waccm_mozart/mo_jlong.F90
. bugfix - need ptarget declared real
models/atm/cam/src/chemistry/waccm_mozart/mo_sad.F90
. replace with version from tag
waccm_qbo_cam3_5_04_tags/waccm_qbo04_cam3_5_18 (per Stacy).
models/atm/cam/src/control/cam_control_mod.F90
. add magfield_fix_year to set year of magnetic field in WACCM
models/atm/cam/src/control/ccsm_msg.F90
. Change check that cell areas sum to 4*pi from absolute to relative error.
models/atm/cam/src/control/interpolate_data.F90
. change check for %domain in extrapolation region -- was causing failure
of 31 level model when interpolating the noaamisc.r8.nc file for simple
GHG chemistry.
models/atm/cam/src/control/runtime_opts.F90
. add 'INTERP_MISSING_MONTHS' interpolation option for prescribed ozone,
tracer_cnst, and tracer_srcs
. replace 'diag TKE' by 'diag_TKE'
models/atm/cam/src/physics/cam/aer_optics.F90
. bugfix - old prescribed dust needs old dust optics
models/atm/cam/src/physics/cam/cloud_fraction.F90
. WACCM tuning mod
. replace 'diag TKE' by 'diag_TKE'
models/atm/cam/src/physics/cam/co2_data_flux.F90
. BGC bugfix - add missing mpibcast
models/atm/cam/src/physics/cam/physpkg.F90
. WACCM bugfix - call qbo_init
. move call aer_optics_initialize to after chem init (needed for the new
function mz_prescribed_dust)
models/atm/cam/src/physics/cam/radae.F90
. WACCM bugfix for H2O saturation table
models/atm/cam/src/physics/cam/convect_shallow.F90
models/atm/cam/src/physics/cam/phys_control.F90
models/atm/cam/src/physics/cam/vertical_diffusion.F90
. replace 'diag TKE' by 'diag_TKE'
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: All PASS except:
023 bl336 TBL.sh wm4h outfrq2h 4s .................................FAIL! rc= 7
037 bl354 TBL.sh fm2dh outfrq2h 4s ................................FAIL! rc= 7
040 bl355 TBL.sh fmo2dh off2x2.5 4s ...............................FAIL! rc= 7
044 bl381 TBL.sh fma1.9h outfrq2h+1870_prog_aero 4s ...............FAIL! rc= 7
051 br532 TBR.sh s32sdh ghgrmp 4+2s ...............................FAIL! rc= 11
bl336 fails due to waccm bugfixes.
bl354, bl355, bl381 fail due to the dust optics bugfix
br532 is a pre-existing failure
tempest: All PASS except:
034 br531 TBR.sh s32dh co2rmp 4+2s ................................FAIL! rc= 11
br531 is a pre-existing failure
bangkok/lf95: All PASS except
038 br511 TBR.sh s8c8mdm ttrac 4+2s ...............................FAIL! rc= 11
br511 is a pre-existing failure
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB except larger than roundoff changes in
- All WACCM configurations
- CAM-CHEM configurations "-chem trop_mozart" and "-chem
trop_mozart_aero" with the prognostic dust radiatively passive and the
old prescribed dust radiatively active.
===============================================================
===============================================================
Tag name: cam3_5_24
Originator(s): tcraig, dennis, eaton
Date: Tue Nov 27 18:32:47 MST 2007
One-line Summary: MPI fixes
Purpose of changes:
. Update the driver external to camxa01_drvseq1_0_57. This tag is
seqmct1_0_57 with mods from John Dennis to fix the
mpicom problems and fix a memory leak.
. Add the file cam.cpl6.template to CAM's bld directory for CCSM build
system modifications being implemented in the ccsm3_9_betaXX series.
Bugs fixed (include bugzilla ID): see above.
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: tcraig
List all subroutines eliminated: none
List all subroutines added and what they do:
models/atm/cam/bld/cam.cpl6.template
. used by new build system being implemented in ccsm3_9_betaXX series.
List all existing files that have been modified, and describe the changes:
SVN_EXTERNAL_DIRECTORIES
. set external for models/drv/seq_mct to
$REPO/drv/seq_mct/branch_tags/camxa_drvseq1_0_57_tags/camxa01_drvseq1_0_57
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: all PASS except:
051 br532 TBR.sh s32sdh ghgrmp 4+2s ...............................FAIL! rc= 11
tempest: all PASS except:
034 br531 TBR.sh s32dh co2rmp 4+2s ................................FAIL! rc= 11
bangkok/lf95: all PASS except:
038 br511 TBR.sh s8c8mdm ttrac 4+2s ...............................FAIL! rc= 11
br532, br531, and br511 are pre-existing failures
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam3_5_23
Originator(s): Art Mirin
Date: November 17, 2007
One-line Summary: Overlap of dynamics and tracer FV subcycles
Purpose of changes:
This version supports overlap of trac2d and cd_core subcycles. This refers to the
subcycles described by the "do 2000 n=1,n2" loop in dyn_run and has nothing to do
with the "do it=1,nsplit" lower-level subcycling. Each trac2d call (n), other
than the last, is overlapped with the subsequent cd_core 'series' (n+1).
This capability becomes relevant as we go to higher and higher resolution and
operate with more and more tracers. Best results are obtained when the dynamics
and tracer times are comparable (for example, operating with more than the default
number of tracers) and when there is a high degrees of subcycling (for example, at
0.47x0.63 with nspltrac=8).
The controlling namelist variable is ct_overlap. The overlapping trac2d calls are
carried out on the second set of npes_yz processes (npes_yz <= iam < 2*npes_yz).
The tracer arrays are sent to the auxiliary processes prior to the do-2000 loop.
During each subcycle (other than the last), the dp0 array is sent prior to the
cd_core series; arrays cx, cy, mfx, mfy are sent directly from cd_core during
the last call in the series (it=nsplit). At the completion of the last auxiliary
trac2d subcycle (n=n2-1), the updated tracer values are returned to the primary
processes; the last tracer subcycle (n=n2) is carried out on the primary processes.
Communication calls are nonblocking, with attempt to overlap computation to the
extent possible. The CCSM mpi layer (wrap_mpi) is used. Tags with values greater
than npes_xy are chosen to avoid possible interference between the messages sent
from cd_core and the geopk-related transpose messages called from cd_core
thereafter. The auxiliary processes must use values of jfirst, jlast, kfirst,
klast corresponding to their primary process antecedents, whereas by design
those values are (1,0,1,0), resp. (set in spmdinit_dyn). We therefore add
auxiliary subdomain limits to the grid datatype: jfirstct, jlastct, kfirstct,
klastct. For the primary processes, these are identical to the actual subdomain
limits; for the secondary processes, these correspond to the subdomain limits of
the antecedent primary process. These values are communicated to the auxiliary
processes during initialization (spmd_vars_init). During the auxiliary
calculations (and allocations) we temporarily set jfirst equal to jfirstct, etc.,
and when done, restore to the original values. Other information needed by the
auxiliary processes is obtained through the grid datatype.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
New variable ct_overlap (default 0). If nonzero, overlap cd_core and trac2d
subcycles as desdribed above; if nonzero, do not overlap. When overlapping,
the number of processes must be at least as large as twice the size of the
latitude-vertical decomposition.
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Sawyer, Worley, Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
utils/pilgrim:
mod_comm.F90 - compute process index relative to local communicator
src/control:
fv_control_mod.F90 - add ct_overlap flag
runtime_opts.F90 - add ct_overlap namelist variable
src/dynamics/fv:
cd_core.F90 - post communication for arrays needed by trac2d on auxiliary processes
dyn_comp.F90 - principal routine modified to accomplish cd_core/trac2d overlap
dynamics_vars.F90 - communicate subdomain limits to auxiliary trac2d processes
spmd_dyn.F90 - define additional 2yz communicator
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista:
051 br532 TBR.sh s32sdh ghgrmp 4+2s ...............................FAIL! rc= 11
tempest:
034 br531 TBR.sh s32dh co2rmp 4+2s ................................FAIL! rc= 11
bangkok/lf95:
038 br511 TBR.sh s8c8mdm ttrac 4+2s ...............................FAIL! rc= 11
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_5_22
Originator(s): mvr, mirin, eaton
Date: 071116
One-line Summary: added tests for running with fv decomposition; various
other bug fixes
Purpose of changes:
needed to beef up testing of the various fv decomposition settings as features
are being added; updating to newer drv tag fixes PGI testing (done post-tag)
which started failing with cam3_5_19; files need to have the same name as the
module they contain - an offender was introduced in cam3_5_20; build-namelist
has not been placing certain variables that are required in multiple namelists
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
build-namelist will now properly place bndtvs and focndomain into both the ocn
and ice namelists
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: mvr,mirin,eaton
List all subroutines eliminated:
D models/atm/cam/src/physics/cam/ghg_defaults.F90
- file renamed to match the module name it contains
List all subroutines added and what they do:
A + models/atm/cam/src/physics/cam/ghg_data.F90
- new filename to match the module name it contains
A models/atm/cam/test/system/config_files/f1.9dm
- new configuration options file for testing fv decomposition in mpi only
A models/atm/cam/test/system/nl_files/fvvp_lb2
A models/atm/cam/test/system/nl_files/fv1d_lb2
A models/atm/cam/test/system/nl_files/fv1d_4tsk
A models/atm/cam/test/system/nl_files/fv1d_8tsk
A models/atm/cam/test/system/nl_files/fvvp_lb0
- new namelist options files for testing fv decomposition in various modes
List all existing files that have been modified, and describe the changes:
M .
M SVN_EXTERNAL_DIRECTORIES
- updated to drvseq1_0_57 which contain bug fixes to map_ocnaocn_mct.F90 - this
was tripping up PGI testing; updated to latest tag for archiving scripts
M models/atm/cam/test/system/tests_pretag_bangkok
M models/atm/cam/test/system/tests_pretag_bluevista
M models/atm/cam/test/system/tests_posttag_blueice
M models/atm/cam/test/system/tests_posttag_calgary
M models/atm/cam/test/system/tests_posttag_lightning
- added new tests of fv decomposition to default pre/post tag test lists
M models/atm/cam/test/system/test_driver.sh
M models/atm/cam/test/system/TER.sh
M models/atm/cam/test/system/CAM_runcmnd.sh
- mods in support of new fv tests; increased wall clock for lightning testing
M models/atm/cam/test/system/input_tests_master
- new fv tests; bug fix with specification of resolution for ccsm tests
M models/atm/cam/bld/run-ibm.csh
M models/atm/cam/bld/run-pc-cice-docn7.csh
M models/atm/cam/bld/run-pc.csh
M models/atm/cam/bld/run-sgi.csh
M models/atm/cam/bld/run-lightning.csh
- modified the default long-term archive root path
M models/atm/cam/bld/SeqCCSM_namelist.pm
- bug fix for namelist variables needing inclusion on multiple namelists (bndtvs, focndomain)
M models/atm/cam/src/physics/cam/volcanicmass.F90
- small bug fix in volcanic_initialize
M models/atm/cam/src/dynamics/fv/sw_core.F90
- fix to bug introduced in cam3_5_21 where the new coding excluded a needed
pole-related computation
M models/atm/cam/src/dynamics/fv/spmd_dyn.F90
- bug fix in call to mpi_comm_split where the value of 'color' was set
to a negative number
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: All PASS except:
030 bl371 TBL.sh f1.9dm fvvp_lb2 4s ...............................FAIL! rc= 5
051 br532 TBR.sh s32sdh ghgrmp 4+2s ...............................FAIL! rc= 11
- bl371 is a new test, so a failure was expected with the baseline code
- br532 is pre-existing failure
tempest: All PASS except:
034 br531 TBR.sh s32dh co2rmp 4+2s ................................FAIL! rc= 11
- br531 is pre-existing failure
bangkok/lf95: All PASS except:
038 br511 TBR.sh s8c8mdm ttrac 4+2s ...............................FAIL! rc= 11
- br511 is pre-existing failure
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate): b4b
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_5_21
Originator(s): Art Mirin
Date: November 6, 2007
One-line Summary: Addition of Rayleigh friction and Putman filtering (FV)
Purpose of changes:
CAM has been augmented with the capability to apply Rayleigh
friction or additional filtering (FV). The purpose is to counteract
the effects of a polar jet that appears in the upper portions
of the model, particularly at higher resolution. The jet, the
extent of which is believed to be non-physical and progressively
worse as the resolution increases, can force the model to run at
a prohibitively low time step.
Additionally, an update to the Pilgrim parutilitiesmodule.F90,
courtesy of John Dennis, is included. The purpose of the
update is to reduce memory requirements. This is needed
particularly for running on BG/L at moderate to high resolution.
Also included in this update is pointing model/drv/seq_mct to
drvseq1_0_56.
Rayleigh friction: Frictional term is applied with an adjustable
vertical profile based on hyperbolic tangent; lost kinetic energy
is converted to potential energy. Namelist variables are as follows:
rayk0 - vertical index of peak (default 2).
raykrange - determines width of profile; if 0., default width
is chosen; see rayleigh_friction.F90.
raytau0 - approximate decay time (days) at top of model;
default of 0. means no Rayleigh friction.
Filtering: Capability to optionally filter intermediate c-grid winds,
courtesy of Bill Putman. Majority of code changes due to necessity
of computing c-grid winds before call to c_sw. Namelist variable is:
filtcw - yes (1), no (0). Default is 0.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
New namelist variables are (for Rayleigh friction) rayk0, raykrange and
raytau0; and (for Putman filtering) filtcw. See above for descriptions.
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Brian Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
src/physics/cam: rayleigh_friction.F90: new Rayleigh friction routine.
List all existing files that have been modified, and describe the changes:
utils/pilgrim/parutilitiesmodule.F90: added capability (courtesy
of John Dennis) to reduce memory of decompositions; see precompile
flag _SMEMORY, which is presently set at the beginning of
parutilitiesmodule.F90.
src/control:
fv_control_mod.F90: added filtcw flag (1 for Putman filtering,
0 for none; default is 0).
runtime_opts.F90: Read in 4 new namelist variables.
src/physics/cam:
physpkg.F90: call rayleigh_friction_init.
tphysac.F90: call rayleigh_friction_tend and subsequent physics_update.
src/dynamics/fv:
cd_core.F90: call new routine for c-grid winds (d2a2c_winds).
Communicate prior to c_sw call. Eliminate ALT_PFT option.
dyn_comp.F90: Pass through filtcw namelist variable.
dynamics_vars.F90: Pass through filtcw namelist variable.
inital.F90: Pass through filtcw namelist variable.
pft_module.F90: Eliminate ALT_PFT option.
restart_dynamics.F90: Pass through filtcw namelist variable.
sw_core.F90: Add subroutine d2a2c_winds. Eliminate coding that it replaced.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: (per Mat Rothstein)
045 br532 TBR.sh s32sdh ghgrmp 4+2s..............FAIL! rc=11
tempest:
034 br531 TBR.sh s32dh co2rmp 4+2s ..............FAIL! rc=11
bangkok/lf95:
037 br511 TBR.sh s8c8mdm ttrac 4+2s .............FAIL! rc=11
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
Tag name: cam3_5_20
Originator(s): Andrew Conley
Date: 11/01/07
One-line Summary:
Move (CAM) data descriptions of gas constituents to pbuf
Purpose of changes:
Preparation for implementation of Radiative Constituents interface.
Note that CAM/CHEM (MOZART) data gasses have not yet been moved.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Memory for prescribed gas fields is now allocated in the
physics buffer rather than module variables. (Not
substantial?)
Code reviewed by: myself, Brian Eaton
List all subroutines eliminated:
D models/atm/cam/src/physics/cam/rad_cnst_data_interface.F90
Module no longer necessary for override from mozart/chem data
specification.
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/src/physics/cam/initindx.F90
data models of gas distributions registrations with pbuf are called
M models/atm/cam/src/physics/cam/ghg_defaults.F90
O2 added as a data model gas
CO2 is now added as a data gas through this interface
Access to pbuf is passed to this module
ghg's register with pbuf
ghg's update values with timestep (advnce)
local memory deleted for data gasses
remove "ghg_defaults_get_cnst()"
M models/atm/cam/src/physics/cam/radheat.F90
add pbuf to radheat_tend argument list so that waccm override has access.
M models/atm/cam/src/physics/cam/radiation.F90
O2 is now accessed from rad_constituents
remove call to unused/inoperative aerosol_indirect method
generalized co2_col_mean code to work for O2 and CO2
pass o2 to radcswmx
M models/atm/cam/src/physics/cam/radsw.F90
pass o2 rather than specifying it as a data statement
M models/atm/cam/src/physics/cam/chem_surfvals.F90
add o2mmr data specification in this module
M models/atm/cam/src/physics/cam/ozone_data.F90
use pbuf to store data description rather than local memory
add "ozone_data_register"
rm unused ozone_data_final
add "ozone_data_get_cnst"
M models/atm/cam/src/physics/cam/physpkg.F90
rm call to "ghg_defaults_init"
add pbuf to call advnce(phys_state, pbuf)
M models/atm/cam/src/physics/cam/rad_constituents.F90
modified logic to use pbuf representations of data gases,
unless a number of complex overrides are present. Read
the logic for your particular case if interested.
M models/atm/cam/src/physics/cam/prescribed_aerosols.F90
rm unused/inoperable aerosol_indirect subroutine
M models/atm/cam/src/physics/cam/advnce.F90
add access to pbuf
M models/atm/cam/src/physics/waccm/nlte_lw.F90
add access to pbuf so code can access data O3
M models/atm/cam/src/physics/waccm/radheat.F90
add access to pbuf
M models/atm/cam/src/chemistry/trop_mozart/rad_cnst_data_interface.F90
remove calls to CAM versions of data gasses.
M models/atm/cam/src/chemistry/waccm_mozart/chem_surfvals.F90
add data prescription of oxygen
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista:
004 bl132 TBL.sh e32sdh ghgrmp 4s .................................FAIL! rc= 7
006 bl133 TBL.sh e32pdh aqpgro 2s .................................FAIL! rc= 7
011 bl151 TBL.sh e64h outfrq24h 2d ................................FAIL! rc= 7
015 bl332 TBL.sh f4sdh ghgrmp 4s ..................................FAIL! rc= 7
017 bl333 TBL.sh f4pdh aqpgro 2s ..................................FAIL! rc= 7
023 bl336 TBL.sh wm4h outfrq2h 4s .................................FAIL! rc= 7
026 bl337 TBL.sh f4dh fv2d_8tsk 4s ................................FAIL! rc= 7
028 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7
031 bl354 TBL.sh fm2dh outfrq2h 4s ................................FAIL! rc= 7
034 bl355 TBL.sh fmo2dh off2x2.5 4s ...............................FAIL! rc= 7
038 bl381 TBL.sh fma1.9h outfrq2h+1870_prog_aero 4s ...............FAIL! rc= 7
042 bl382 TBL.sh fmgpa1.9h outfrq2h+1870_control 4s ...............FAIL! rc= 7
045 br532 TBR.sh s32sdh ghgrmp 4+2s ...............................FAIL! rc= 11
046 bl532 TBL.sh s32sdh ghgrmp 4s .................................FAIL! rc= 7
053 bl551 TBL.sh s64h outfrq24h 2d ................................FAIL! rc= 7
tempest:
004 bl131 TBL.sh e32dh co2rmp 4s ..................................FAIL! rc= 7
007 bl132 TBL.sh e32sdh ghgrmp 4s .................................FAIL! rc= 7
009 bl133 TBL.sh e32pdh aqpgro 2s .................................FAIL! rc= 7
014 bl152 TBL.sh e64o outfrq24h 2d ................................FAIL! rc= 7
017 bl314 TBL.sh wg10dh outfrq2h 4s ...............................FAIL! rc= 7
021 bl331 TBL.sh f4gdh co2rmp 4s ..................................FAIL! rc= 7
024 bl332 TBL.sh f4sdh ghgrmp 4s ..................................FAIL! rc= 7
026 bl333 TBL.sh f4pdh aqpgro 2s ..................................FAIL! rc= 7
031 bl376 TBL.sh f1.9o outfrq24h 2d ...............................FAIL! rc= 7
034 br531 TBR.sh s32dh co2rmp 4+2s ................................FAIL! rc= 11
035 bl531 TBL.sh s32dh co2rmp 4s ..................................FAIL! rc= 7
038 bl532 TBL.sh s32sdh ghgrmp 4s .................................FAIL! rc= 7
045 bl552 TBL.sh s64o outfrq24h 2d ................................FAIL! rc= 7
calgary/lf95:
004 bl111 TBL.sh e8c8mdm ttrac 4s .................................FAIL! rc= 7
008 bl112 TBL.sh e8sdm ghgrmp 4s ..................................FAIL! rc= 7
010 bl113 TBL.sh e8pdm aqpgro 2s ..................................FAIL! rc= 7
012 bl153 TBL.sh e64m outfrq24h 2d ................................FAIL! rc= 7
016 bl311 TBL.sh f10c8mdm ttrac 4s ................................FAIL! rc= 7
024 bl312 TBL.sh f10sdm ghgrmp 4s .................................FAIL! rc= 7
026 bl313 TBL.sh f10pdm aqpgro 2s .................................FAIL! rc= 7
029 bl315 TBL.sh f10c5dm microp_mg 4s .............................FAIL! rc= 7
032 bl317 TBL.sh f10dm fv2d_4tsk 4s ...............................FAIL! rc= 7
034 bl377 TBL.sh f1.9m outfrq24h 2d ...............................FAIL! rc= 7
037 br511 TBR.sh s8c8mdm ttrac 4+2s ...............................FAIL! rc= 11
038 bl511 TBL.sh s8c8mdm ttrac 4s .................................FAIL! rc= 7
042 bl512 TBL.sh s8sdm ghgrmp 4s ..................................FAIL! rc= 7
046 bl553 TBL.sh s64m outfrq24h 2d ................................FAIL! rc= 7
The following failures were preexisting:
bluevista
045 br532 TBR.sh s32sdh ghgrmp 4+2s ...............................FAIL! rc= 11
tempest
034 br531 TBR.sh s32dh co2rmp 4+2s ................................FAIL! rc= 11
calgary/lf95
037 br511 TBR.sh s8c8mdm ttrac 4+2s ...............................FAIL! rc= 11
Almost all baseline tests fail since this commits a roundoff level change.
This code produces roundoff level changes since we use a mass-weighted
average value of a constant value instead of the constant value for CO2
and O2.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: waccm/mozart prognostic O2 is now used by
radiation
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate): roundoff, verified by cprncdf and visual inspection.
larger than "roundoff" for waccm/mozart
If bitwise differences were observed, how did you show they were no worse
than roundoff? cprncdf on Temperature field
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
Summarize any changes to answers: roundoff
===============================================================
===============================================================
Tag name: cam3_5_19
Originator(s): mvertens
Date: Wed Oct 31 13:59:52 MDT 2007
One-line Summary: updated externals to ccsm4_0_alpha07
Purpose of changes: Bring cam trunk up to date with sequential ccsm
code base
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: see below
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: tcraig
List all subroutines eliminated: none
List all subroutines added and what they do:
A bld/run-ibm-cice-docn7.csh
- temporary script to run with cice-docn7 outside of cpl7 scripts
(recommendation is to always use cpl7 scripts)
List all existing files that have been modified, and describe the changes:
M models/ocn/dom/ocn_comp_mct.F90
M models/ocn/som/ocn_comp_mct.F90
M models/ice/csim4/ice_comp_mct.F90
M models/atm/cam/src/control/atm_comp_mct.F90
- made routines compatible with drvseq1_0_55
M models/atm/cam/src/control/con_cam.F90
- removed call to shr_inputInfo_initRestArchive
M models/atm/cam/src/control/print_memusage.F90
- added BGL flag
M models/atm/cam/bld/configure
- stand-alone: added drv/seq_mct/driver to Filepath
- cpl7: added drv/seq_mct/driver to Filepath and ccsm_seq flag to specify that cam
is being built from cpl7 scripts
M models/atm/cam/bld/config_definition.xml
- added ccsm_seq switch
M SVN_EXTERNAL_DIRECTORIES
- using following new externals (https://svn-ccsm-models.cgd.ucar.edu is implied in path below)
models/drv/seq_mct drv/seq_mct/trunk_tags/drvseq1_0_55
models/lnd/clm2/src clm2/branch_tags/seqccsm_clm3_5_08_tags/seqccsm08_clm3_5_11/src
models/ice/cice/src cice/trunk_tags/cice4_0_20071022
models/ocn/docn7 docn7/branch_tags/drva_docn7_070824_tags/drva05_docn7_070824
models/csm_share csm_share/branch_tags/loga_share3_070903_tags/loga15_share3_071012
models/utils/esmf_wrf_timemgr esmf_wrf_timemgr/trunk_tags/esmf_wrf_timemgr_060616
models/utils/timing timing/trunk_tags/timing_070525
models/utils/mct mct/branch_tags/seqa_MCT2_3_0_070524_tags/seqa06_MCT2_4_2_071026
models/atm/cam/archiving scripts/trunk_tags/scripts_070907b/ccsm_utils/Tools/archiving
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: all tests passed except for
045 br532 TBR.sh s32sdh ghgrmp 4+2s ...............................FAIL! rc= 9
tempest: all tests passed except for
045 br532 TBR.sh s32sdh ghgrmp 4+2s ...............................FAIL! rc= 11
bangkok/lf95: all tests passed except for
045 br532 TBR.sh s32sdh ghgrmp 4+2s ...............................FAIL! rc= 9
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: none, answers are bfb
===============================================================
===============================================================
Tag name: cam3_5_18
Originator(s): Jim Edwards
Date: 10/24/07
One-line Summary: update of homme dycore
Purpose of changes: merge recent development of homme dycore into trunk
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
added support for homme ne21np4 resolution
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: myself, Brian Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
get_dyn_grid_parm added to dyn_grid.F90 for all dycores
given a string name of an integer parameter, return the value
of that parameter, return -1 if that parameter is not valid in a
given dycore.
List all existing files that have been modified, and describe the changes:
Bulk update of HOMME dycore from development in the standalone homme
repository and in branch homme_cam3_5_01
M 6985 models/atm/cam/src/dynamics/homme/restart_dynamics.F90
M 6985 models/atm/cam/src/dynamics/homme/external/filter_mod.F90
M 6985 models/atm/cam/src/dynamics/homme/external/diffusion_mod.F90
M 6985 models/atm/cam/src/dynamics/homme/external/forcing_mod.F90
M 6985 models/atm/cam/src/dynamics/homme/external/prim_advance_mod.F90
M 6985 models/atm/cam/src/dynamics/homme/external/parallel_mod.F90
M 6985 models/atm/cam/src/dynamics/homme/external/reduction_mod.F90
M 6985 models/atm/cam/src/dynamics/homme/external/schedule_mod.F90
M 6985 models/atm/cam/src/dynamics/homme/external/derivative_mod.F90
M 6985 models/atm/cam/src/dynamics/homme/external/prim_si_mod.F90
M 6985 models/atm/cam/src/dynamics/homme/external/domain_mod.F90
M 6985 models/atm/cam/src/dynamics/homme/external/prim_state_mod.F90
M 6985 models/atm/cam/src/dynamics/homme/external/dimensions_mod.F90
M 6985 models/atm/cam/src/dynamics/homme/external/surfaces_mod.F90
M 6985 models/atm/cam/src/dynamics/homme/external/global_norms_mod.F90
M 6985 models/atm/cam/src/dynamics/homme/external/control_mod.F90
M 6985 models/atm/cam/src/dynamics/homme/external/physics_mod.F90
M 6985 models/atm/cam/src/dynamics/homme/external/prim_driver_mod.F90
M 6985 models/atm/cam/src/dynamics/homme/external/interpolate_mod.F90
M 6985 models/atm/cam/src/dynamics/homme/external/element_mod.F90
M 6985 models/atm/cam/src/dynamics/homme/external/time_mod.F90
M 6985 models/atm/cam/src/dynamics/homme/external/hybvcoord_mod.F90
M 6985 models/atm/cam/src/dynamics/homme/external/baroclinic_inst_mod.F90
M 6985 models/atm/cam/src/dynamics/homme/external/prim_advection_mod.F90
M 6985 models/atm/cam/src/dynamics/homme/external/dof_mod.F90
M 6985 models/atm/cam/src/dynamics/homme/external/solver_mod.F90
M 6985 models/atm/cam/src/dynamics/homme/external/namelist_mod.F90
M 6985 models/atm/cam/src/dynamics/homme/external/bndry_mod.F90
M 6985 models/atm/cam/src/dynamics/homme/dycore.F90
M 6990 models/atm/cam/src/dynamics/homme/dyn_grid.F90
M 6985 models/atm/cam/src/dynamics/homme/initcom.F90
M 6985 models/atm/cam/src/dynamics/homme/dp_coupling.F90
M 6985 models/atm/cam/src/dynamics/homme/stepon.F90
M 6985 models/atm/cam/src/dynamics/homme/dyn_comp.F90
Added get_dyn_grid_parm which replaces use pmgrid functionality
and allows the removal of several #ifdef macros in cam_history
M 6985 models/atm/cam/src/dynamics/sld/dyn_grid.F90
M 6985 models/atm/cam/src/dynamics/eul/dyn_grid.F90
M 6985 models/atm/cam/src/dynamics/fv/dyn_grid.F90
M 6985 models/atm/cam/src/control/cam_history.F90
Support for homme grid resolution ne21np4
M 6985 models/atm/cam/bld/DefaultCAM_INPARM_Namelist.xml
M 6985 models/atm/cam/bld/config_horiz_grid.xml
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: All pass except preexisting and homme
055 bl711 TBL.sh h5x8dm adia 4s ...................................FAIL! rc= 7
since have not yet established a climate for the homme dycore no perturbation growth was run.
tempest: All pass except preexisting.
bangkok/lf95: All pass except preexisting.
The following failures were preexisting:
bluevista
045 br532 TBR.sh s32sdh ghgrmp 4+2s ...............................FAIL! rc= 11
tempest
034 br531 TBR.sh s32dh co2rmp 4+2s ................................FAIL! rc= 11
bangkok/lf95
037 br511 TBR.sh s8c8mdm ttrac 4+2s ...............................FAIL! rc= 11
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_5_17
Originator(s): Pat Worley, Brian Eaton, James B. White III
Date: 10/18/07
One-line Summary: introduction of scalable gmean algorithms;
replacement of gavglook by gmean_mass (which calls gmean)
Purpose of changes: improve performance and memory scalability of
field mean calculations in the physics
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: cam_inparm (in atm_in)
has optional parameters phys_float_repro_gmean, phys_fixed_repro_gmean,
phys_nonrepro_gmean, phys_rdiff_warning_gmean (see descriptions below)
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: for large processor counts,
significant improvement in performance
Code reviewed by: Worley and Eaton
List all subroutines eliminated:
gavglook (in physpkg.F90)
- replaced by gmean_mass
List all subroutines added and what they do:
phys_gmean_defaultopts (in phys_gmean.F90)
- get default phys_gmean runtime options
phys_gmean_setopts (in phys_gmean.F90)
- set phys_gmean runtime options
gmean_mass (in phys_gmean.F90)
- compute global mean mass of constituent fields on physics decomposition
- replacement for gavglook; calls gmean
gmean_float_repro (in phy_gmean.F90 - private to module)
- gmean algorithm used in cam3_5_16 (preserved until future clean-up check-in)
- implementation calls gather_chunk_to_field and computes 2D mean on masterproc
- selected by setting phys_float_repro_gmean = .true. in cam_inparm namelist
- if selected, this value is used instead of that produced by other gmean
algorithms
gmean_fixed_repro (in phy_gmean.F90 - private to module)
- fixed-point-based, scalable, reproducible gmean algorithm
- implementation calls mpi_allreduce twice
- on by default, but can be turned on/off by setting
phys_fixed_repro_gmean = .true./.false. in cam_inparm namelist
gmean_nonrepro (in phy_gmean.F90 - private to module)
- floating-point-based, scalable, nonreproducible gmean algorithm
- implementation calls mpi_allreduce once
- on by default, but can be turned on/off by setting
phys_nonrepro_gmean = .true./.false. in cam_inparm namelist
- when used with gmean_fixed_repro, result from gmean_nonrepro used only
to check that gmean_fixed_repro result is accurate; error criterion
set using phys_rdiff_warning_gmean mac_inparm namelist. Default value is
phys_rdiff_warning_gmean = 1.e-13_r8
List all existing files that have been modified, and describe the changes:
atm/cam/src/control/cam_comp.F90
atm/cam/src/physics/cam/physpkg.F90
- gmean_mass replacement for gavglook
atm/cam/src/control/cam_history.F90
- fix of SYPD performance calculation when doing restarts
atm/cam/src/control/ccsm_msg.F90
atm/cam/src/dynamics/eul/dyn_grid.F90
atm/cam/src/dynamics/fv/dyn_grid.F90
atm/cam/src/dynamics/homme/dyn_grid.F90
atm/cam/src/dynamics/sld/dyn_grid.F90
atm/cam/src/physics/cam/phys_grid.F90
- change in get_horiz_grid_d call to return both areas and integration
weights (optional parameters); used to eliminate necessity of dycore-specific
implementations in new gmean algorithms
atm/cam/src/control/mpishorthand.F
atm/cam/src/utils/spmd_utils.F90
- support for fixed-precision gmean algorithm
atm/cam/src/control/runtime_opts.F90
- gmean runtime options
atm/cam/src/physics/cam/phys_gmean.F90
- new and old gmean algorithms and routines to control which to use
- gmean_mass replacement for gavglook
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: ALL PASS except
015 bl332 TBL.sh f4sdh ghgrmp 4s ..................................FAIL! rc= 7
017 bl333 TBL.sh f4pdh aqpgro 2s ..................................FAIL! rc= 7
023 bl336 TBL.sh wm4h outfrq2h 4s .................................FAIL! rc= 7
026 bl337 TBL.sh f4dh fv2d_8tsk 4s ................................FAIL! rc= 7
028 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7
031 bl354 TBL.sh fm2dh outfrq2h 4s ................................FAIL! rc= 7
038 bl381 TBL.sh fma1.9h outfrq2h+1870_prog_aero 4s ...............FAIL! rc= 7
042 bl382 TBL.sh fmgpa1.9h outfrq2h+1870_control 4s ...............FAIL! rc= 7
045 br532 TBR.sh s32sdh ghgrmp 4+2s ...............................FAIL! rc= 11
tempest: ALL PASS except
017 bl314 TBL.sh wg10dh outfrq2h 4s ...............................FAIL! rc= 7
021 bl331 TBL.sh f4gdh co2rmp 4s ..................................FAIL! rc= 7
024 bl332 TBL.sh f4sdh ghgrmp 4s ..................................FAIL! rc= 7
026 bl333 TBL.sh f4pdh aqpgro 2s ..................................FAIL! rc= 7
031 bl376 TBL.sh f1.9o outfrq24h 2d ...............................FAIL! rc= 7
034 br531 TBR.sh s32dh co2rmp 4+2s ................................FAIL! rc= 11
bangkok/lf95: ALL PASS except
016 bl311 TBL.sh f10c8mdm ttrac 4s ................................FAIL! rc= 7
024 bl312 TBL.sh f10sdm ghgrmp 4s .................................FAIL! rc= 7
026 bl313 TBL.sh f10pdm aqpgro 2s .................................FAIL! rc= 7
029 bl315 TBL.sh f10c5dm microp_mg 4s .............................FAIL! rc= 7
032 bl317 TBL.sh f10dm fv2d_4tsk 4s ...............................FAIL! rc= 7
034 bl377 TBL.sh f1.9m outfrq24h 2d ...............................FAIL! rc= 7
037 br511 TBR.sh s8c8mdm ttrac 4+2s ...............................FAIL! rc= 11
The following failures were preexisting:
bluevista
045 br532 TBR.sh s32sdh ghgrmp 4+2s ...............................FAIL! rc= 11
tempest
034 br531 TBR.sh s32dh co2rmp 4+2s ................................FAIL! rc= 11
bangkok/lf95
037 br511 TBR.sh s8c8mdm ttrac 4+2s ...............................FAIL! rc= 11
Changing the gmean calculation is not bit-for-bit when using the
FV dycore (but is for the spectral dycores). Thus all FV comparisons
to the baseline failed.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: FV dycore using new gmean algorithms. Using old
algorithm (phys_float_repro_gmean), the results are bit-for-bit
- what platforms/compilers: all
- nature of change (roundoff; larger than roundoff but same climate; new
climate): roundoff
If bitwise differences were observed, how did you show they were no worse
than roundoff? perturbation growth test on bluevista
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
Tag name: cam3_5_16
Originator(s): mvr, mvertens, mirin
Date: 071009
One-line Summary: bug fix for gensom utility (temporary); cleanup of
getfil routine; line length bug fix for code entered in prev tag; new
resolutions for ocnfrac input files
Purpose of changes: the gensom utility broke with cam3_5_01 where
the string variables passed between components were being truncated
when the max length was exceeded - a temporary fix, extending the
length of the strings was put into the drv code...the getfil routine
still had logic trying to locate files remotely if not present on
local disc - this should have been cleaned up with cam3_5_11...lines
of fortran code cannot exceed 132 chars on tempest compilers - new
code from cam3_5_15 needed to be split into two lines
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
new ocnfrac datasets at 1,1/2,1/4 deg fv
Describe any substantial timing or memory changes:
Code reviewed by: mvr,mvertens
List all subroutines eliminated:
none
List all subroutines added and what they do:
none
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/DefaultDOM_INPARM_Namelist.xml
M models/atm/cam/bld/DefaultCSIM_INPARM_Namelist.xml
M models/atm/cam/bld/DefaultSOM_INPARM_Namelist.xml
- new resolutions added to default settings for ocnfrac files
M models/atm/cam/src/control/ioFileMod.F90
- further cleanup of getfil to only search local disc for file
M models/atm/cam/src/physics/cam/stratiform.F90
- shortened length of line that trips up compiler on tempest (sgi/irix)
M SVN_EXTERNAL_DIRECTORIES
M .
- updated drv branch tag to pull in temporary fix for lengths of strings
passed between components
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
blueice: All PASS except:
045 br532 TBR.sh s32sdh ghgrmp 4+2s ...............................FAIL! rc= 11
tempest: All PASS except:
034 br531 TBR.sh s32dh co2rmp 4+2s ................................FAIL! rc= 11
bangkok/lf95: All PASS except:
037 br511 TBR.sh s8c8mdm ttrac 4+2s ...............................FAIL! rc= 11
The failures are pre-existing.
CAM tag used for the baseline comparison tests if different than previous
tag: needed to use cam3_5_14 on tempest to get around compiler problem
with line length
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_5_15
Originator(s): andrew, eaton
Date: Thu Oct 4 18:44:20 MDT 2007
One-line Summary: fix MG microphysics diagnostics; update CLM datasets; new
use cases
Purpose of changes:
. MG microphysics: Add fixes for diagnostics and add new radar reflectivity
diagnostics.
. Update the CLM datasets for 1 degree and higher resolutions.
. Add new build-namelist -use_case options:
1870-2000_control -- CAM configuration for running 1870-2000 CCSM controls
waccm_1995_climo -- WACCM 1995 climatology
. Modify co2vmr value in 1990_control use case to be consistent with the
value in the CCSM scripts.
. Modify WACCM code not to use iyear_ad because it is not available when
running in the CCSM environment.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets:
. Updated the CLM datasets for 1 degree and higher resolutions
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do:
models/atm/cam/bld/use_cases/1870-2000_control.nl
models/atm/cam/bld/use_cases/waccm_1995_climo.nl
. new use case files
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/build-namelist
. look for valid use case names in the use_cases directory. Now this file
doesn't need to be updated when new use cases are added.
models/atm/cam/bld/DefaultCAM_INPARM_Namelist.xml
. update a WACCM lbc_file entry
models/atm/cam/bld/DefaultCLM_INPARM_Namelist.xml
. update CLM files for 1 degree and higher resolutions
models/atm/cam/bld/SeqCCSM_namelist.pm
. add code to add the data root directory to the relative pathnames
that are specified in the use case namelist files.
models/atm/cam/bld/cam_inparm.pm
models/atm/cam/bld/dom_inparm.pm
. mods so that relative paths can be specified in use case files
models/atm/cam/bld/namelist.pm
. modify code in sub parse that's responsible for finding the line of the
input file that contains the namelist group name.
models/atm/cam/bld/use_cases/1990_control.nl
. update co2vmr to match specification in CCSM
models/atm/cam/src/chemistry/waccm_mozart/exbdrift.F90
. use the model year instead of iyear_ad in call to subsol. iyear_ad is
currently not available to CAM when running in CCSM.
models/atm/cam/src/control/srfxfer.F90
. always set cam_out%co2diag; not just when co2_transport==.true.
models/atm/cam/src/physics/cam/cldwat2m.F90
. mods for radar reflectivity calculation
models/atm/cam/src/physics/cam/prescribed_aerosols.F90
. fix units of aerosol number concentration fields in history files
models/atm/cam/src/physics/cam/stratiform.F90
. add radar reflectivity fields
. use state instead of state1 in call to collect_sw_aer_masses
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: All PASS except:
023 bl336 TBL.sh wm4h outfrq2h 4s .................................FAIL! rc= 7
045 br532 TBR.sh s32sdh ghgrmp 4+2s ...............................FAIL! rc= 11
tempest: All PASS except:
034 br531 TBR.sh s32dh co2rmp 4+2s ................................FAIL! rc= 11
bangkok/lf95: All PASS except:
037 br511 TBR.sh s8c8mdm ttrac 4+2s ...............................FAIL! rc= 11
bl336 is failing due to using the model year rather than iyear_ad in the
WACCM code for the ExB calculation.
br511, br531, br532 are pre-existing failures.
CAM tag used for the baseline comparison tests if different than previous
tag: cam3_5_10
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam3_5_14
Originator(s): mvr
Date: 070925
One-line Summary: all writes of model log output to unit 6 were replaced
with writes to a variable logical unit - allows for a move to a unique
unit number with assigned log file in the future
Purpose of changes: sorting through the log output of a coupled
run was complicated by various components all dumping to stdout;
assigning unique unit numbers with associated log files for each
component will clear some of the confusion
Bugs fixed (include bugzilla ID): 447 (partially)
Describe any changes made to build system:
- added option to configure to indicate that CAM is being built from within
sequential CCSM scripts
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: myself, mvertens, eaton
List all subroutines eliminated:
D models/atm/cam/src/control/linebuf_stdout.c
- removed obsolete code
List all subroutines added and what they do:
A models/atm/cam/src/control/cam_logfile.F90
- added new module for managing the logical unit of cam's output log
List all existing files that have been modified, and describe the changes:
- too many to list!!
- every file with writes to stdout was modified to use a variable unit number
instead...what follows are mods made in addition to these
M models/atm/cam/bld/configure
M models/atm/cam/bld/config_definition.xml
- added option to indicate that CAM is being built from within sequential
CCSM scripts
M models/atm/cam/src/control/con_cam.F90
M models/atm/cam/src/control/atm_comp_mct.F90
M models/atm/cam/src/control/cam_comp.F90
- moved call to spmdinit up one level to ease the redirection of share
output to cam's log output
M models/atm/cam/src/physics/cam/convect_ke_intr.F90
M models/atm/cam/src/physics/cam/icarus_scops.F90
- replaced usage of hardcoded unit numbers with calls to share code for
unique unit numbers
M input_tests_master
- modified resolutions used for testing of ccsm through cam
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: All PASS except:
045 br532 TBR.sh s32sdh ghgrmp 4+2s ...............................FAIL! rc= 11
tempest: All PASS except:
034 br531 TBR.sh s32dh co2rmp 4+2s ................................FAIL! rc= 11
calgary/lf95: All PASS except:
037 br511 TBR.sh s8c8mdm ttrac 4+2s ...............................FAIL! rc= 11
The failures are pre-existing.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_5_13
Originator(s): eaton
Date: Thu Sep 20 17:45:00 MDT 2007
One-line Summary: Reduce memory use
Purpose of changes:
. Reduce memory use.
. Update the csm_share tag to trunk_tags/share3_070918
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes:
. Memory use in the FV core was reduced by eliminating the allocation of 4
unnecessary 3D arrays in inidat.
. Memory use was reduced when absorptivity/emissivity restart files are
read/written by breaking the read and write of the abs/ems arrays into
smaller pieces so no more than a single 3D global array of temporary
storage is required. Previously PLEVP 3D arrays were being
simultaneously allocated.
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
SVN_EXTERNAL_DIRECTORIES
. change csm_share to trunk_tags/share3_070918
models/atm/cam/src/dynamics/fv/inidat.F90
. remove 4 unneeded 3D arrays
models/atm/cam/src/physics/cam/restart_physics.F90
. Break the read and write of the abs/ems arrays into smaller pieces so no
more than a single 3D global array of temporary storage is required.
Previously PLEVP 3D arrays were being simultaneously allocated.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: All PASS except:
045 br532 TBR.sh s32sdh ghgrmp 4+2s ...............................FAIL! rc= 11
tempest: All PASS except:
034 br531 TBR.sh s32dh co2rmp 4+2s ................................FAIL! rc= 11
bangkok/lf95: All PASS except:
037 br511 TBR.sh s8c8mdm ttrac 4+2s ...............................FAIL! rc= 11
The failures are pre-existing.
CAM tag used for the baseline comparison tests if different than previous
tag: cam3_5_10
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam3_5_12
Originator(s): jet, mariana
Date: Tue Sep 18 15:00:16 MDT 2007
One-line Summary: SCM fixes, brought clm external up to latest commit,
added focndomain changes.
Purpose of changes: Fixed outstanding problem with SCM not being able
to read land initial data, also refactored the way scm mode
determines whether land/ocn/ice is present by using new focndomain
boundary dataset. Required analogous changes to drv,share,clm
externals.
Bugs fixed (include bugzilla ID): 480, 612
Describe any changes made to build system: new focndomain file is now
provided by configure
Describe any changes made to the namelist: ocn_in has focndomain boundary
data file
List any changes to the defaults for the boundary datasets: new focndomain
file is now required to have ocean fraction data.
Describe any substantial timing or memory changes:none
Code reviewed by:jet
List all subroutines eliminated:
D models/atm/cam/src/control/setlatlonidx.F90
moved this file's functionality into a share module that all
components can use.
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/ocn/dom/sst_data.F90 - uses new shr_scam_mod_file
M models/ocn/dom/ocn_comp.F90 - fixes for focndomain
M models/ocn/dom/ocn_comp_mct.F90 - fixes for focndomain
M models/ocn/som/ocn_comp.F90 - fixes for focndomain
M models/ocn/som/ocn_comp_mct.F90 - fixes for focndomain
M models/ice/csim4/ice_data.F90 - uses new shr_scam_mod_file
M models/ice/csim4/ice_comp.F90 - fixes for focndomain
M models/ice/csim4/ice_comp_mct.F90 - fixes for focndomain
M models/atm/cam/bld/som_inparm.pm - fixes for focndomain
M models/atm/cam/bld/configure - fixes for focndomain
M models/atm/cam/bld/run-pc-cice-docn7.csh - fixes for focndomain
M models/atm/cam/bld/DefaultCAM_INPARM_Namelist.xml - fixes for focndomain
M models/atm/cam/bld/cam_inparm.pm - fixes for focndomain
M models/atm/cam/bld/DefaultDOM_INPARM_Namelist.xml - fixes for focndomain
M models/atm/cam/bld/dom_inparm.pm - fixes for focndomain
M models/atm/cam/bld/DefaultCSIM_INPARM_Namelist.xml - fixes for focndomain
M models/atm/cam/bld/DefaultSOM_INPARM_Namelist.xml - fixes for focndomain
M models/atm/cam/bld/csim_inparm.pm - fixes for focndomain
M models/atm/cam/src/control/runtime_opts.F90 - fixes for focndomain
M models/atm/cam/src/control/ncdio_atm.F90 - scam bug fix
M models/atm/cam/src/control/scamMod.F90 - uses new shr_scam_mod_file
M models/atm/cam/src/control/startup_initialconds.F90 - merge marianas mods for focndomain
M models/atm/cam/src/control/filenames.F90 - merge marianas mods for focndomain
M models/atm/cam/src/control/cam_history.F90 - uses new shr_scam_mod_file
M models/atm/cam/src/physics/cam/prescribed_aerosols.F90 - uses new shr_scam_mod_file
M models/atm/cam/src/physics/cam/gw_drag.F90 - scam bug fix
M models/atm/cam/src/dynamics/eul/getinterpnetcdfdata.F90 - uses new shr_scam_mod_file
M models/atm/cam/src/dynamics/eul/iop.F90 - scam refactor changes
M SVN_EXTERNAL_DIRECTORIES - point to new external directories
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
All ccsm tests fail due to new external shr code requirements for clm, atm, and, drv
bluevista:
045 br532 TBR.sh s32sdh ghgrmp 4+2s ...............................FAIL! rc= 11
tempest:
034 br531 TBR.sh s32dh co2rmp 4+2s ................................FAIL! rc= 11
bangkok/lf95:
037 br511 TBR.sh s8c8mdm ttrac 4+2s ...............................FAIL! rc= 11
CAM tag used for the baseline comparison tests if different than previous
tag: cam3_5_08
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff? bfb
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_5_11
Originator(s): mvr
Date: 7 Sep 2007
One-line Summary: replacing inline archiving of model output with
scripts for archiving after completion of model run
Purpose of changes: it's been a long-time desire to make the archiving
of model output a post-processing activity, similar to running with ccsm;
ideally, we'd like both systems to use the same scripts; this should
allow for adding support for more machines and storage systems, inside
and eventually outside ncar...
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: removed share code from
filepath for coupled runs
Describe any changes made to the namelist: archiving options removed
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: myself, eaton
List all subroutines eliminated:
D models/atm/cam/test/system/nl_files/off2x2.5p
- removed namelist options file no longer used by test scripts
List all subroutines added and what they do:
none
List all existing files that have been modified, and describe the changes:
M SVN_EXTERNAL_DIRECTORIES
M .
- added external for archiving scripts which will reside in the ccsm scripts;
updated to new share tag which has cleanup of restart pointer files for drv
M models/ocn/dom/ocn_comp.F90
M models/ocn/dom/ocn_comp_mct.F90
M models/ocn/som/ocn_comp.F90
M models/ocn/som/ocn_comp_mct.F90
M models/ice/csim4/ice_comp.F90
M models/ice/csim4/ice_comp_mct.F90
M models/atm/cam/src/control/runtime_opts.F90
M models/atm/cam/src/control/ccsm_msg.F90
M models/atm/cam/src/control/atm_comp_mct.F90
M models/atm/cam/src/control/filenames.F90
M models/atm/cam/src/control/cam_restart.F90
M models/atm/cam/src/control/ioFileMod.F90
M models/atm/cam/src/control/cam_comp.F90
M models/atm/cam/src/control/cam_history.F90
M models/atm/cam/src/physics/cam/restart_physics.F90
M models/atm/cam/bld/NamelistsDescriptions.xml
- removed code related to inline archiving of model output; now forcing restart
pointer files to reside in run directory
M models/atm/cam/test/system/test_driver.sh
- made 'regular' the default queue for bluevista testing; increased wall clock
limit for lightning testing; removed redundant setting of LD_LIBRARY_PATH on bangkok
M models/atm/cam/test/system/nl_files/ghgrmp
M models/atm/cam/test/system/nl_files/aqpgro
M models/atm/cam/test/system/nl_files/adia
M models/atm/cam/test/system/nl_files/outfrq24h
M models/atm/cam/test/system/nl_files/idphys
M models/atm/cam/test/system/nl_files/co2rmp
M models/atm/cam/test/system/nl_files/fv2d_4tsk
M models/atm/cam/test/system/nl_files/scm_b4b_o1
M models/atm/cam/test/system/nl_files/scm_prep
M models/atm/cam/test/system/nl_files/no_ttrac
M models/atm/cam/test/system/nl_files/off2x2.5
M models/atm/cam/test/system/nl_files/fv2d_8tsk
M models/atm/cam/test/system/nl_files/outfrq2h
M models/atm/cam/test/system/nl_files/ttrac_lb1
M models/atm/cam/test/system/nl_files/microp_mg
M models/atm/cam/test/system/nl_files/ttrac_lb2
M models/atm/cam/test/system/nl_files/ttrac
M models/atm/cam/test/system/nl_files/ttrac_lb3
- removed specifications of restart pointer files in test scripts and any archive settings
M models/atm/cam/test/system/input_tests_master
- post-tag testing of ccsm will now exercise 1870_control compset
M models/atm/cam/bld/configure
- removed share code from filepath in support of having csm_share build as library in
coupled mode
M models/atm/cam/bld/run-ibm.csh
M models/atm/cam/bld/run-pc-cice-docn7.csh
M models/atm/cam/bld/run-pc.csh
M models/atm/cam/bld/run-sgi.csh
M models/atm/cam/bld/run-lightning.csh
- enabled template run scripts with calls to new archive scripts
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista:
045 br532 TBR.sh s32sdh ghgrmp 4+2s ...............................FAIL! rc= 11
tempest:
034 br531 TBR.sh s32dh co2rmp 4+2s ................................FAIL! rc= 11
bangkok/lf95:
037 br511 TBR.sh s8c8mdm ttrac 4+2s ...............................FAIL! rc= 11
- branch tests br511, br531, br532 failed in the previous tag
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_5_10
Originator(s): Art Mirin, Pat Worley
Date: 9/5/07
One-line Summary: Enable different numbers of processes for different portions of the calculation
Purpose of changes: Improve scalability and make minor changes to physics
tuning for 0.5-deg and 0.25-deg FV.
CAM is now able to utilize different numbers of processes for different
portions of the calculation. More specifically:
(1) CAM/FV can run with a smaller latitude-vertical decomposition than
longitude-latitude decomposition.
(2) CAM can run with more physics processes than dynamics processes, or with
inactive processes. Inactive physics processes are enabled by assigning zero chunks.
The current restrictions are as follows: If phys_loadbalance equals 0 or 3, then the
physics process count will equal the dynamics process count, and additional processes will be
idle throughout. If phys_loadbalance equals 2, all processes will be assigned to
the physics and a prescribed subset to the dynamics. The current capability holds for FV,
EUL and SLD dynamical cores.
Also, a few physics tuning parameters relevant to running the FV dycore at
0.5-deg or finer have been modified.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist: added nspltrac, dyn_npes, dyn_npes_stride
nspltrac: number of tracer timesteps per physics timestep for FV; if 0,
resolution-dependent defaults are enacted
dyn_npes: number of dynamics processes for EUL/SLD
dyn_npes_stride: stride for dynamics processes for EUL/SLD
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes: ability to use more physics
than dynamics processes improves throughput of physics; ability to use more
lon-lat processes than lat-vert processes for FV improves throughput of parts of FV dynamics
Code reviewed by: Brian Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
src/physics/cam:
check_energy.F90 - inactive physics processes
cldwat.F90 - tuning; icritc for 0.5- and 0.25-deg FV
cloud_fraction.F90 - tuning; premit for 0.5- and 0.25-deg FV
gw_drag.F90 -writing effgw_oro and kwv to stdout
hk_conv.F90 - tuning; c0 for 0.5- and 0.25-deg FV, and writing cmftau and c0 to stdout
phys_grid.F90 -chunk creation and allocation changes to support inactive processes in
the dynamics; use correct communicator and dimensions with mod_comm options in
transpose routines.
physpkg.F90 - automatic => allocatable arrays in gavglook (Note: if/when gavglook is
optimized a la phys_gmean, this will be unnecessary)
src/control:
cam_comp.F90 - fix timers bug
fv_control_mod.F90 - declare nspltrac for tracer subcycling
runtime_opts.F90 - add namelist variables for active dynamics processes (EUL/SLD) and
tracer subcycling (FV)
src/utils:
spmd_utils.F90 - trivial clean-up of appearance and comments
src/dynamics/fv:
FVperfmod.F90 - add communicators to barrier calls
benergy.F90 - add communicator to calls
cd_core.F90 - flexible dynamics decompositions and add communicators to calls
d2a3d_ijk.F90 - add communicator to calls
d2a3d_ikj.F90 - add communicator to calls
diag_module.F90 - flexible dynamics decompositions and add communicator to calls
dp_coupling.F90 - flexible dynamics and physics decompositions
dyn_comp.F90 - flexible dynamics decompositions and add communicator to calls
dynamics_vars.F90 - flexible dynamics decompositions, add communicator to calls, and
tracer subcycling
epvd.F90 - add communicator to calls
fv_prints.F90 - flexible dynamics and physics decompositions
geopk.F90 - flexible dynamics decompositions and add communicator to calls; fix
placement of precompile flag
inidat.F90 - flexible dynamics decompositions and add communicator to calls
inital.F90 - tracer subcycling
io_dist.F90 - add communicator to calls
mapz_module.F90 - add diagnostics
mean_module.F90 - flexible dynamics decompositions
metdata.F90 - add communicator to calls
pfixer.F90 - add communicator to calls
pmgrid.F90 - remove superfluous decompositions
restart_dynamics.F90 - flexible dynamics decompositions and tracer subcycling
spmd_dyn.F90 - flexible dynamics decompositions, adding new EUL/SLD namelist
arguments in spmd_dyn_defaultopts and spmd_dyn_setopts (as no-ops)
stepon.F90 - add communicator to calls and tracer subcycling
te_map.F90 - add communicator to calls
trac2d.F90 - add communicator to calls
uv3s_update.F90 - add communicator to calls
src/dynamics/eul:
bndexch.F90 - support for inactive processes in dynamics (communication restricted to
active processes)
comspe.F90 - support for inactive processes in dynamics (wave number decomposition
only over active processes)
realloc4.F90 - support for inactive processes in dynamics (communication restricted to
active processes for grid-point/spectral transposes)
realloc7.F90 - support for inactive processes in dynamics (array initialization)
scan2.F90 - support for inactive processes in dynamics (array initialization)
scanslt.F90 - support for inactive processes in dynamics (communication restricted to
active processes for halo updates)
spmd_dyn.F90 - support for inactive processes in dynamics (defining decomposition,
assigning zero work to idle processes, and adding new namelist options to control active
process placement)
src/dynamics/sld:
comspe.F90 - support for inactive processes in dynamics (wave number decomposition
only over active processes)
realloc4.F90 - support for inactive processes in dynamics (communication restricted to
active processes for grid-point/spectral transposes)
realloc7.F90 - support for inactive processes in dynamics (array initialization)
scan2.F90 - support for inactive processes in dynamics (array initialization)
scanslt.F90 - support for inactive processes in dynamics (defining decomposition,
communication restricted to active processes for halo updates)
spmd_dyn.F90 - support for inactive processes in dynamics (defining decomposition,
assigning zero work to idle processes, and adding new namelist options to control active
process placement)
src/dynamics/homme:
spmd_dyn.F90 - adding new EUL/SLD namelist arguments in spmd_dyn_defaultopts
and spmd_dyn_setopts (as no-ops)
models/utils/pilgrim:
mod_comm.F90 - numerous changes to support different sized decompositions and
multiple communicators
parutilitesmodule.F90 - numerous changes to support different sized decompositions and
multiple communicators
puninterfaces.F90 - add communicator to calls
models/ocn/dom:
ocn_spmd.F90 - corrected comments
models/ice/csim4:
ice_spmd.F90 - corrected comments
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: br532 (known failure)
tempest: br531 (known failure)
bangkok/lf95: br511 (known failure)
CAM tag used for the baseline comparison tests if different than previous
tag: 3_5_08; actually, this tag 3_5_10 and 3_5_09 are identical except
for the ChangeLog file
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
Answers are BFB except for 0.5-deg and 0.25-deg FV due to different
tuning parameters.
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
Tag name: cam3_5_09
See tag cam3_5_10; only difference between 3_5_10 and 3_5_09 is ChangeLog file.
===============================================================
Tag name: cam3_5_08
Originator(s): andrew, eaton
Date: Wed Aug 15 15:44:58 MDT 2007
One-line Summary: add Morrison-Gettleman microphysics code; 2-bin sea salt optics
Purpose of changes:
. The Morrison-Gettleman microphysics code has been added as an option to
replace the default Rasch-Kristjansson microphysics. To enable this
option:
-- configure must add 2 additional constituents. If no other optional
constituents are enabled then the number of constituents needs to be
increased from its default value of 3 to 5 via setting the configure
arg "-nadv 5".
-- set the namelist variable microp_scheme='MG'
. An option to produce radiation fields used to diagnose the indirect
effect has been added. It is off by default. To enable it set the
namelist variable indforce=.true.
. The aerosol optics for sea salt has been changed to use 2 sea salt size
bins. The previous sea salt optics assumed only 1 bin. This results in
larger than roundoff changes to the climate simulation.
. Add restart and baseline tests to the regression suite for running with
the MG microphysics enabled.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: added microp_scheme and indforce
List any changes to the defaults for the boundary datasets:
. add default CLM fsurdat files for 0.23x0.31 and 0.47x0.63 resolutions
Describe any substantial timing or memory changes: none (new microphysics
not on by default)
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do:
models/atm/cam/src/physics/cam/cldwat2m.F90
. new MG microphysics parameterization
models/atm/cam/test/system/config_files/f10c5dm
models/atm/cam/test/system/nl_files/microp_mg
. new tests for MG microphysics
List all existing files that have been modified, and describe the changes:
models/atm/cam/src/physics/cam/cloud_fraction.F90
models/atm/cam/src/physics/cam/phys_control.F90
models/atm/cam/src/physics/cam/physics_types.F90
models/atm/cam/src/physics/cam/pkg_cldoptics.F90
models/atm/cam/src/physics/cam/radsw.F90
models/atm/cam/src/physics/cam/stratiform.F90
. mods for MG microphysics
models/atm/cam/src/chemistry/trop_mozart/mz_aerosols_intr.F90
models/atm/cam/src/physics/cam/aer_optics.F90
models/atm/cam/src/physics/cam/aerosol_index.F90
models/atm/cam/src/physics/cam/aerosol_intr.F90
models/atm/cam/src/physics/cam/aerosol_radiation_interface.F90
models/atm/cam/src/physics/cam/prescribed_aerosols.F90
models/atm/cam/src/physics/cam/seasalt_intr.F90
. mods for 2-bin sea salt optics
models/atm/cam/src/control/runtime_opts.F90
. add namelist variables microp_scheme and indforce
models/atm/cam/src/physics/cam/cam_diagnostics.F90
. clarify long_names of precip fields
models/atm/cam/src/physics/cam/radiation.F90
. add radiation calc for indirect effect diagnostics
models/atm/cam/src/physics/cam/sulchem.F90
. limiter to keep pH positive
models/atm/cam/src/physics/cam/wetdep.F90
. bound variable fracev
models/atm/cam/src/physics/cam/zm_conv.F90
. remove unused variable conke
models/atm/cam/bld/DefaultCLM_INPARM_Namelist.xml
. add fsurdat defaults for 0.23x0.31 and 0.47x0.63 resolutions
models/atm/cam/bld/Makefile.in
. modify linux section to allow not specifying the mpi location when
SPMD=TRUE
models/atm/cam/test/system/input_tests_master
models/atm/cam/test/system/tests_posttag_calgary
models/atm/cam/test/system/tests_posttag_lightning
models/atm/cam/test/system/tests_pretag_bangkok
. add new tests for MG microphysics
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: all PASS except
004 bl132 TBL.sh e32sdh ghgrmp 4s .................................FAIL! rc= 7
006 bl133 TBL.sh e32pdh aqpgro 2s .................................FAIL! rc= 7
011 bl151 TBL.sh e64h outfrq24h 2d ................................FAIL! rc= 7
015 bl332 TBL.sh f4sdh ghgrmp 4s ..................................FAIL! rc= 7
017 bl333 TBL.sh f4pdh aqpgro 2s ..................................FAIL! rc= 7
023 bl336 TBL.sh wm4h outfrq2h 4s .................................FAIL! rc= 7
026 bl337 TBL.sh f4dh fv2d_8tsk 4s ................................FAIL! rc= 7
028 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7
031 bl354 TBL.sh fm2dh outfrq2h 4s ................................FAIL! rc= 7
034 bl355 TBL.sh fmo2dh off2x2.5 4s ...............................FAIL! rc= 7
038 bl381 TBL.sh fma1.9h outfrq2h+1870_prog_aero 4s ...............FAIL! rc= 7
042 bl382 TBL.sh fmgpa1.9h outfrq2h+1870_control 4s ...............FAIL! rc= 7
045 br532 TBR.sh s32sdh ghgrmp 4+2s ...............................FAIL! rc= 11
046 bl532 TBL.sh s32sdh ghgrmp 4s .................................FAIL! rc= 7
053 bl551 TBL.sh s64h outfrq24h 2d ................................FAIL! rc= 7
tempest: all PASS except
004 bl131 TBL.sh e32dh co2rmp 4s ..................................FAIL! rc= 7
007 bl132 TBL.sh e32sdh ghgrmp 4s .................................FAIL! rc= 7
009 bl133 TBL.sh e32pdh aqpgro 2s .................................FAIL! rc= 7
014 bl152 TBL.sh e64o outfrq24h 2d ................................FAIL! rc= 7
017 bl314 TBL.sh wg10dh outfrq2h 4s ...............................FAIL! rc= 7
021 bl331 TBL.sh f4gdh co2rmp 4s ..................................FAIL! rc= 7
024 bl332 TBL.sh f4sdh ghgrmp 4s ..................................FAIL! rc= 7
026 bl333 TBL.sh f4pdh aqpgro 2s ..................................FAIL! rc= 7
031 bl376 TBL.sh f1.9o outfrq24h 2d ...............................FAIL! rc= 7
034 br531 TBR.sh s32dh co2rmp 4+2s ................................FAIL! rc= 11
035 bl531 TBL.sh s32dh co2rmp 4s ..................................FAIL! rc= 7
038 bl532 TBL.sh s32sdh ghgrmp 4s .................................FAIL! rc= 7
045 bl552 TBL.sh s64o outfrq24h 2d ................................FAIL! rc= 7
bangkok/lf95: all PASS except
004 bl111 TBL.sh e8c8mdm ttrac 4s .................................FAIL! rc= 7
008 bl112 TBL.sh e8sdm ghgrmp 4s ..................................FAIL! rc= 7
010 bl113 TBL.sh e8pdm aqpgro 2s ..................................FAIL! rc= 7
012 bl153 TBL.sh e64m outfrq24h 2d ................................FAIL! rc= 7
016 bl311 TBL.sh f10c8mdm ttrac 4s ................................FAIL! rc= 7
024 bl312 TBL.sh f10sdm ghgrmp 4s .................................FAIL! rc= 7
026 bl313 TBL.sh f10pdm aqpgro 2s .................................FAIL! rc= 7
029 bl315 TBL.sh f10c5dm microp_mg 4s .............................FAIL! rc= 5
032 bl317 TBL.sh f10dm fv2d_4tsk 4s ...............................FAIL! rc= 7
034 bl377 TBL.sh f1.9m outfrq24h 2d ...............................FAIL! rc= 5
037 br511 TBR.sh s8c8mdm ttrac 4+2s ...............................FAIL! rc= 11
038 bl511 TBL.sh s8c8mdm ttrac 4s .................................FAIL! rc= 7
042 bl512 TBL.sh s8sdm ghgrmp 4s ..................................FAIL! rc= 7
046 bl553 TBL.sh s64m outfrq24h 2d ................................FAIL! rc= 7
These are the expected results. The baseline tests fail due to the change
in the sea salt optical treatment. The SLD branch test failures are a
pre-existing problem.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers:
- what code configurations: default
- what platforms/compilers: all
- nature of change (roundoff; larger than roundoff but same climate; new
climate): larger than roundoff, but same climate
10 year runs were done (by Andrew Gettleman) with cam3_5_04 and
nmicro2_3_cam3_5_04 in their default configurations. The only difference
between these runs is the treatment of sea salt optical properties (use 2
size bins instead of 1). The output from the AMWG diagnostics package is
here:
http://www.cgd.ucar.edu/cms/andrew/diag/nmicro2_3_cam3_5_04_2-cam3_5_04
===============================================================
===============================================================
Tag name: cam3_5_07
Originator(s): eaton, mirin, fvitt
Date: Tue Aug 14 13:38:57 MDT 2007
One-line Summary: increase max number of history fields; modify
tracer_suite; add namelist utility; fix cyclic mode
Purpose of changes:
. Increase size of master field list. This was needed to allow running a
CCSM configuration that turns on full trop_mozart chemistry.
Also made a couple of changes to eliminate adding names to the master
field list that have no corresponding outfld calls.
. Fix a bug in trop_mozart data reading code. This affects performance but
the answers are identical.
. Modifications to the tracers_suite module to allow easily creating many
test tracers. The original 5 tracers (which have no emission, source, or
sink terms) are copied as many time as necessary to produce the requested
number of test tracers.
This option is used by setting the configure -nadv option to the total
number of advected tracers, and by setting the namelist variable
tracers_flag=.true. The total number of test tracers will be the number
set by the -nadv option, minus the number present by default (the water
substance tracers) and the number registered by any other packages.
. Utility code has been added which will position an input file at the
first record of a specified namelist group. This simplifies reading a
specific namelist from an input file that contains multiple namelists.
Bugs fixed (include bugzilla ID): #563
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes:
. The size of the history master field list was increased.
. The size of the primary hash table for the master field list was
increased.
Note that the primary reason these data array sizes need to be increased
is that there are 10 or so diagnostic fields that occur for each
constituent, and they are registered in the master list whether the
diagnostic is being computed or not. Better control logic needs to be
added to the diagnostic code so that these fields are not registered in
the master list unless the diagnostic is going to be computed and written
to a history file.
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do:
models/atm/cam/src/utils/namelist_utils.F90
. add utility code to search for a specified namelist group name in an
input file and leave the file positioned at the record containing that
group name
List all existing files that have been modified, and describe the changes:
models/atm/cam/src/control/cam_history.F90
. increase pflds to 10000
. increase tbl_hash_pri_sz_lg2 to 16.
models/atm/cam/src/control/runtime_opts.F90
. use subroutine find_group_name from new namelist_utils module to find the
cam_inparm namelist group in the input file.
models/atm/cam/src/dynamics/fv/dyn_comp.F90
. comment out addfld calls for per constituent diagnostics of
hadvnam, vadvnam, tendnam, fixcnam. There are no matching outfld calls
for these fields. (The spectral cores do have outfld calls for these
fields.)
models/atm/cam/src/physics/cam/cam_diagnostics.F90
. if diag_cnst_conv_tend='none' or 'q_only' don't add the per constituent
names for DCxxxx to the master field list
models/atm/cam/src/physics/cam/constituents.F90
. add cnst_num_avail function to return the number of available slots in
the constituent array.
models/atm/cam/src/physics/cam/tracers_suite.F90
. modify to enable up to 999 copies of each of the 5 basic tracer types.
That's a total of up to 5000 tracers.
models/atm/cam/src/physics/cam/tracers.F90
. modify tracers_register to query the constituent module for the number of
available slots in the constituent array, and to register test tracers in
all available slots. This assumes that the test tracers are registering
last.
models/atm/cam/src/chemistry/trop_mozart/tracer_data.F90
. bug fix for cyclic mode.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: All PASS except
045 br532 TBR.sh s32sdh ghgrmp 4+2s ...............................FAIL! rc= 11
tempest: All PASS except
034 br531 TBR.sh s32dh co2rmp 4+2s ................................FAIL! rc= 11
bangkok/lf95: All PASS except
034 br511 TBR.sh s8c8mdm ttrac 4+2s ...............................FAIL! rc= 11
These are expected failures.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: none.
===============================================================
===============================================================
Tag name: cam3_5_06
Originator(s): mvertens
Date: Mon Aug 6 10:07:59 MDT 2007
One-line Summary: cam and surface components may now couple every cam time step
Purpose of changes: There is now the capability to have cam and the surface components
couple every cam time step. This capability can be invoked by simply setting the
namelist variable, atm_cpl_dt in drv_in equal to the cam timestep (by default it is
set to 3600 seconds). The default behavior of the system is to still couple on the
cam shortwave radiation timestep. All tests have maintained this behavior.
Bugs fixed (include bugzilla ID): 493
Describe any changes made to build system: None
Describe any changes made to the namelist: None
List any changes to the defaults for the boundary datasets: None
Describe any substantial timing or memory changes: None expected
Code reviewed by: myself
List all subroutines eliminated: None
List all subroutines added and what they do: None
List all existing files that have been modified, and describe the changes:
M SVN_EXTERNAL_DIRECTORIES
- updated models/lnd/clm code to pftintdat08_clm3_5_05 (modified lnd_comp_mct.F90 to permit
either hourly coupling or coupling on the cam time step)
- update drv/seq_mct code to drvseq1_0_42 (seq_flux_mct.F90 only does
albedo calculation for ocn or ice if nextsw_cday is not -1
M models/atm/cam/src/control/atm_comp_mct.F90
- put in capability to couple either on the short wave radiative time step or
every cam time step
M atm/cam/src/control/ccsm_msg.F90
- one line fix for bugzilla #493
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista:
045 br532 TBR.sh s32sdh ghgrmp 4+2s ...............................FAIL! rc= 11
tempest:
034 br531 TBR.sh s32dh co2rmp 4+2s ................................FAIL! rc= 11
bangkok/lf95:
034 br511 TBR.sh s8c8mdm ttrac 4+2s ...............................FAIL! rc= 11
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: bit-for-bit with cam3_5_05.
In addition, bfb restults with cam3_4_15 can be shown with coupling
every cam timestep, if the cam3_4_15 ice_comp.F90 code is modified
such that the zenith angle is updated on the timestep before cam needs
the albedos (as is done in clm).
===============================================================
===============================================================
Tag name: cam3_5_05
Originator(s): eaton
Date: Tue Jul 24 09:17:11 MDT 2007
One-line Summary: add use case for 1990 control. bug fixes.
Purpose of changes:
. add use case for 1990 control runs to build namelist
Bugs fixed (include bugzilla ID):
. fix bug in configure's -test option
. fix bug in setting nlend
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do:
models/atm/cam/bld/use_cases/1990_control.nl
. ghg values, ozone, and aerosol datasets for 1990.
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/build-namelist
. add 1990_control as use case
models/atm/cam/bld/configure
. fix bug in configure's -test option
models/atm/cam/src/control/atm_comp_mct.F90
. fix bug in setting nlend
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: all PASS except
045 br532 TBR.sh s32sdh ghgrmp 4+2s ...............................FAIL! rc= 11
tempest: all PASS except
034 br531 TBR.sh s32dh co2rmp 4+2s ................................FAIL! rc= 11
bangkok/lf95: all PASS except
034 br511 TBR.sh s8c8mdm ttrac 4+2s ...............................FAIL! rc= 11
These branch test failures are pre-existing.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: bit-for-bit
===============================================================
===============================================================
Tag name: cam3_5_04
Originator(s): mvr,rneale,jrichter
Date: 18 Jul 2007
One-line Summary: Addition of convective momentum transport fixes
plus tuning parameter updates
Purpose of changes: a bug was discovered in the cmt code and
energy balancing was required for the spinup of biogeochemistry
3.5 coupled runs
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: rneale,jrichter
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
M cloud_fraction.F90
- tuning of cloud parameters
M zm_conv_intr.F90
M zm_conv.F90
- cmt bug fixes
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
blueice:
bl132 TBL.sh e32sdh ghgrmp 4s .................................FAIL! rc= 7
bl133 TBL.sh e32pdh aqpgro 2s .................................FAIL! rc= 7
bl151 TBL.sh e64h outfrq24h 2d ................................FAIL! rc= 7
bl332 TBL.sh f4sdh ghgrmp 4s ..................................FAIL! rc= 7
bl333 TBL.sh f4pdh aqpgro 2s ..................................FAIL! rc= 7
bl336 TBL.sh wm4h outfrq2h 4s .................................FAIL! rc= 7
bl337 TBL.sh f4dh fv2d_8tsk 4s ................................FAIL! rc= 7
bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7
bl354 TBL.sh fm2dh outfrq2h 4s ................................FAIL! rc= 7
bl355 TBL.sh fmo2dh off2x2.5 4s ...............................FAIL! rc= 7
bl381 TBL.sh fma1.9h outfrq2h+1870_prog_aero 4s ...............FAIL! rc= 7
bl382 TBL.sh fmgpa1.9h outfrq2h+1870_control 4s ...............FAIL! rc= 7
br532 TBR.sh s32sdh ghgrmp 4+2s ...............................FAIL! rc= 11
bl532 TBL.sh s32sdh ghgrmp 4s .................................FAIL! rc= 7
bl533 TBL.sh s32pdh aqpgro 2s .................................FAIL! rc= 7
bl551 TBL.sh s64h outfrq24h 2d ................................FAIL! rc= 7
tempest:
bl131 TBL.sh e32dh co2rmp 4s ..................................FAIL! rc= 7
bl132 TBL.sh e32sdh ghgrmp 4s .................................FAIL! rc= 7
bl133 TBL.sh e32pdh aqpgro 2s .................................FAIL! rc= 7
bl152 TBL.sh e64o outfrq24h 2d ................................FAIL! rc= 7
bl314 TBL.sh wg10dh outfrq2h 4s ...............................FAIL! rc= 7
bl331 TBL.sh f4gdh co2rmp 4s ..................................FAIL! rc= 7
bl332 TBL.sh f4sdh ghgrmp 4s ..................................FAIL! rc= 7
bl333 TBL.sh f4pdh aqpgro 2s ..................................FAIL! rc= 7
bl376 TBL.sh f1.9o outfrq24h 2d ...............................FAIL! rc= 7
br531 TBR.sh s32dh co2rmp 4+2s ................................FAIL! rc= 11
bl531 TBL.sh s32dh co2rmp 4s ..................................FAIL! rc= 7
bl532 TBL.sh s32sdh ghgrmp 4s .................................FAIL! rc= 7
bl533 TBL.sh s32pdh aqpgro 2s .................................FAIL! rc= 7
bl552 TBL.sh s64o outfrq24h 2d ................................FAIL! rc= 7
bangkok/lf95:
bl111 TBL.sh e8c8mdm ttrac 4s .................................FAIL! rc= 7
bl112 TBL.sh e8sdm ghgrmp 4s ..................................FAIL! rc= 7
bl113 TBL.sh e8pdm aqpgro 2s ..................................FAIL! rc= 7
bl153 TBL.sh e64m outfrq24h 2d ................................FAIL! rc= 7
bl311 TBL.sh f10c8mdm ttrac 4s ................................FAIL! rc= 7
bl312 TBL.sh f10sdm ghgrmp 4s .................................FAIL! rc= 7
bl313 TBL.sh f10pdm aqpgro 2s .................................FAIL! rc= 7
bl317 TBL.sh f10dm fv2d_4tsk 4s ...............................FAIL! rc= 7
bl377 TBL.sh f1.9m outfrq24h 2d ...............................FAIL! rc= 7
br511 TBR.sh s8c8mdm ttrac 4+2s ...............................FAIL! rc= 11
bl511 TBL.sh s8c8mdm ttrac 4s .................................FAIL! rc= 7
bl512 TBL.sh s8sdm ghgrmp 4s ..................................FAIL! rc= 7
bl513 TBL.sh s8pdm aqpgro 2s ..................................FAIL! rc= 7
bl553 TBL.sh s64m outfrq24h 2d ................................FAIL! rc= 7
-all baseline tests (other than ideal physics or adiabatic tests) were
expected to fail...the branch test with sld has been failing for reasons
unrelated to this commit
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change : new climate!
If bitwise differences were observed, how did you show they were no worse
than roundoff?
- cmt mods were validated in a 10yr stand-alone cam run using cam3_3_3
- tuning mods are being validated in a 18mo F run, using ccsm3_5_beta06
plus the mods for this commit
- climate to be verified in an upcoming 50yr ccsm coupled run
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_5_03
Originator(s): eaton, fvitt, jwolfe
Date: Thu Jul 5 18:58:40 MDT 2007
One-line Summary: Vertical interp of prescribed constituents modified; PGI
compiler workarounds; tuning mod; new dust optics.
Purpose of changes:
. Use hybrid coefficients and surface pressure if available to
construct a pressure field to vertically interpolate the prescribed
constituents. Otherwise use the variable "lev" to vertically
interpolate the prescribed constituents. (fvitt)
. Mod to cam_diagnostics so that diagnostics needed by the AMWG package are
always on the default history file.
. Mod to check_energy so that when printing of detailed column by column
conservation errors is on (it's off by default), it's on for all
processes and not just the masterproc.
. Tuning mod: reduce rhminh from .80 to .79
. Add default 30 level initial files for FV, 1.9x2.5, for running w/ UW
PBL scheme.
. Change CLM to branch_tags/pftintdat06_clm3_5_05/. This provides a fix
for effective porosity having zero values, which apparently was not
affecting the simulation, but was preventing us from running some tests
in debug mode.
. PGI compiler (6.2-5) was failing to recognize aliases made in use statements.
Workaround by using CAM's physconst module in place of share_const_mod. (jwolfe)
Also modify code to avoid creating zero length substrings which are
causing problems for the pgi compiler.
. Update the aerosol optics file. The new file is backwards compatible
with the previous one when using the old prescribed aerosol distributions
(which is still the default). It contains new optics for the prognostic
dust (and the new prescribed dust) due to differences in how the size
distribution is binned.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
. SVN_EXTERNAL_DIRECTORIES
- change CLM to branch_tags/pftintdat04_clm3_5_04
. models/atm/cam/src/chemistry/trop_mozart/tracer_cnst.F90
- remove '_trcnst' suffix from fields output by this module
. models/atm/cam/src/chemistry/trop_mozart/tracer_data.F90
- mods to use PS in vertical interpolation if present
. models/atm/cam/src/control/wrap_nf.F90
- add optional arg to wrap_inq_varid to allow return of control to caller
rather than just aborting if a variable isn't in the netcdf file
- add wrap_get_scalar_realx
. models/atm/cam/src/physics/cam/aer_optics.F90
- replace the netcdf variable names *_dust by *_dust_pres
- replace the netcdf variable names *_dust_mod by *_dust_prog
- The get variable id call for abs_cff_mss_aer needs the same logic
as the shortwave optics.
Use abs_cff_mss_aer_prog for the prognostic and new prescribed dust, and
abs_cff_mss_aer for the old prescribed dust.
. models/atm/cam/src/physics/cam/cam_diagnostics.F90
- Mod to cam_diagnostics so that diagnostics needed by the AMWG package are
always on the default history file.
. models/atm/cam/src/physics/cam/check_energy.F90
- remove "if (masterproc)" from prints in check_energy_chng
. models/atm/cam/src/chemistry/trop_mozart/mz_aerosols_intr.F90
. models/atm/cam/src/chemistry/trop_mozart/prescribed_ozone.F90
. models/atm/cam/src/chemistry/trop_mozart/tracer_cnst.F90
- use physconst module instead of shr_const_mod. This is a workaround for
a PGI compiler problem encountered on jaguar.
. models/atm/cam/src/chemistry/trop_mozart/tracer_data.F90
- Mod to avoid creating zero length substring. Using the zero length
substring in a fortran intrinsic function resulted in an allocation
error using pgi6.2-5.
. models/atm/cam/src/physics/cam1/cloud_fraction.F90
- change rhminh from .80 to .79
. models/atm/cam/bld/DefaultCAM_INPARM_Namelist.xml
- add initial files for FV, 1.9x2.5, 30 levels, Jan 1 and Sep 1 start dates.
- update aeroptics file to use AerosolOptics_c060817nat.nc
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: All PASS except:
028 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7
031 bl354 TBL.sh fm2dh outfrq2h 4s ................................FAIL! rc= 7
034 bl355 TBL.sh fmo2dh off2x2.5 4s ...............................FAIL! rc= 7
038 bl381 TBL.sh fma1.9h outfrq2h+1870_prog_aero 4s ...............FAIL! rc= 7
042 bl382 TBL.sh fmgpa1.9h outfrq2h+1870_control 4s ...............FAIL! rc= 7
045 br532 TBR.sh s32sdh ghgrmp 4+2s ...............................FAIL! rc= 11
tempest: All PASS except:
031 bl376 TBL.sh f1.9o outfrq24h 2d ...............................FAIL! rc= 7
034 br531 TBR.sh s32dh co2rmp 4+2s ................................FAIL! rc= 11
bangkok/lf95: All PASS except:
031 bl377 TBL.sh f1.9m outfrq24h 2d ...............................FAIL! rc= 7
034 br511 TBR.sh s8c8mdm ttrac 4+2s ...............................FAIL! rc= 11
. The branch tests br511, br531, br532 also failed in the previous tag.
. The baseline tests bl381 and bl382 fail due to
- the vertical interpolation change for prescribed constituents
- new CAM tuning for FV 1.9x2.5 and 2x2.5
- new dust optics
. The baseline tests bl375, bl376, bl377 failed due to
- new CAM tuning for FV 1.9x2.5 and 2x2.5
. The baseline tests bl354 and bl355 failed due to
- new CAM tuning for FV 1.9x2.5 and 2x2.5
- new dust optics
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB except for FV 1.9x2.5 and 2x2.5 due
to CAM tuning, and all trop_mozart_* runs due to new dust optics.
===============================================================
===============================================================
Tag name: cam3_5_02
Originator(s): bundy
Date: Fri Jun 29 08:30:08 MDT 2007
One-line Summary: Modifications to UW PBL scheme (non-default option)
Purpose of changes: Update the version of the UW PBL scheme in CAM to
the latest from the UW development group.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
M models/atm/cam/src/physics/cam/mcshallow.F90
M models/atm/cam/src/physics/cam/convect_shallow.F90
M models/atm/cam/src/physics/cam/cloud_fraction.F90
M models/atm/cam/src/physics/cam/eddy_diff.F90
M models/atm/cam/src/physics/cam/vertical_diffusion.F90
All changes affect only the CAM UW scheme (shallow convection and eddy
diffusivity).
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: 045 br532 TBR.sh s32sdh ghgrmp 4+2s
...............................FAIL! rc= 11
tempest: 034 br531 TBR.sh s32dh co2rmp 4+2s
................................FAIL! rc= 11
bangkok/lf95: 034 br511 TBR.sh s8c8mdm ttrac 4+2s
...............................FAIL! rc= 11
These tests are expected to fail although the branch capability that
is being tested is apparently okay.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers:
None to default CAM
New climate for UW PBL configuration
===============================================================
Tag name: cam3_5_01
Originator(s): mvertens
Date: Thu Jun 28 09:50:19 MDT 2007
One-line Summary: CAM now couples to surface components only on shortwave timestep
Purpose of changes:
* The following changes force CAM to couple to the surface components on
its shortwave radiation timestep only (currently this is hourly). This
is the same coupling constraint that is currently used in ccsm
mode. CAM now averages its output precipitation fluxes over the hour
period and receives hourly averaged fluxes from the surface
components. All states exchanged between CAM and the surface
componenents are instantaneous. CAM also sends to all the surface
components the calendar to use for the next zenith angle contribution
to the albedo computation.
* The CAM-csim and CAM-dom/som components now have an hourly timestep
and only calculate the albedos hourly. Previously, the CAM-csim and
CAM-dom/som components updated the zenith angle every CAM timestep,
which was incorrect, since CAM was updating its zenith angle only
hourly (i.e. on the shortwave radiation computation).
* As a result of these changes, the CCSM CICE and DOCN7 componenents can
now couple to CAM via the sequential driver (either on the CAM grid or
on the displaced pole grid). CICE and DOCN7 will now be checked out
with the CAM code and a new script in the bld directory,
run-pc-cice-docn7.csh, accompanies this tag. This script permits CAM
to couple to CICE/DOCN7 at T31 (thermodynamic ice mode only) or at
T31_gx3v5. Furthermore, the DOCN7 can be run either in prescribed SST
mode or som mode. This script is a starting point for bringing the
CCSM ice and ocn surface components into stand-alone CAM.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system:
* Changed the specification of coupling fields so that only one file is
needed now, seq_flds_mod.F90. Optional coupling fields are now
specified via the use of CPP directives in this file (i.e. GENSOM,
DUST, PROGSSLT). This removes the needed for a different file and
therefore a different filepath for every new coupling scenario. Changed
configure to reflect this modification.
* Changed configure to permit the addition of CICE and DOCN7 as alternative
ice and ocean components. The following new options can be provided to configure
-ocn Build CAM with data SSTs or slab ocean model [dom | som | docn7]
-ice Build CAM with sea ice model [csim4 | cice]
* If cice is selected for the -ice option then the following additional settings can be added
-cice_nbx Number of cice blocks in x on 1 processor (default 4)
-cice_nby Number of cice blocks in y on 1 processor (default 4)
-cice_nlon Number of ocean/ice longitudes (only used for cice/docn7)
-cice_nlat Number of ocean/ice latitudes (only used for cice/docn7)
-cice_ntasks Number of MPI tasks used to set cice block decomposition (default not set)
Describe any changes made to the namelist:
* The atm_cpl_dt is now set to 3600 by default in DefaultTIMEMGR_INPARM_Namelist.xml
* In additin, in SeqCCSM_namelist.pm, the coupling frequency of all
components is now set to atm coupling frequency. In the future the ocn
component (i.e. POP2) will be allowed to couple at a different
frequency.
List any changes to the defaults for the boundary datasets: None
Describe any substantial timing or memory changes: None expected (not tested yet)
Code reviewed by: myself, Brian Eaton, Erik Kluzek
List all subroutines eliminated:
(all removed files are in the testing system and described below)
List all subroutines added and what they do:
(all added files are in the testing system and described below)
List all existing files that have been modified, and describe the changes:
M SVN_EXTERNAL_DIRECTORIES
- updated to the following externals
models/drv/seq_mct https://svn-ccsm-models.cgd.ucar.edu/drv/seq_mct/trunk_tags/drvseq1_0_37
models/lnd/clm2/src https://svn-ccsm-models.cgd.ucar.edu/clm2/trunk_tags/clm3_5_04/src
models/ice/cice/src https://svn-ccsm-models.cgd.ucar.edu/cice/trunk_tags/cice4_0_20070605/src
models/ocn/docn7 https://svn-ccsm-models.cgd.ucar.edu/docn7/trunk_tags/docn7_070605
models/csm_share https://svn-ccsm-models.cgd.ucar.edu/csm_share/trunk_tags/share3_070626
models/utils/esmf_wrf_timemgr https://svn-ccsm-models.cgd.ucar.edu/esmf_wrf_timemgr/trunk_tags/esmf_wrf_timemgr_060616
models/utils/timing https://svn-ccsm-models.cgd.ucar.edu/timing/trunk_tags/timing_070525
models/utils/mct https://svn-ccsm-models.cgd.ucar.edu/mct/trunk_tags/MCT2_3_0_070524
- note that lnd_comp_mct.F90 now performs time averages over the coupling interval for its output fluxes)
- note that the cice and docn7 timesteps are hourly)
- note that seq_flds_mod.F90 i drvseq1_0_37 has a different specification for optional fields, see above)
- note that driver sync clock is advanced at the beginning of the time step now
M models/ocn/dom/ocn_filenames.F90
- removed previous interface and put in new interface for
specifying restart filename from input year, month, day, secs
M models/ocn/dom/ocn_time_manager.F90
- time step is no longer advanced on restart
M models/ocn/dom/ocn_comp.F90
- sync clock dtime is now the atm coupling interval (1 hour) and not
the atm time step
- no longer read FLAND as option (not needed)
- restart filename determined from sync clock time and no longer from rsfilename_spec_ocn
M models/ocn/dom/ocn_comp_mct.F90
- since driver sync clock is advanced at the beginning of the time step now, check of internal
clock with sync clock is now made at the end of ocn_run_mct
M models/ocn/som/ocn_filenames.F90
- removed previous interface and put in new interface for
specifying restart filename from input year, month, day, secs
M models/ocn/som/ocn_time_manager.F90
- time step is no longer advanced on restart
M models/ocn/som/ocn_comp.F90
- sync clock dtime is now the atm coupling interval (1 hour) and not
the atm time step
- no longer read FLAND as option (not needed)
- restart filename determined from sync clock time and no longer from rsfilename_spec_ocn
M models/ocn/som/ocn_comp_mct.F90
- since driver sync clock is advanced at the beginning of the time step now, check of internal
clock with sync clock is now made at the end of ocn_run_mct
M models/ice/csim4/ice_filenames.F90
- removed previous interface and put in new interface for
specifying restart filename from input year, month, day, secs
M models/ice/csim4/ice_types.F90
- added fswabs to ice_out derived type
M models/ice/csim4/ice_srf.F90
- added fsns update for dom mode (currently only done for som mode)
M models/ice/csim4/ice_comp.F90
- sync clock dtime is now the atm coupling interval (1 hour) and not
the atm time step
- restart filename determined from sync clock time and no longer from rsfilename_spec_ice
M models/ice/csim4/albice.F90
- removed zenith angle calculation from code (now done independently in driver as is the case
in cpl6
- note again that ice zenith angle and albedo now updated only on the atm/surface model coupling interval,
(i.e. cam sw radiation) and not every cam time step
M models/ice/csim4/ice_comp_mct.F90
- the merge to ocean now multiplies melth by aice (to be able to couple in cice) - so now divide melth by
aice in the ice_import_mct routine since the merge to ocn will now multiply it back in
- since driver sync clock is advanced at the beginning of the time step now, check of internal
clock with sync clock is now made at the end of ice_run_mct
M models/atm/cam/src/control/atm_comp_mct.F90
- removed all references to infobuf (shr_intputInfo will now be used for this information)
- CAM now returns control to the driverafter it completes its shortwave radiation computation (i.e. hourly).
However, since the sw computation is done one timestep into the hour (except for SLD), the CAM time
will generally be shifted relative to the sync clock time step by one CAM timestep. As a result,
if iradsw is not 1, then an offset of -dtime is used to check consistency with sync clock at the
end of atm_run_mct routine.
- CAM now averages precipation between coupling intervals and sends averaged precipitation back to the
surface components (as is done in cpl6).
- CAM sends calendar day of next radation computation back to the surface components, and albedos will be
computed for the surface components utilizing this calendar day.
- restart filenames are now set by the sync clock date and not by the CAM internal date (see below)
M models/atm/cam/src/control/filenames.F90
M models/atm/cam/src/control/cam_restart.F90
M models/atm/cam/src/control/cam_comp.F90
M models/atm/cam/src/control/cam_history.F90
M models/atm/cam/src/physics/cam/restart_physics.F90
- for above five routines, now pass in
[yr_spec, mon_spec, day_spec, sec_spec]
as optional arguments to set the name of the restart file
(this will enable restart file name to trigger off of the sync clock time
rather than the internal cam time, which for non-sld mode will be shifted by
a time step ahead of the sync clock time)
M models/atm/cam/src/utils/time_manager.F90
- removed dtime_in as argument (note that for cam_comp.F90, dtime_cam is no longer
passed as an argument to eshr_timemgr_clockGet, but is now set to get_step_size()
after the call to timemgr_init
M models/atm/cam/bld/configure
A + models/atm/cam/bld/run-pc-cice-docn7.csh
M models/atm/cam/bld/DefaultTIMEMGR_INPARM_Namelist.xml
M models/atm/cam/bld/timemgr_inparm.pm
M models/atm/cam/bld/SeqCCSM_namelist.pm
M models/atm/cam/bld/config_definition.xml
- see above description of bld changes
M models/atm/cam/test/system/TSB.ccsm.sh
D models/atm/cam/test/system/config_files/fmo2m
D models/atm/cam/test/system/config_files/f2c11m
D models/atm/cam/test/system/config_files/e128m
A + models/atm/cam/test/system/config_files/f4gdh
D models/atm/cam/test/system/config_files/f4c11gdh
D models/atm/cam/test/system/config_files/e128c11m
A + models/atm/cam/test/system/config_files/fma1.9h
D models/atm/cam/test/system/config_files/fm1.9m
D models/atm/cam/test/system/config_files/f1.9pm
D models/atm/cam/test/system/config_files/e32c11dh
D models/atm/cam/test/system/config_files/s32c11dh
A + models/atm/cam/test/system/config_files/fmgpa1.9h
D models/atm/cam/test/system/config_files/e128pm
M models/atm/cam/test/system/TCT.ccsm.sh
D models/atm/cam/test/system/tests_posttag_phoenix
M models/atm/cam/test/system/TBR.sh
D models/atm/cam/test/system/tests_posttag_bluesky
D models/atm/cam/test/system/tests_pretag_bluesky
M models/atm/cam/test/system/TER.sh
M models/atm/cam/test/system/test_driver.sh
D models/atm/cam/test/system/posttag_cron_bluesky.sh
M models/atm/cam/test/system/tests_pretag_bluevista
M models/atm/cam/test/system/nl_files/ghgrmp
D models/atm/cam/test/system/nl_files/outfrq3s
M models/atm/cam/test/system/nl_files/aqpgro
M models/atm/cam/test/system/nl_files/adia
M models/atm/cam/test/system/nl_files/idphys
A + models/atm/cam/test/system/nl_files/co2rmp
M models/atm/cam/test/system/nl_files/fv2d_4tsk
M models/atm/cam/test/system/nl_files/scm_prep
D models/atm/cam/test/system/nl_files/pghgsul
M models/atm/cam/test/system/nl_files/off2x2.5
M models/atm/cam/test/system/nl_files/no_ttrac
M models/atm/cam/test/system/nl_files/fv2d_8tsk
A + models/atm/cam/test/system/nl_files/outfrq2h
M models/atm/cam/test/system/nl_files/ttrac_lb1
M models/atm/cam/test/system/nl_files/ttrac_lb2
M models/atm/cam/test/system/nl_files/ttrac
M models/atm/cam/test/system/nl_files/ttrac_lb3
M models/atm/cam/test/system/tests_posttag_blueice
M models/atm/cam/test/system/input_tests_master
D models/atm/cam/test/system/tests_posttag_robin
M models/atm/cam/test/system/CAM_runcmnd.sh
M models/atm/cam/test/system/TSM.sh
- the above testing changes needed to account for the fact that atm->surface
component coupling only occurs on the cam shortwave radiation time
step and that the GENSOM and DUST fields transferred no longer involve a
change to the filepath but rather require the specification of a CPP variable
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista:
045 br532 TBR.sh s32sdh ghgrmp 4+2s ...............................FAIL! rc= 11
(note that the branch actually was bfb, but a history restart file was written on
the initial part of this run, giving rise to a fail status for this test)
tempest:
034 br531 TBR.sh s32dh co2rmp 4+2s ................................FAIL! rc= 11
(note that the branch actually was bfb, but a history restart file was written on
the initial part of this run, giving rise to a fail status for this test)
calgary/lf95:
034 br511 TBR.sh s8c8mdm ttrac 4+2s ...............................FAIL! rc= 11
(note that the branch actually was bfb, but a history restart file was written on
the initial part of this run, giving rise to a fail status for this test)
Summarize any changes to answers:
- what code configurations:
NOTE ***all SLD configurations will be bfb with cam3_4_15***
NOTE ***all non-SLD runs are larger than roundoff but same climate (see below) ***
- what platforms/compilers: All
- nature of change: larger than roundoff but same climate
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced:
- source tag (all code used must be in the repository):
hourly coupling: https://svn-ccsm-models.cgd.ucar.edu/cam1/branch_tags/seqflxave_cam3_4_01_tags/seqflxave10_cam3_4_11
control: https : https://svn-ccsm-models.cgd.ucar.edu/cam1/branch_tags/conv_cam3_4_03_tags/conv09_cam3_4_03
- platform/compilers:
ibm-sp( blueice )
- configure commandline:
configure -spmd -smp -dyn fv -res 1.9x2.5 -ocn dom (for hourly_cpl-f1.9m_conv09 below)
configure -spmd -smp -dyn fv -res 1.9x2.5 -ocn som (for hourly_cpl_som-cam_som below)
- build-namelist command (or complete namelist):
for dom:
$cfgdir/build-namelist -s -case $case -runtype $runtype -config $blddir/config_cache.xml \
-namelist "&camexp mss_irt=0 stop_option='nmonths', stop_n=$stop_n doisccp=.true. fincl1='CLDST' \
ncdata='/fis/cgd/cms/rneale/cam_mydata/atm/cam3_2_59_bvc.cam2.i.0023-01-01-00000.nc' start_ymd=101 \
restart_pfile='$rundir/rpointer.drv' rest_pfile='$rundir/rpointer.cam' / \
&dom_inparm rest_pfile='$rundir/rpointer.dom' / \
&som_inparm rest_pfile='$rundir/rpointer.som' / \
&csim_inparm rest_pfile='$rundir/rpointer.csim' / \
&clmexp rpntpath='$rundir/rpointer.clm' \
finidat='/fs/cgd/csm/inputdata/lnd/clm2/inidata_3.1/ccsm/clmi_0000-01-01_1.9x2.5_gx1v4_c070311.nc' /" \
|| echo "build-namelist failed" && exit 1
for som:
$cfgdir/build-namelist -s -case $case -runtype $runtype -config $blddir/config_cache.xml \
-namelist "&camexp mss_irt=0 stop_option='nmonths', stop_n=$stop_n doisccp=.true. fincl1='CLDST' \
ncdata='/fis/cgd/cms/rneale/cam_mydata/atm/cam3_2_59_bvc.cam2.i.0023-01-01-00000.nc' start_ymd=101 \
bndtvs='/fis/cgd/cseg/people/mvr/cam_mydata/som/cam3_4_11_with_correct_pft_0000-0020_qflux.nc' \
restart_pfile='$rundir/rpointer.drv' rest_pfile='$rundir/rpointer.cam' / \
&dom_inparm rest_pfile='$rundir/rpointer.dom' / \
&som_inparm rest_pfile='$rundir/rpointer.som' / \
&csim_inparm rest_pfile='$rundir/rpointer.csim' / \
&clmexp rpntpath='$rundir/rpointer.clm' \
finidat='/fs/cgd/csm/inputdata/lnd/clm2/inidata_3.1/ccsm/clmi_0000-01-01_1.9x2.5_gx1v4_c070311.nc' /" \
|| echo "build-namelist failed" && exit 1
MSS location of output:
/MVR/csm/seqflxave10_cam3_4_11/hour_cpl
/MVR/csm/seqflxave10_cam3_4_11/hour_cpl_som
MSS location of control simulations used to validate new climate:
*** to be filled in by mvr ***
URL for AMWG diagnostics output used to validate new climate:
http://www.cgd.ucar.edu/cms/rneale/runs/cam/hourly_cpl-f1.9m_conv09
http://www.cgd.ucar.edu/cms/rneale/runs/cam/hourly_cpl_som-cam_som
===============================================================
Tag name: cam3_4_15
Originator(s): eaton, fvitt
Date: Fri Jun 15 15:30:19 MDT 2007
One-line Summary: mod to trop_mozart boundary dataset reader
Purpose of changes:
. Modify the dataset reader in the trop_mozart tracer_data module so that the
dataset does not need a time coordinate variable. This reader is being used
for oxidant datasets that use an integer date variable instead of a time
coordinate.
. update the use case "1870_prog_aero" to specify that the oxidant dataset
should be treated as cyclic for year 1870.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
. models/atm/cam/src/chemistry/trop_mozart/tracer_data.F90
- mod to relax requirement that datasets have a time coordinate variable
. models/atm/cam/bld/use_cases/1870_prog_aero.nl
- specify the oxidant dataset as cyclic. The earlier version of
tracer_data.F90 didn't support a cyclic dataset option.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: All PASS.
tempest: All PASS.
bangkok/lf95: All PASS.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: NONE
===============================================================
===============================================================
Tag name: cam3_4_14
Originator(s): eaton, fvitt
Date: Fri Jun 8 10:42:45 MDT 2007
One-line Summary: add initial file; trop_mozart prescribed aerosol bug fix
Purpose of changes:
. Add an FV 1.9x2.5 initial file valid for Jan 01 which is the initial date
normally used for CCSM runs.
. Changed the 1870 use cases for build-namelist to start simulations in
year 0 rather than 1870, which is the convention for climatological runs.
Bugs fixed (include bugzilla ID):
. Apply bug fix from Francis to the trop_mozart prescribed aerosol code
(fix the molecular weights of the aerosol species).
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
. models/atm/cam/bld/DefaultCAM_INPARM_Namelist.xml
- add Jan 1 initial file for FV 1.9x2.5
. models/atm/cam/bld/use_cases/1870_control.nl
. models/atm/cam/bld/use_cases/1870_prog_aero.nl
- Removed the setting start_ymd=18700901 from these files which are used to
implement the build-namelist use cases. So by default the simulations
will start in year 0 which is the convention for climatological runs.
. models/atm/cam/src/chemistry/trop_mozart_prescribed_aero/mo_imp_sol.F90
- new preprocessor output -- doesn't affect prescribed aerosols
. models/atm/cam/src/chemistry/trop_mozart_ghg_paero/mo_sim_dat.F90
. models/atm/cam/src/chemistry/trop_mozart_prescribed_aero/mo_sim_dat.F90
- fix molecular weights of aerosol species
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: All PASS.
tempest: All PASS.
bangkok/lf95: All PASS.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: none
===============================================================
===============================================================
Tag name: cam3_4_13
Originator(s): eaton, fvitt
Date: Tue Jun 5 09:26:59 MDT 2007
One-line Summary: updates to trop_mozart; add build-namelist use cases; misc bug fixes
Purpose of changes:
. Updates to trop_mozart code to assume constituents use dry mass mixing
ratios (fvitt).
. Fix the sea salt mass passed to the aerosol optics code by summing masses
from the 4 sea salt bins into the 1 bin currently used by the optics
(fvitt).
. Add new chemistry option which includes a simple GHG chemistry (CH4, N2O,
CFC11, CFC12) in addition to the prescribed aerosols. This is intended
to replace the simple GHG chemistry that was enabled in CAM by setting
the namelist variable trace_gas=.true.. This option is set by invoking
configure with the arg "-chem trop_mozart_ghg_paero". This option
additionally allows reading the prescribed ozone from a 3D dataset which
is spatially interpolated to the model grid. (fvitt)
. Add ability to cycle the prescribed aerosol datasets. This is needed for
the 1870 and present day control runs. (fvitt)
. Add "-use_case" option to build-namelist to provide a single switch for
setting up runs that require setting many namelist parameters. Two use
cases have been implemented:
1870_prog_aero -- sets up run that was used to produce the 1870 aerosol
climatology
1870_control -- sets up the 1870 control run which uses a simple GHG
chemistry, the 1870 prescribed aerosol climatology, an
1870 prescribed ozone climatology, and sets the solar
constant to 1366.5 W/m^2
. Update to mct/trunk_tags/MCT2_3_0_070524. This fixes a memory leak.
. Update to timing/trunk_tags/timing_070525. This fixes a problem with the
timers that caused a seg fault on lightning (w/ pathscale) when a t_stopf
call was made without a matching t_startf.
Bugs fixed (include bugzilla ID):
. defineqflux was broken on 64-bit platforms due to stdlib.h not being
included (it's needed to provide the correct declaration of malloc).
. Fix compilation failure on IBM when compiling for pure OMP.
Describe any changes made to build system: none
Describe any changes made to the namelist:
. New namelist vars for trop_mozart chemistry
character(len=16) :: prescribed_ozone_name ! prescribed ozone field name in netCDF
character(len=256) :: prescribed_ozone_file ! prescribed ozone file
character(len=256) :: prescribed_ozone_filelist ! list of prescribed data files (series of files)
character(len=256) :: prescribed_ozone_datapath ! absolute path of prescribed data files
character(len=8) :: prescribed_ozone_type ! 'CYCLICAL' | 'SERIAL' (default)
logical :: prescribed_ozone_rmfile ! remove data file from local disk (default .false.)
integer :: prescribed_ozone_ymd ! yyyymmdd - start date of the prescribed data (default is current model date)
integer :: prescribed_ozone_tod ! start time of day (seconds) of the prescribed data (default is current model time)
character(len=256) :: tracer_cnst_datapath ! absolute path of prescribed data files
character(len=8) :: tracer_cnst_type ! 'CYCLICAL' | 'SERIAL' (default)
character(len=256) :: tracer_srcs_datapath ! absolute path of prescribed data files
character(len=8) :: tracer_srcs_type ! 'CYCLICAL' | 'SERIAL' (default)
List any changes to the defaults for the boundary datasets:
. Defaults have been changed for emission datasets used by trop_mozart_aero
to match the datasets used to produce the prescribed aerosol distributions
to be used in future CCSM simulations.
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do:
models/atm/cam/bld/config_defaults_trop_mozart_ghg_paero.xml
models/atm/cam/src/chemistry/trop_mozart/mo_ghg_chem.F90
models/atm/cam/src/chemistry/trop_mozart_ghg_paero
models/atm/cam/src/chemistry/trop_mozart_ghg_paero/mo_lu_solve.F90
models/atm/cam/src/chemistry/trop_mozart_ghg_paero/mo_setrxt.F90
models/atm/cam/src/chemistry/trop_mozart_ghg_paero/m_rxt_id.F90
models/atm/cam/src/chemistry/trop_mozart_ghg_paero/mo_phtadj.F90
models/atm/cam/src/chemistry/trop_mozart_ghg_paero/mo_nln_matrix.F90
models/atm/cam/src/chemistry/trop_mozart_ghg_paero/mo_adjrxt.F90
models/atm/cam/src/chemistry/trop_mozart_ghg_paero/m_het_id.F90
models/atm/cam/src/chemistry/trop_mozart_ghg_paero/mo_lu_factor.F90
models/atm/cam/src/chemistry/trop_mozart_ghg_paero/mo_indprd.F90
models/atm/cam/src/chemistry/trop_mozart_ghg_paero/mo_sim_dat.F90
models/atm/cam/src/chemistry/trop_mozart_ghg_paero/m_spc_id.F90
models/atm/cam/src/chemistry/trop_mozart_ghg_paero/mo_imp_sol.F90
models/atm/cam/src/chemistry/trop_mozart_ghg_paero/chem_mods.F90
models/atm/cam/src/chemistry/trop_mozart_ghg_paero/mo_prod_loss.F90
models/atm/cam/src/chemistry/trop_mozart_ghg_paero/mo_lin_matrix.F90
. implement new trop_mozart_ghg_paero chemistry option
models/atm/cam/src/chemistry/trop_mozart/prescribed_ozone.F90
. allows trop_mozart options to read the 3D prescribed ozone datasets and
spatially interpolate them to the model grid.
models/atm/cam/bld/use_cases/1870_prog_aero.nl
. implements the use case 1870_prog_aero
models/atm/cam/bld/use_cases/1870_control.nl
. implements the use case 1870_control
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/build-namelist
. add "-use_case" commandline arg to be used to specify a use case
. add use case for 1870_prog_aero
. add use case for 1870_control
. add "-ignore_ic_year" commandline arg. This is analogous to the
-ignore_ic_date option except that the month/day part of the date are
still matched. This allows a run starting at 18700901 to use a
climatological initial file with a date 901.
models/atm/cam/bld/DefaultCAM_INPARM_Namelist.xml
. update names of some trop_mozart_aero files
. add prescribed aerosol dataset for 1870 climatology
models/atm/cam/bld/cam_inparm.pm
. changed defaults so that prognostice aerosols are radiatively passive
. update trop_mozart_aero
. update trop_mozart_prescribed_aero
. add trop_mozart_ghg_paero
models/atm/cam/bld/config_definition.xml
models/atm/cam/bld/configure
. add new -chem option trop_mozart_ghg_paero
models/atm/cam/bld/SeqCCSM_namelist.pm
. add code to build method to parse the namelist file that's providing the
use case defaults
models/atm/cam/bld/compnl.pm
. modify attribute matching in defaults files to allow only a month/day
match of the ic_ymd attribute
models/atm/cam/src/chemistry/trop_mozart/chemistry.F90
. specify dry mmr when adding constituents
models/atm/cam/src/chemistry/trop_mozart/mz_aerosols_intr.F90
. fix mass calculations to assume dry mixing ratios
. sum the masses of the 4 sea salt constituents to provide the single sea
salt bin requested by current radiative optics package.
models/atm/cam/src/chemistry/trop_mozart/tracer_cnst.F90
. fix mass to volume mixing ratio conversions to assume dry mixing ratios
models/atm/cam/src/chemistry/trop_mozart/tracer_data.F90
. add fix to init of file%filenames
. Replace hardcoded Fortran unit numbers by calls to the shr_file_getUnit
and shr_file_freeUnit routines in the shr_file_mod module.
. add '/' separator between path and fname parts of filepath.
models/atm/cam/src/chemistry/trop_mozart/chemistry.F90
models/atm/cam/src/chemistry/trop_mozart/mo_chemini.F90
models/atm/cam/src/chemistry/trop_mozart/mo_fstrat.F90
models/atm/cam/src/chemistry/trop_mozart/mo_gas_phase_chemdr.F90
models/atm/cam/src/chemistry/trop_mozart/mo_mean_mass.F90
models/atm/cam/src/chemistry/trop_mozart/mo_sulf.F90
models/atm/cam/src/chemistry/trop_mozart/mo_tropopause.F90
models/atm/cam/src/chemistry/trop_mozart/rad_cnst_data_interface.F90
models/atm/cam/src/chemistry/trop_mozart/tracer_cnst.F90
models/atm/cam/src/chemistry/trop_mozart/tracer_data.F90
models/atm/cam/src/chemistry/trop_mozart/tracer_srcs.F90
. mods for trop_mozart_ghg_paero package
models/atm/cam/src/control/runtime_opts.F90
. remove initialization of scon from preset. Move to cam_control_mod.F90.
. add namelist variables for reading prescribed ozone
. add namelist variables to specify whether the prescribed data is serial
or cyclical
models/atm/cam/src/control/cam_control_mod.F90
. add initializer for scon (1367 W/m^2)
models/atm/cam/src/control/ccsm_msg.F90
. Move the t_startf/t_stopf calls to fix a bug that resulted from the calls
to t_adj_detailf(+2) during initialization (this prevented a t_startf
call from being made in initialization whose matching t_stopf call during
the run phase then became unmatched).
models/atm/cam/src/dynamics/fv/dyn_comp.F90
. Move !$omp directive inside the SPMD ifdef at line 1392. Otherwise a
syntax error (on ibm) was encountered when compiling for no SPMD.
models/atm/cam/tools/defineqflux/timediddle_mavg.c
. Include the stdlib.h header file -- needed for correct malloc declaration.
models/drv/seq_mct/seq_flux_mct.F90
. Move calls to mct_gGrid_exportRAttr so they are only called once rather
than every step. This avoids a memory leak discovered in the mct code.
It's also a better to just make these calls once even without a memory
leak.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: All PASS except:
037 bl354 TBL.sh fm2dh outfrq3s 9s ................................FAIL! rc= 7
040 bl355 TBL.sh fmo2dh off2x2.5 9s ...............................FAIL! rc= 7
These baseline tests should fail due to changing the treatment of constituents
by trop_mozart from wet to dry mixing ratio.
tempest: All PASS.
bangkok/lf95: All PASS.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB except for trop_mozart_* chemistry
===============================================================
===============================================================
Tag name: cam3_4_12
Originator(s): Francis Vitt, eaton
Date: Wed May 2 16:02:20 MDT 2007
One-line Summary:
Addition of prescribed aerosols chemistry package with changes
to dust and trop_mozart external forcings.
Purpose of changes:
The new chemistry package and code changes are needed to be able to
read the prescribed aerosol distributions that will be generated by
runs with the prognostic aerosol package which was added at cam3_4_11.
Note that both new aerosol packages are undergoing testing and
should not be used for scientific purposes. CAM by default is still
using the CAM-3.0 prescribed aerosol climatology.
Update CLM to clm3_4_1.
Update the CLM namelist defaults file.
Bugs fixed (include bugzilla ID):
- vertical emissions altitude (external forcing) is now set to
height relative to surface
- trop_mozart interactive dry deposition broadcast days of
climatological soil water data to all MPI tasks
Describe any changes made to build system:
chemistry option added to configure script
-chem trop_mozart_prescribed_aero
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets:
Update the CLM namelist defaults to use the correct PFT file,
the latest fsurdat file for FV 1.9x2.5, and a fully spun up IC file
for FV 1.9x2.5.
Describe any substantial timing or memory changes: not yet tested...
Code reviewed by: Brian Eaton, Francis Vitt
List all subroutines eliminated:
List all subroutines added and what they do:
--- models/atm/cam/bld/config_defaults_trop_mozart_prescribed_aero.xml
needed to configure new trop_mozart_prescribed_aero chemistry package
--- models/atm/cam/src/chemistry/trop_mozart_prescribed_aero/mo_lu_solve.F90
--- models/atm/cam/src/chemistry/trop_mozart_prescribed_aero/mo_setrxt.F90
--- models/atm/cam/src/chemistry/trop_mozart_prescribed_aero/m_rxt_id.F90
--- models/atm/cam/src/chemistry/trop_mozart_prescribed_aero/mo_phtadj.F90
--- models/atm/cam/src/chemistry/trop_mozart_prescribed_aero/mo_nln_matrix.F90
--- models/atm/cam/src/chemistry/trop_mozart_prescribed_aero/mo_adjrxt.F90
--- models/atm/cam/src/chemistry/trop_mozart_prescribed_aero/m_het_id.F90
--- models/atm/cam/src/chemistry/trop_mozart_prescribed_aero/mo_lu_factor.F90
--- models/atm/cam/src/chemistry/trop_mozart_prescribed_aero/mo_indprd.F90
--- models/atm/cam/src/chemistry/trop_mozart_prescribed_aero/mo_sim_dat.F90
--- models/atm/cam/src/chemistry/trop_mozart_prescribed_aero/m_spc_id.F90
--- models/atm/cam/src/chemistry/trop_mozart_prescribed_aero/mo_imp_sol.F90
--- models/atm/cam/src/chemistry/trop_mozart_prescribed_aero/chem_mods.F90
--- models/atm/cam/src/chemistry/trop_mozart_prescribed_aero/mo_prod_loss.F90
--- models/atm/cam/src/chemistry/trop_mozart_prescribed_aero/mo_lin_matrix.F90
code files for the new trop_mozart_prescribed_aero chemistry package
List all existing files that have been modified, and describe the changes:
--- models/atm/cam/bld/configure
--- models/atm/cam/bld/config_definition.xml
added -chem trop_mozart_prescribed_aero configure option
--- models/atm/cam/bld/DefaultCAM_INPARM_Namelist.xml
--- models/atm/cam/bld/cam_inparm.pm
default namelist settings for the trop_mozart_prescribed_aero chemistry
--- models/atm/cam/bld/DefaultCLM_INPARM_Namelist.xml
new PFT file, updated fsurdat file for FV 1.9x2.5, new spun-up IC
file for FV 1.9x2.5
--- models/atm/cam/src/control/runtime_opts.F90
replaced pcnst with MAXTRCS to allow then number of prescribed tracers
to excede the number of prognostic tracers
--- models/atm/cam/src/physics/cam/dust_intr.F90
Natalie's tuned dust module
--- models/atm/cam/src/physics/cam/aerosol_radiation_interface.F90
allow radiative feedbacks from prescribed aerosols
--- models/atm/cam/src/chemistry/trop_mozart/mo_setext.F90
--- models/atm/cam/src/chemistry/trop_mozart/mo_gas_phase_chemdr.F90
invoke external forcing routine (vert emis) with relative height
--- models/atm/cam/src/chemistry/trop_mozart/mo_tropopause.F90
don't do anything if trop_mozart does not solve for any species (gascnt < 1)
--- models/atm/cam/src/chemistry/trop_mozart/mo_sulf.F90
don't do anything if trop_mozart does not have any "user" reactions
--- models/atm/cam/src/chemistry/trop_mozart/mo_chemini.F90
changed some log messages
--- models/atm/cam/src/chemistry/trop_mozart/tracer_cnst.F90
changed some log messages
--- models/atm/cam/src/chemistry/trop_mozart/mo_extfrc.F90
added diagnostic output to history
--- models/atm/cam/src/chemistry/trop_mozart/chemistry.F90
chem_is_active = .false. if imozart < 1
--- models/atm/cam/src/chemistry/trop_mozart/mo_drydep.F90
don't try to open data files if number of dry dep tracers is zero
MPI broadcast days of the climatology soil data
--- models/atm/cam/src/chemistry/trop_mozart_aero/mo_setrxt.F90
change rates of CB1->CB2 and OC1->OC2 transitions
--- models/atm/cam/src/chemistry/trop_mozart_aero/mo_imp_sol.F90
xnox species added to preprossor -- no real change for trop_mozart_aero
--- models/atm/cam/src/chemistry/trop_mozart_aero/chem_mods.F90
preprocessor changes for label rxt rates output -- no real change for trop_mozart_aero
--- SVN_EXTERNAL_DIRECTORIES
new CLM external
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista:
037 bl354 TBL.sh fm2dh outfrq3s 9s ................................FAIL! rc= 7
040 bl355 TBL.sh fmo2dh off2x2.5 9s ...............................FAIL! rc= 7
Changes to the dust caused changes to the chemical species via the
interactive photolysis rates which in turn changed radiation
feedbacks via the sulfur aerosol cycle.
tempest: All Pass
bangkok/lf95: All Pass
CAM tag used for the baseline comparison tests if different than previous
tag: cam3_4_11 with the CLM namelist defaults modified to use the same new
files as this tag so that the baseline tests would pass.
Summarize any changes to answers: BFB except for trop_mozart
===============================================================
===============================================================
Tag name: cam3_4_11
Originator(s): eaton, fvitt
Date: Fri Apr 20 16:40:53 MDT 2007
One-line Summary: mods to deep convection, cloud fraction and aerosols;
update CLM and timing lib externals
Purpose of changes:
. Add dilute plume and convective momentum transport mods from Neale and
Richter to ZM deep convection parameterization.
. Add a call to the cloud fraction routine after pcond (cloud microphysics)
and move the outfld calls for the cloud fraction fields to be after this
new call to cldfrc (Sungsu Park).
. Add the Vavrus "freeze dry" mod.
. Update to latest timing library tag (timing_070328). This fixes the
regression tests that were failing on tempest.
. Update to the latest CLM trunk tag (clm3_expa_98).
. The initial version of a new aerosol package is included, but should not
be used for scientific purposes. It will be updated shortly with mods
required to produce an acceptable aerosol simulation.
. Add debug flags for pathscale to fix a problem with the regression tests
on lightning (fix from Jim Edwards).
. Update the sample run scripts (from Mat Rothstein)
Bugs fixed (include bugzilla ID):
ID 409 - aqua-planet tests in cam3_4_04 fail on tempest
ID 434 - fix fsnsoi gensom problem
Describe any changes made to build system:
. the new aerosol package is enabled by specifying "-chem trop_mozart_aero"
to configure
Describe any changes made to the namelist:
. some emissions datasets names have been added to build-namelist, but the
datasets have not yet been moved to the standard locations.
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do:
models/atm/cam/bld/config_defaults_trop_mozart_aero.xml
. configuration file for new chemistry package
models/atm/cam/src/chemistry/trop_mozart_aero/chem_mods.F90
models/atm/cam/src/chemistry/trop_mozart_aero/m_het_id.F90
models/atm/cam/src/chemistry/trop_mozart_aero/m_rxt_id.F90
models/atm/cam/src/chemistry/trop_mozart_aero/m_spc_id.F90
models/atm/cam/src/chemistry/trop_mozart_aero/mo_adjrxt.F90
models/atm/cam/src/chemistry/trop_mozart_aero/mo_imp_sol.F90
models/atm/cam/src/chemistry/trop_mozart_aero/mo_indprd.F90
models/atm/cam/src/chemistry/trop_mozart_aero/mo_lin_matrix.F90
models/atm/cam/src/chemistry/trop_mozart_aero/mo_lu_factor.F90
models/atm/cam/src/chemistry/trop_mozart_aero/mo_lu_solve.F90
models/atm/cam/src/chemistry/trop_mozart_aero/mo_nln_matrix.F90
models/atm/cam/src/chemistry/trop_mozart_aero/mo_phtadj.F90
models/atm/cam/src/chemistry/trop_mozart_aero/mo_prod_loss.F90
models/atm/cam/src/chemistry/trop_mozart_aero/mo_setrxt.F90
models/atm/cam/src/chemistry/trop_mozart_aero/mo_sim_dat.F90
. source code for new chemistry package
List all existing files that have been modified, and describe the changes:
SVN_EXTERNALS_DIRECTORIES
. update CLM and timing lib externals
models/atm/cam/bld/DefaultCAM_INPARM_Namelist.xml
. new files for trop_mozart_aero package
models/atm/cam/bld/DefaultCLM_INPARM_Namelist.xml
. update fsurdat file for FV 1.9x2.5 only
models/atm/cam/bld/cam_inparm.pm
. add support for trop_mozart_aero namelist vars
models/atm/cam/bld/config_definition.xml
models/atm/cam/bld/configure
. add "trop_mozart_aero" option to -chem arg of configure
models/atm/cam/bld/Makefile.in
. add -trapuv -Wuninitialized to debug flags for pathf90
models/atm/cam/bld/run-ibm.csh
models/atm/cam/bld/run-pc.csh
models/atm/cam/bld/run-sgi.csh
models/atm/cam/bld/run-lightning.csh
models/atm/cam/bld/run-darwin.csh
. updates to the sample run scripts
models/atm/cam/src/chemistry/trop_mozart/mo_airplane.F90
models/atm/cam/src/chemistry/trop_mozart/mo_drydep.F90
models/atm/cam/src/chemistry/trop_mozart/mo_extfrc.F90
models/atm/cam/src/chemistry/trop_mozart/mo_gas_phase_chemdr.F90
models/atm/cam/src/chemistry/trop_mozart/mo_lightning.F90
models/atm/cam/src/chemistry/trop_mozart/mo_setext.F90
models/atm/cam/src/chemistry/trop_mozart/mo_strato_sad.F90
models/atm/cam/src/chemistry/trop_mozart/mo_sulf.F90
models/atm/cam/src/chemistry/trop_mozart/mo_usrrxt.F90
models/atm/cam/src/chemistry/trop_mozart/mz_aerosols_intr.F90
models/atm/cam/src/physics/cam/sulchem.F90
. mods for new chem package
models/atm/cam/src/physics/cam/cloud_fraction.F90
. add Vavrus "freeze dry" mod
models/atm/cam/src/physics/cam/physconst.F90
. add some constants for water (values from shr_const_mod)
models/atm/cam/src/physics/cam/radiation.F90
. remove concld from dummy args of radiation_tend -- use physics buffer
models/atm/cam/src/physics/cam/stratiform.F90
. remove concld from dummy args of stratiform_tend -- use physics buffer
. add call to cldfrc after pcond
. move outfld calls for cloud fraction fields to after cldfrc after pcond
models/atm/cam/src/physics/cam/tphysbc.F90
. remove local var concld as actual arg to stratiform_tend and
radiation_tend calls
models/atm/cam/src/physics/cam/zm_conv.F90
. add mods for convective momentum transport and dilute plumes
models/atm/cam/src/physics/cam/zm_conv_intr.F90
. add calls for convective momentum transport to zm_conv_tend
. add diagnostic output
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: All PASS except:
014 bl151 TBL.sh e64h outfrq24h 2d ................................FAIL! rc= 7
018 bl331 TBL.sh f4c11gdh pghgsul 9s ..............................FAIL! rc= 7
034 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7
056 bl711 TBL.sh h5x8dm adia 9s ...................................FAIL! rc= 7
. The Eulerian and FV baselines fail (bl151, bl375) due to a roundoff diff in
a single diagnostic field (SWCF) introduced at cam3_4_07
. The HOMME baseline fails (bl711) due to answer changes introduced at cam3_4_10.
. The FV baseline that tests the gensom mode (bl331) fails due to the gensom
bugfix.
tempest: All PASS except:
021 bl331 TBL.sh f4c11gdh pghgsul 9s ..............................FAIL! rc= 7
. bl331 fails for the same reason as on bluevista.
. Note that updating to the new timing lib fixed the problems with the
aqua-planet tests.
bangkok/lf95: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag: conv09_cam3_4_03 -- *** NOTE *** this tag was used for baseline
comparisons to validate that the new physics was merged from the
branch to the trunk correctly. All baseline tests that exercise the
standard physics will fail when this code is compared to cam3_4_10.
Summarize any changes to answers, i.e.,
- what code configurations: All that use standard cam physics
- what platforms/compilers: All
- nature of change (roundoff; larger than roundoff but same climate; new
climate): new climate
If bitwise differences were observed, how did you show they were no worse
than roundoff? N.A.
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository): conv09_cam3_4_03
- platform/compilers: CrayXT (jaguar) / pgi-6.2.5 (-fast optimization)
- configure commandline:
configure -fc ftn -dyn fv -res 1.9x2.5 -fopt "-fast" -spmd -nosmp -cppdefs "-DCATAMOUNT"
- build-namelist command (or complete namelist):
build-namelist -s -case f1.9m_conv09 -config $blddir/config_cache.xml \
-namelist "&camexp start_type='startup' stop_option='nmonths' stop_n=1
mss_irt=0 restart_option='monthly' restart_n=1
phys_loadbalance=2 npr_yz=32,8,8,32
rest_pfile='./rpointer.cam' /
&clmexp rpntpath='./rpointer.clm'
finidat='/tmp/work/eaton/inputdata/clmi_0000-09-01_1.9x2.5_gx1v4_c070311.nc' /"
- MSS location of output:
Only climo files used to produce AMWG diagnostics were saved.
mss:/EATON/CASE/conv09_cam3_4_03/f1.9m_conv09/climo_0001-0006.tar
MSS location of control simulations used to validate new climate:
This run was compared against the simulation done with conv07_cam3_4_03
mss:/EATON/CASE/conv07_cam3_4_03/f1.9h_conv07/climo_0001-0006.tar
URL for AMWG diagnostics output used to validate new climate:
http://www.cgd.ucar.edu/cms/rneale/runs/cam/cam3_4_03_conv09-cam3_4_03_conv07/
===============================================================
===============================================================
Tag name: cam3_4_10
Originator(s): Jim Edwards
Date: 4-16-07
One-line Summary: Fix problems commited in cam3_4_09
Purpose of changes: cam3_4_09 had problems which made it unusable
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
models/atm/cam/src/dynamics/fv/inidat.F90
models/atm/cam/src/physics/cam/co2_cycle.F90
models/atm/cam/src/control/ncdio_atm.F90
Fixed formatting and module use errors that managed to get past the
precommit testing for cam3_4_09
models/atm/cam/test/system/CAM_runcmnd.sh
models/atm/cam/test/system/test_driver.sh
fixes to test system on bangkok
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: All passed except:
bl711 TBL.sh h5x8dm adia 9s
HOMME Dycore test expected failure due to restructuring
tempest: all pass except
008 sm133 TSM.sh e32pdh aqpgro 3s .................................FAIL! rc= 8
009 bl133 TBL.sh e32pdh aqpgro 3s .................................FAIL! rc= 4
025 sm333 TSM.sh f4pdh aqpgro 3s ..................................FAIL! rc= 8
026 bl333 TBL.sh f4pdh aqpgro 3s ..................................FAIL! rc= 4
039 sm533 TSM.sh s32pdh aqpgro 3s .................................FAIL! rc= 8
040 bl533 TBL.sh s32pdh aqpgro 3s .................................FAIL! rc= 4
bangkok/lf95: all pass
CAM tag used for the baseline comparison tests if different than previous
tag: cam3_4_08
Summarize any changes to answers, i.e.,
- what code configurations: - what platforms/compilers: - nature
of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_4_09
Originator(s): Jim Edwards
Date: 04-13-07
One-line Summary: Merge of HOMME dycore development branch
Purpose of changes: Update with latest HOMME development
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes:
Code reviewed by: myself, Eaton
List all subroutines eliminated:
D models/atm/cam/src/dynamics/homme/external/math_constants.F90
D models/atm/cam/src/dynamics/homme/external/stats.h
D models/atm/cam/src/dynamics/homme/external/preq_init_mod.F90
D models/atm/cam/src/dynamics/homme/external/utils_mod.F90
D models/atm/cam/src/dynamics/homme/external/locate.F90
D models/atm/cam/src/dynamics/homme/external/torus_mod.F90
D models/atm/cam/src/dynamics/homme/external/stats.F90
D models/atm/cam/src/dynamics/homme/external/field_mod.F90
D models/atm/cam/src/dynamics/homme/external/types_mod.F90
D models/atm/cam/src/dynamics/homme/external/generic_list.F90
D models/atm/cam/src/dynamics/homme/external/ref_state_mod.F90
D models/atm/cam/src/dynamics/homme/external/vertex_mod.F90
D models/atm/cam/src/dynamics/homme/external/rotation_init.F90
D models/atm/cam/src/dynamics/homme/external/vertical_mod.F90
D models/atm/cam/src/dynamics/fv/xpavg.F90
List all subroutines added and what they do:
A models/atm/cam/src/dynamics/homme/external/prim_driver_mod.F90
driver module for the HOMME dycore
A models/atm/cam/src/utils/xpavg_mod.F90
moved from dynamics/fv/xpavg.F90 to facilitate use in ncdio.F90
List all existing files that have been modified, and describe the changes:
U models/atm/cam/test/system/config_files/h5x8dm
U models/atm/cam/bld/DefaultCTL_NL_Namelist.xml
Added -lapack_libdir argument and modified namelist for h5xdm testcase
U models/atm/cam/src/control/ncdio_atm.F90
U models/atm/cam/src/control/startup_initialconds.F90
U models/atm/cam/src/control/cam_comp.F90
U models/atm/cam/src/physics/cam/aer_optics.F90
U models/atm/cam/src/physics/cam/physpkg.F90
U models/atm/cam/src/physics/cam/prescribed_aerosols.F90
U models/atm/cam/src/physics/cam/phys_buffer.F90
U models/atm/cam/src/physics/cam/phys_grid.F90
U models/atm/cam/src/dynamics/sld/inidat.F90
U models/atm/cam/src/dynamics/sld/inital.F90
U models/atm/cam/src/dynamics/eul/inidat.F90
U models/atm/cam/src/dynamics/eul/inital.F90
U models/atm/cam/src/dynamics/fv/inidat.F90
U models/atm/cam/src/dynamics/fv/p_d_adjust.F90
U models/atm/cam/src/dynamics/fv/inital.F90
moved initialization of fields used only in physics from
dynamics/*/inidat.F90 to physpkg.F90 and moved around
initialization to handle this change this allowed us to remove
code duplicated for each dycore and avoid a special case for HOMME
initialization. However a special case was inserted in ncdio.F90
for the fv core initialization to handle the pole average scheme
xpavg.F90, this was done to maintain bfb results with that core.
U models/atm/cam/src/physics/cam/co2_cycle.F90
fixed a bug in constituent initialization
U models/atm/cam/src/dynamics/homme/external/flops_mod.F90
U models/atm/cam/src/dynamics/homme/external/filter_mod.F90
U models/atm/cam/src/dynamics/homme/external/diffusion_mod.F90
U models/atm/cam/src/dynamics/homme/external/mass_matrix_mod.F90
U models/atm/cam/src/dynamics/homme/external/forcing_mod.F90
U models/atm/cam/src/dynamics/homme/external/prim_advance_mod.F90
U models/atm/cam/src/dynamics/homme/external/cg_mod.F90
U models/atm/cam/src/dynamics/homme/external/reduction_mod.F90
U models/atm/cam/src/dynamics/homme/external/schedule_mod.F90
U models/atm/cam/src/dynamics/homme/external/spacecurve_mod.F90
U models/atm/cam/src/dynamics/homme/external/derivative_mod.F90
U models/atm/cam/src/dynamics/homme/external/prim_si_mod.F90
U models/atm/cam/src/dynamics/homme/external/gridgraph_mod.F90
U models/atm/cam/src/dynamics/homme/external/prim_state_mod.F90
U models/atm/cam/src/dynamics/homme/external/metagraph_mod.F90
U models/atm/cam/src/dynamics/homme/external/dimensions_mod.F90
U models/atm/cam/src/dynamics/homme/external/surfaces_mod.F90
U models/atm/cam/src/dynamics/homme/external/quadrature_mod.F90
U models/atm/cam/src/dynamics/homme/external/global_norms_mod.F90
U models/atm/cam/src/dynamics/homme/external/coordinate_systems_mod.F90
U models/atm/cam/src/dynamics/homme/external/cube_mod.F90
U models/atm/cam/src/dynamics/homme/external/control_mod.F90
U models/atm/cam/src/dynamics/homme/external/interpolate_mod.F90
U models/atm/cam/src/dynamics/homme/external/linear_algebra_mod.F90
U models/atm/cam/src/dynamics/homme/external/element_mod.F90
U models/atm/cam/src/dynamics/homme/external/prim_si_ref_mod.F90
U models/atm/cam/src/dynamics/homme/external/time_mod.F90
U models/atm/cam/src/dynamics/homme/external/hybvcoord_mod.F90
U models/atm/cam/src/dynamics/homme/external/baroclinic_inst_mod.F90
U models/atm/cam/src/dynamics/homme/external/checksum_mod.F90
U models/atm/cam/src/dynamics/homme/external/prim_advection_mod.F90
U models/atm/cam/src/dynamics/homme/external/edge_mod.F90
U models/atm/cam/src/dynamics/homme/external/ll_mod.F90
U models/atm/cam/src/dynamics/homme/external/thread_mod.F90
U models/atm/cam/src/dynamics/homme/external/dof_mod.F90
U models/atm/cam/src/dynamics/homme/external/physical_constants.F90
U models/atm/cam/src/dynamics/homme/external/solver_mod.F90
U models/atm/cam/src/dynamics/homme/external/timer.h
U models/atm/cam/src/dynamics/homme/external/namelist_mod.F90
U models/atm/cam/src/dynamics/homme/external/bndry_mod.F90
U models/atm/cam/src/dynamics/homme/external/timer_mod.F90
U models/atm/cam/src/dynamics/homme/external/metis_mod.F90
U models/atm/cam/src/dynamics/homme/external/hybrid_mod.F90
U models/atm/cam/src/dynamics/homme/dyn_grid.F90
U models/atm/cam/src/dynamics/homme/initcom.F90
U models/atm/cam/src/dynamics/homme/dp_coupling.F90
U models/atm/cam/src/dynamics/homme/stepon.F90
U models/atm/cam/src/dynamics/homme/inidat.F90
U models/atm/cam/src/dynamics/homme/dyn_comp.F90
U models/atm/cam/src/dynamics/homme/restart_dynamics.F90
U models/atm/cam/src/dynamics/homme/inital.F90
A cleanup and restructuring of the HOMME dycore interface.
Variables that do not need to be exposed to the model interface
are now hidden. A number of subroutines that are not used by
cam have been removed.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: All passed except:
bl711 TBL.sh h5x8dm adia 9s
HOMME Dycore test expected failure due to restructuring
tempest: All passed except problems carried over from previous commit:
008 sm133 TSM.sh e32pdh aqpgro 3s .................................FAIL! rc= 8
009 bl133 TBL.sh e32pdh aqpgro 3s .................................FAIL! rc= 4
025 sm333 TSM.sh f4pdh aqpgro 3s ..................................FAIL! rc= 8
026 bl333 TBL.sh f4pdh aqpgro 3s ..................................FAIL! rc= 4
039 sm533 TSM.sh s32pdh aqpgro 3s .................................FAIL! rc= 8
040 bl533 TBL.sh s32pdh aqpgro 3s .................................FAIL! rc= 4
bangkok/lf95: All passed
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_4_08
Originator(s): mvr, eaton
Date: 29 Mar 2007
One-line Summary:
svn external definitions simplified; changes to ccsm testing from within cam;
directory name change: physics/cam1 -> physics/cam; ccsm filepath will now
be identical to stand-alone cam; several unrelated bug fixes
Purpose of changes:
attempting to clean up some long standing issues with regards to multiple
external definitions for a single component and directory structure
differences between ccsm and stand-alone cam...this will help with testing
ccsm from within cam
Bugs fixed (include bugzilla ID): 420 & 425
Describe any changes made to build system:
ccsm filepath now identical to stand-alone cam...the external definitions for
ccsm tags using this cam tag (and thereafter) will need to reflect these mods
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: mvr, eaton
List all subroutines eliminated:
D models/atm/cam/src/physics/cam1
- directory and contents moved to ~physics/cam
List all subroutines added and what they do:
A models/atm/cam/src/physics/cam
- directory and contents moved from ~physics/cam1
List all existing files that have been modified, and describe the changes:
M .
M SVN_EXTERNAL_DIRECTORIES
- condensed number of external definitions; now pulling in all csm_share code
M models/ocn/dom/ocn_comp_mct.F90
- bug fix for restart_option="none" (bug# 425)
M models/atm/cam/test/system/TSB.ccsm.sh
- mod to run ccsm tests in regular queue by default rather than premium
M models/atm/cam/test/system/TCT.ccsm.sh
- mods to run ccsm tests either with just the cam code from users sandbox or
with all available sandbox code if env var is set
M models/atm/cam/test/system/test_driver.sh
- will now grab latest ccsm tag in 3_5 beta sequence for testing; increase
job time for posttag testing on lightning
M models/atm/cam/test/system/tests_posttag_bluevista
M models/atm/cam/test/system/tests_pretag_bluevista
M models/atm/cam/test/system/tests_posttag_blueice
- change default test lists to move all ccsm testing to posttag
M models/atm/cam/bld/configure
- mods to support directory name change of physics/cam1 -> physics/cam;
ccsm filepath will now be the same as for stand-alone cam
M models/atm/cam/src/control/cam_history.F90
- bug fix for restarts of fixed length intervals (bug# 420)
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: all PASS
tempest:
008 sm133 TSM.sh e32pdh aqpgro 3s .................................FAIL! rc= 8
009 bl133 TBL.sh e32pdh aqpgro 3s .................................FAIL! rc= 4
025 sm333 TSM.sh f4pdh aqpgro 3s ..................................FAIL! rc= 8
026 bl333 TBL.sh f4pdh aqpgro 3s ..................................FAIL! rc= 4
039 sm533 TSM.sh s32pdh aqpgro 3s .................................FAIL! rc= 8
040 bl533 TBL.sh s32pdh aqpgro 3s .................................FAIL! rc= 4
- carried over from previous commit
bangkok/lf95: all PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_4_07
Originator(s): John Truesdale
Date:
One-line Summary: SCM mode cleanup and additional namelist
functionality
Purpose of changes: replace scm specific netcdf routines with
standard model calls and add surface forcing functionality
to scm cam mode. Also added hooks for relaxation to
obs and column radiation only modes to scm.
Bugs fixed (include bugzilla ID): na
Describe any changes made to build system: na
Describe any changes made to the namelist: Added some new
namelist variables for control of single column mode
of scam. New namelist variables are
scm_iop_srf_prop = .TRUE. if using iop surface properties
scm_relaxation = .TRUE. if relaxing solution back to obs
scm_diurnal_avg = .TRUE. currently just a place holder for
diurnal averaging
scm_crm_mode = .TRUE. run scm in column radiation mode
List any changes to the defaults for the boundary datasets: na
Describe any substantial timing or memory changes: na
Code reviewed by: truesdale
List all subroutines eliminated:
getnetcdfdata.F90 - extracted the setlatlonidx functionality from this
routine and got rid of the rest. Created new
subroutine for setlatlonidx.
List all subroutines added and what they do:
setlatlonidx.F90 - determines closest boundary dataset lat lon to
that specified in the namelist for single column mode.
List all existing files that have been modified, and describe the changes:
The following changes removed scam specific netcdf code and
replaced it with standard cam netcdf calls.
models/atm/cam/src/control/ncdio_atm.F90
models/atm/cam/src/dynamics/eul/getinterpnetcdfdata.F90
models/atm/cam/src/physics/cam1/iop.F90
models/atm/cam/src/physics/cam1/prescribed_aerosols.F90
models/lnd/clm2/src/biogeochem/STATICEcosysDynMod.F90
models/lnd/clm2/src/main/iniTimeConst.F90
models/lnd/clm2/src/main/ncdio.F90
models/lnd/clm2/src/main/surfrdMod.F90
models/ice/csim4/ice_data.F90
models/ocn/dom/sst_data.F90
The following changes added new scm functionality to CAM.
models/atm/cam/src/control/atm_comp_mct.F90
models/atm/cam/src/control/history_scam.F90
models/atm/cam/src/control/runtime_opts.F90
models/atm/cam/src/control/scamMod.F90
models/atm/cam/src/dynamics/eul/dynpkg.F90
models/atm/cam/src/dynamics/eul/forecast.F90
models/atm/cam/src/physics/cam1/physics_types.F90
models/atm/cam/src/physics/cam1/physpkg.F90
models/atm/cam/src/physics/cam1/radiation.F90
models/atm/cam/src/physics/cam1/radlw.F90
models/atm/cam/src/physics/cam1/radsw.F90
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista:
035 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7
059 cs998 TCS.ccsm.sh ERS f19_g13 K ...............................FAIL! rc= 2
Notes: test 059 is a known failure for the last few commits.
test 035 failed on the history tape compare for one value of one
the SWCF field. I reran the test per Eaton's suggestion
compiling with debug on and it passed. Conclusion is
that the models are bit for bit in the prognostics but
because of a compiler optimization are getting a least
significant bit flip in one diagnostic field.
tempest:
008 sm133 TSM.sh e32pdh aqpgro 3s .................................FAIL! rc= 8
009 bl133 TBL.sh e32pdh aqpgro 3s .................................FAIL! rc= 4
025 sm333 TSM.sh f4pdh aqpgro 3s ..................................FAIL! rc= 8
026 bl333 TBL.sh f4pdh aqpgro 3s ..................................FAIL! rc= 4
039 sm533 TSM.sh s32pdh aqpgro 3s .................................FAIL! rc= 8
040 bl533 TBL.sh s32pdh aqpgro 3s .................................FAIL! rc= 4
These aquaplanet tests are known failures for the previous few commits.
bangkok/lf95:
All tests pass
CAM tag used for the baseline comparison tests if different than previous
tag: cam3_4_06
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate): bit-for-bit on all platforms. One roundoff diagnostic on bluevista
which subsequently passed when run with debug turned on.
If bitwise differences were observed, how did you show they were no worse
than roundoff? Turn on debug and they went away. Also the difference in
absolute values was in the least significant bit of one number in
the domain.
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_4_06
Originator(s): eaton
Date: Wed Mar 14 12:12:51 MDT 2007
One-line Summary: refactor configure script
Purpose of changes:
. Refactor the configure script by moving the generic functionality of
managing a configuration file into a new module (Build::Config), and
leave the CAM specific parts in the configure script. A detailed design
with instructions for how to modify and extend configure is being
prepared as part of a new CAM Developer's Guide.
configure's commandline interface contains the following changes:
. -nlat, -nlon, -trk, -trm, -trn have been removed as well as removing the
"custom" option to -res. A new resolution is added by putting the
appropriate entry in the config_horiz_grid.xml file.
. The -res argument is being replaced by the option -hgrid to specify the
horizontal grid. The -res argument will be supported for backwards
compatibility.
. -nnadv has been removed since the feature of "non-advected
constituents" was removed at cam3_3_47.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system:
. new configure script.
. build-namelist updated to use new Build::Config module to read the
config_cache.xml file (which has a new structure).
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
models/atm/cam/bld/config_cam_eul_defaults.xml
models/atm/cam/bld/config_cam_fv_defaults.xml
models/atm/cam/bld/config_cam_homme_defaults.xml
models/atm/cam/bld/config_cam_sld_defaults.xml
models/atm/cam/bld/config_trop_chem_mozart_defaults.xml
models/atm/cam/bld/config_waccm_ghg_defaults.xml
models/atm/cam/bld/config_waccm_mozart_defaults.xml
models/atm/cam/bld/resolution_parameters.xml
models/atm/cam/bld/system_defaults.xml
List all subroutines added and what they do:
models/atm/cam/bld/config_defaults_eul.xml
models/atm/cam/bld/config_defaults_fv.xml
models/atm/cam/bld/config_defaults_homme.xml
models/atm/cam/bld/config_defaults_sld.xml
models/atm/cam/bld/config_defaults_trop_mozart.xml
models/atm/cam/bld/config_defaults_waccm_ghg.xml
models/atm/cam/bld/config_defaults_waccm_mozart.xml
. The new defaults files contain just the values that need to be set
models/atm/cam/bld/config_definition.xml
. The configuration definition lives in this file. This file contains
information about the config parameters such as valid values, and whether
the parameter can take a list of values. The config_cache.xml file
output by configure has the same structure and will contain the same
parameters as this file (though not in the same order).
models/atm/cam/bld/config_definition.xsl
. An XSLT stylesheet for the config_definition.xml file.
models/atm/cam/bld/config_horiz_grid.xml
. The horizonal grid specifiers are defined here.
models/atm/cam/bld/config_sys_defaults.xml
. This is the old system_defaults.xml file renamed for consistency with the
other config_* files.
models/atm/cam/bld/perl5lib/Build/Config.pm
. New Build::Config module.
models/atm/cam/bld/perl5lib/t/01.t
models/atm/cam/bld/perl5lib/t/config_cache.xml
models/atm/cam/bld/perl5lib/t/config_definition.xml
models/atm/cam/bld/perl5lib/t/config_setup_eul.xml
models/atm/cam/bld/perl5lib/t/debug.t
. test code for the Build::Config module
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/perl5lib/XML/Changes
models/atm/cam/bld/perl5lib/XML/Lite.pm
models/atm/cam/bld/perl5lib/XML/Lite/Element.pm
models/atm/cam/bld/perl5lib/XML/README
models/atm/cam/bld/perl5lib/XML/man3/XML::Lite.3
models/atm/cam/bld/perl5lib/XML/man3/XML::Lite::Element.3
. The XML::Lite module has just been moved under perl5lib
models/atm/cam/bld/CAM_config.pm
models/atm/cam/bld/build-namelist
. Update build-namelist to use the Build::Config module to read the
config_cache.xml file produced by configue.
models/atm/cam/bld/configure
. complete refactoring to make use of new Build::Config module
models/atm/cam/bld/mkDepends
. add fix for emacs fontlock
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: All PASS except:
059 cs998 TCS.ccsm.sh ERS f19_g13 K ...............................FAIL! rc= 2
The CCSM test is failing due to a share code inconsistency.
tempest: All PASS except:
008 sm133 TSM.sh e32pdh aqpgro 3s .................................FAIL! rc= 8
009 bl133 TBL.sh e32pdh aqpgro 3s .................................FAIL! rc= 4
025 sm333 TSM.sh f4pdh aqpgro 3s ..................................FAIL! rc= 8
026 bl333 TBL.sh f4pdh aqpgro 3s ..................................FAIL! rc= 4
039 sm533 TSM.sh s32pdh aqpgro 3s .................................FAIL! rc= 8
040 bl533 TBL.sh s32pdh aqpgro 3s .................................FAIL! rc= 4
The smoke tests started failing in cam3_4_04. Still looking into this.
bangkok/lf95: All PASS.
CAM tag used for the baseline comparison tests if different than previous
tag: cam3_4_03.
Summarize any changes to answers: none -- BFB.
===============================================================
===============================================================
Tag name: cam3_4_05
Originator(s): mvertens
Date: Tue Mar 13 11:04:53 MDT 2007
One-line Summary: upgraded to drvseq1_0_12
Purpose of changes:Updated component codes and driver to have flexibility
to recognize active versus data/dead atm/land components. This is needed to
optimize performance for4 certain component combinations such as datm/land coupling.
Bugs fixed (include bugzilla ID): None
Describe any changes made to build system: None
Describe any changes made to the namelist: None
List any changes to the defaults for the boundary datasets: None
Describe any substantial timing or memory changes: None
Code reviewed by: myself
List all subroutines eliminated: None
List all subroutines added and what they do: None
List all existing files that have been modified, and describe the changes:
M SVN_EXTERNAL_DIRECTORIES
- updated drv/seqmct to drvseq1_0_12
- updated csm_share/[shr/eshf/flds] to share3_070305
- updated clm2 to prof_clm3_expa_92_tags/prof02_clm3_expa_92
M models/ocn/dom/ocn_time_manager.F90
M models/ocn/dom/ocn_comp_mct.F90
updated to drvseq1_0_12
M models/ocn/som/ocn_time_manager.F90
M models/ocn/som/ocn_comp_mct.F90
updated to drvseq1_0_12
M models/ice/csim4/ice_time_manager.F90
M models/ice/csim4/ice_comp_mct.F90
updated to drvseq1_0_12
M models/atm/cam/src/control/atm_comp_mct.F90
updated to drvseq1_0_12
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: All PASS except
059 cs998 TCS.ccsm.sh ERS f19_g13 K ...............................FAIL! rc= 2
tempest: All PASS except
008 sm133 TSM.sh e32pdh aqpgro 3s .................................FAIL! rc= 8
009 bl133 TBL.sh e32pdh aqpgro 3s .................................FAIL! rc= 4
025 sm333 TSM.sh f4pdh aqpgro 3s ..................................FAIL! rc= 8
026 bl333 TBL.sh f4pdh aqpgro 3s ..................................FAIL! rc= 4
039 sm533 TSM.sh s32pdh aqpgro 3s .................................FAIL! rc= 8
040 bl533 TBL.sh s32pdh aqpgro 3s .................................FAIL! rc= 4
bangkok/lf95: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: none
===============================================================
===============================================================
Tag name: cam3_4_04
Originator(s): pworley
Date: Fri Mar 9 09:00:00 EST 2007
One-line Summary: new profiling interface; one processor EUL/SLD MPI fix
Purpose of changes:
. Add new profiling interface, isolating all calls to underlying timing
library in perf_mod.F90
. Replace compile time profiling options with runtime options via
a new namelist prof_inparm (in drv_in)
. Add support for one process MPI runs in the EUL and SLD dycores
Bugs fixed (include bugzilla ID):
. disabled a few shr_sys_flush(6) calls that caused the code
to crash on UNICOSMP systems (Cray X1/X1E)
Describe any changes made to build system:
Describe any changes made to the namelist:
. Added prof_inparm namelist:
profile_disable, profile_barrier, profile_single_file,
profile_depth_limit, profile_detail_limit, profile_timer
to drv_in
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: pworley
List all subroutines eliminated:
. ccsm_seq_timer_init (in con_cam.F90 and in seq_ccsm_drv.F90)
List all subroutines added and what they do:
in perf_mod.F90:
t_initf: replacement for ccsm_seq_timer_init, including support
for new namelist prof_inparm
t_profile_onf: logical function indicating whether profiling is active
at this point in the code
t_barrier_onf: logical function indicating whether timing barriers are
enabled
t_single_filef: logical function indicating whether one file or one
file per process should be generated for timing and spmdstats data
t_adj_detailf: increment or decrement timing detail level
t_barrierf: if enabled, t_startf(XXX), shr_sys_barrier, t_stopf(XXX)
t_finalizef: timing library clean-up
in FVperf_module.F90:
FVbarrierclock: if enabled, t_startf(XXX), mp_barrier, t_stopf(XXX)
List all existing files that have been modified, and describe the changes:
in CAM:
atm/cam/bld/NamelistsDescriptions.xml
atm/cam/bld/SeqCCSM_namelist.pm
atm/cam/bld/camexp.pm
atm/cam/bld: prof_inparm.pm
. added support to generate new namelist (prof_inparm) within drv_in
(contributed by Brian Eaton)
atm/cam/src/dynamics/eul/spmd_dyn.F90
atm/cam/src/dynamics/sld/spmd_dyn.F90
. added support for 1 processor MPI (SPMD) runs
atm/cam/src/chemistry/trop_mozart/chemistry.F90
atm/cam/src/chemistry/trop_mozart/mo_gas_phase_chemdr.F90
atm/cam/src/chemistry/trop_mozart/tracer_data.F90
atm/cam/src/chemistry/waccm_mozart/chemistry.F90
atm/cam/src/chemistry/waccm_mozart/iondrag.F90
atm/cam/src/chemistry/waccm_mozart/mo_gas_phase_chemdr.F90
atm/cam/src/chemistry/waccm_mozart/spedata.F90
atm/cam/src/control/cam_history.F90
atm/cam/src/control/ccsm_msg.F90
atm/cam/src/control/wrap_mpi.F90 (when WRAP_MPI_TIMING defined)
atm/cam/src/dynamics/eul/dyndrv.F90
atm/cam/src/dynamics/eul/dynpkg.F90
atm/cam/src/dynamics/eul/scandyn.F90
atm/cam/src/dynamics/eul/stepon.F90
atm/cam/src/dynamics/fv/metdata.F90
atm/cam/src/dynamics/homme/dp_coupling.F90
atm/cam/src/dynamics/homme/dyn_comp.F90
atm/cam/src/dynamics/homme/stepon.F90
atm/cam/src/dynamics/sld/dyndrv.F90
atm/cam/src/dynamics/sld/dynpkg.F90
atm/cam/src/dynamics/sld/scandyn.F90
atm/cam/src/dynamics/sld/stepon.F90
atm/cam/src/physics/cam1/advnce.F90
atm/cam/src/physics/cam1/cloudsimulator.F90
atm/cam/src/physics/cam1/convect_ke_intr.F90
atm/cam/src/physics/cam1/molec_diff.F90
atm/cam/src/physics/cam1/phys_gmean.F90
atm/cam/src/physics/cam1/prescribed_aerosols.F90
atm/cam/src/physics/cam1/radiation.F90
atm/cam/src/physics/cam1/stratiform.F90
atm/cam/src/physics/cam1/sulchem.F90
atm/cam/src/physics/cam1/tphysac.F90
atm/cam/src/physics/cam1/tphysbc.F90
atm/cam/src/physics/cam1/vertical_diffusion.F90
atm/cam/src/physics/cam1/zm_conv_intr.F90
atm/cam/src/physics/waccm/radheat.F90
ice/csim4/ice_comp.F90
ice/csim4/ice_comp_mct.F90
ice/csim4/ice_dh.F
ice/csim4/ice_srf.F90
ocn/dom/ocn_comp.F90
ocn/dom/ocn_comp_mct.F90
ocn/som/mixed_layer.F90
ocn/som/ocn_comp.F90
ocn/som/ocn_comp_mct.F90
utils/pilgrim/mod_comm.F90 (when TIMING defined)
utils/pilgrim/unit_testers/ghosttest.F90 (when MODCM_TIMING defined)
utils/pilgrim/unit_testers/parpatterntest.F90 (when MODCM_TIMING defined)
utils/pilgrim/unit_testers/redistributetest.F90 (when MODCM_TIMING defined)
. added 'use perf_mod'
atm/cam/src/control/atm_comp_mct.F90
. added 'use perf_mod' and one new timer event
atm/cam/src/dynamics/eul/courlim.F90
atm/cam/src/dynamics/eul/linemsdyn.F90
atm/cam/src/dynamics/eul/scan2.F90
atm/cam/src/dynamics/eul/scanslt.F90
atm/cam/src/dynamics/eul/spegrd.F90
atm/cam/src/dynamics/sld/courlim.F90
atm/cam/src/dynamics/sld/scan2.F90
atm/cam/src/dynamics/sld/scanslt.F90
atm/cam/src/dynamics/sld/spegrd.F90
atm/cam/src/physics/cam1/phys_grid.F90
atm/cam/src/dynamics/eul/dp_coupling.F90
atm/cam/src/dynamics/fv/dp_coupling.F90
atm/cam/src/dynamics/fv/dyn_comp.F90
atm/cam/src/dynamics/fv/stepon.F90
atm/cam/src/dynamics/sld/dp_coupling.F90
. added 'use perf_mod', added mpicom definition for no-SPMD case
where necessary, and replaced TIMING_BARRIERS logic with t_barrierf
atm/cam/src/control/cam_comp.F90
. added 'use perf_mod', added mpicom definition for no-SPMD case,
replaced TIMING_BARRIERS logic with t_barrierf, eliminated (via ifdef)
shr_sys_flush(6) for UNICOSMP, and replaced MULTIPLE_PERF_FILES
logic with t_single_filef logic for spmdstats
atm/cam/src/control/con_cam.F90
. added 'use perf_mod', removed call to usetlbuf (on CATAMOUNT),
replaced definition and call to ccsm_seq_timer_init with call to
t_initf, moved timer initialization, added t_adj_detailf(+2/-2)
to 'hide' initialization in default timing, added additional
timers (many sets, to match seq_ccsm_drv), modified t_prf call,
and added new t_finalizef call
atm/cam/src/dynamics/fv/FVperf_module.F90
. added 'use perf_mod' (when not in GEOS_MODE) and added
FVbarrierclock (as interface to t_barrierf, eventually;
implemented exlicitly until mpicom is exposed in Pilgrim
and modcomm calls)
atm/cam/src/dynamics/fv/cd_core.F90
atm/cam/src/dynamics/fv/trac2d.F90
. replaced TIMING_BARRIERS logic with FVbarrierclock
atm/cam/src/physics/cam1/physpkg.F90
. added 'use perf_mod', added mpicom definition for no-SPMD case,
replaced TIMING_BARRIERS logic with t_barrierf, added t_adj_detailf(+1/-1)
to hide instrumentation inside chunking loops
in CLM2
lnd/clm2/src/main/atmdrvMod.F90
lnd/clm2/src/main/clm_comp.F90
lnd/clm2/src/main/clm_csmMod.F90
lnd/clm2/src/main/program_csm.F90
lnd/clm2/src/main/program_off.F90
. added 'use perf_mod'
lnd/clm2/src/main/lnd_comp_mct.F90
lnd/clm2/src/riverroute/RtmMod.F90
lnd/clm2/src/main/driver.F90
. added 'use perf_mod', added mpicom definition for no-SPMD case where
necessary, and replaced TIMING_BARRIERS logic with t_barrierf
lnd/clm2/src/main/decompMod.F90
. eliminated (via ifdef) shr_sys_flush(6) for UNICOSMP
in DRV
drv/seq_mct/seq_ccsm_drv.F90
. added 'use perf_mod', removed call to usetlbuf (on CATAMOUNT),
replaced definition and call to ccsm_seq_timer_init with call to
t_initf, moved timer initialization, added t_adj_detailf(+2/-2)
to 'hide' initialization in default timing, replaced TIMING_BARRIERS
logic with t_barrierf, remove MULTIPLE_PERF_FILES logic and use
modified t_prf call, added new t_finalizef call, deleted stop 0.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: All PASS except
059 cs998 TCS.ccsm.sh ERS f19_g13 K ...............................FAIL! rc= 2
This test was also failing in the previous tag.
tempest: All PASS
bangkok/lf95: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
===============================================================
===============================================================
Tag name: cam3_4_03
Originator(s): mvertens, andrew, eaton
Date: Fri Mar 2 08:15:42 MST 2007
One-line Summary: restart bug fix; mod in radforce for lf95/debug
Purpose of changes:
. Fix restart bug (see below).
. Add mod to radforce code to make lf95/debug work. This is just changing
a full array copy (1:pcols) to a copy of only the subsection (1:ncol).
Bugs fixed (include bugzilla ID):
. [ID 405] The code currently crashes when writing the restart pointer
files for the dom, som, and csim components unless the user has
explicitly set these names via the namelist. This has be fixed by
providing appropriate default values.
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/ice/csim4/ice_comp.F90
models/ocn/dom/ocn_comp.F90
models/ocn/som/ocn_comp.F90
. add default for rest_pfile
models/atm/cam/test/system/nl_files/aqpgro
models/atm/cam/test/system/nl_files/fv2d_4tsk
models/atm/cam/test/system/nl_files/fv2d_8tsk
models/atm/cam/test/system/nl_files/ghgrmp
models/atm/cam/test/system/nl_files/no_ttrac
models/atm/cam/test/system/nl_files/off2x2.5
models/atm/cam/test/system/nl_files/off2x2.5p
models/atm/cam/test/system/nl_files/outfrq24h
models/atm/cam/test/system/nl_files/outfrq3s
models/atm/cam/test/system/nl_files/pghgsul
models/atm/cam/test/system/nl_files/scm_b4b_o1
models/atm/cam/test/system/nl_files/scm_prep
models/atm/cam/test/system/nl_files/ttrac
models/atm/cam/test/system/nl_files/ttrac_lb1
models/atm/cam/test/system/nl_files/ttrac_lb2
models/atm/cam/test/system/nl_files/ttrac_lb3
. remove setting of rest_pfile (the default is what we want to use)
models/atm/cam/src/physics/cam1/radiation.F90
. replace whole array assignments of pmxrgnrf and nmxrgnrf with versions
that copy the array section 1:ncol. This is to make lf95/debug happy.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: All PASS except
059 cs998 TCS.ccsm.sh ERS f19_g13 K ...............................FAIL! rc= 2
This test was also failing in the previous tag.
tempest: All PASS
bangkok/lf95: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: none
===============================================================
===============================================================
Tag name: cam3_4_02
Originator(s): cchen, sawyer, eaton
Date: Wed Feb 28 08:37:57 MST 2007
One-line Summary: fix FV omega calc, misc bug fixes
Purpose of changes:
. A new implementation of the omega calculation fix originally applied in
cam3_3_3 was provided by Jack Chen, and made to work with OMP by Will
Sawyer.
. A new subroutine, pbuf_setval, was added to the physics buffer module to
allow fields to be initialized to values other than the default (NaN).
Bugs fixed (include bugzilla ID):
. [ID387] Fix bug in FV mode reading an initial file that contains the
field DELP.
. Fix for subroutine fstrat_inti in mo_fstrat.F90 (trop_mozart chem).
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: cchen, sawyer, eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/src/dynamics/fv/dyn_comp.F90
models/atm/cam/src/dynamics/fv/diag_module.F90
models/atm/cam/src/dynamics/fv/cd_core.F90
models/atm/cam/src/dynamics/fv/te_map.F90
. FV omega fix
models/atm/cam/src/physics/cam1/phys_buffer.F90
. add pbuf_setval method for initializing physics buffer fields to values
other than the default (NaN).
models/atm/cam/src/chemistry/trop_mozart/mo_fstrat.F90 (fstrat_inti)
. remove inappropriate uses of beglat, endlat. This fix didn't affect
results in the configurations tested by the regression tests.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: PASS except
001 sb998 TSB.ccsm.sh ERS f19_g13 K ...............................FAIL! rc= 5
019 bl331 TBL.sh f4c11gdh pghgsul 9s ..............................FAIL! rc= 7
022 bl332 TBL.sh f4sdh ghgrmp 9s ..................................FAIL! rc= 7
027 bl334 TBL.sh f4dh adia 9s .....................................FAIL! rc= 7
030 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7
033 bl337 TBL.sh f4dh fv2d_8tsk 9s ................................FAIL! rc= 7
035 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7
038 bl354 TBL.sh fm2dh outfrq3s 9s ................................FAIL! rc= 7
041 bl355 TBL.sh fmo2dh off2x2.5 9s ...............................FAIL! rc= 7
059 cs998 TCS.ccsm.sh ERS f19_g13 K ...............................FAIL! rc= 2
The CCSM test fails for same reason as in previous tag.
The failed baseline tests are expected failures due to the change in the
diagnostic FV omega calculation. The only fields with diffs are OMEGA,
OMEGAT, UW3d and WTH3d.
tempest: PASS except
017 bl314 TBL.sh wg10dh outfrq3s 9s ...............................FAIL! rc= 7
021 bl331 TBL.sh f4c11gdh pghgsul 9s ..............................FAIL! rc= 7
024 bl332 TBL.sh f4sdh ghgrmp 9s ..................................FAIL! rc= 7
029 bl335 TBL.sh f4dh idphys 9s ...................................FAIL! rc= 7
031 bl376 TBL.sh f1.9o outfrq24h 2d ...............................FAIL! rc= 7
The failed baseline tests are expected failures due to the change in the
diagnostic FV omega calculation. The only fields with diffs are OMEGA,
OMEGAT, UW3d and WTH3d.
bangkok/lf95: PASS except
016 bl311 TBL.sh f10c8mdm ttrac 9s ................................FAIL! rc= 7
024 bl312 TBL.sh f10sdm ghgrmp 9s .................................FAIL! rc= 7
029 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 7
031 bl377 TBL.sh f1.9m outfrq24h 2d ...............................FAIL! rc= 7
These are expected failures due to the change in the diagnostic FV omega
calculation. The only fields with diffs are OMEGA and OMEGAT.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: Answers are BFB except for the diagnostic
FV omega calculation.
===============================================================
===============================================================
Tag name: cam3_4_01
Originator(s): mvr, mvertens
Date: 27 Feb 2007
One-line Summary: cam dom/som now called after cam run phase
Purpose of changes:
Summary: Moved the ocean component to now be run after atmospheric component to
permit incorporation of an active ocean component that is only called
once per day. Made the driver code and specification of surface fields
exchanged between components part of SVN_EXTERNAL_DIRECTORIES.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
- Introduced separate hard-wired namelist files for cam (atm_in),
cam-dom/som (ocn_in), cam-csim (ice_in) and the sequential driver (drv_in).
- The new cam-csim namelist is as follows:
namelist /csim_inparm/ prognostic_icesnow, reset_csim_iceprops, &
ice_conschk_frq, icecyc, rest_pfile, nrevsn
- The new cam-dom namelist is as follows:
namelist /dom_inparm/ sstcyc, nrevsn, rest_pfile
- The new cam-som namelist is as follows:
namelist /som_inparm/ sstcyc, som_conschk_frq, rest_pfile, nrevsn
- The new driver namelist file contains the namelists
ccsm_inparm and timemgr_inparm (which were previous in single namelist file "namelist")
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: mvertens, mvr
List all subroutines eliminated:
D models/drv
- directory removed and made external
D models/atm/cam/bld/atmlndnl.pm
List all subroutines added and what they do:
A + models/ocn/dom/ocn_filenames.F90
A + models/ocn/dom/ocn_time_manager.F90
A + models/ocn/som/ocn_filenames.F90
A + models/ocn/som/ocn_time_manager.F90
A + models/ice/csim4/ice_filenames.F90
A + models/ice/csim4/ice_time_manager.F90
- Made cam-dom, cam-som and cam-csim no longer dependent on the cam time
manager. Now each of these components has a separate time manager module
that is initialized from the sync clock the same way as the cam time manager
is initialized from the sync clock. Also made cam-dom, cam-som and cam-csim
have a cam-independent specification for their restart file names. These
routines (ocn_filenames.F90 and ice_filenames.F90 now utilize the component
specific time manager routines rather than the cam time manager).
A + models/ocn/som/ocn_constants.F90
- Added new module to specify tfrez and cpw independent of the cam-csim code.
This removed an implicit dependency between the cam-som and cam-csim code.
A + models/atm/cam/bld/som_inparm.pm
A + models/atm/cam/bld/DefaultDOM_INPARM_Namelist.xml
A + models/atm/cam/bld/dom_inparm.pm
A + models/atm/cam/bld/DefaultCSIM_INPARM_Namelist.xml
A + models/atm/cam/bld/compnl.pm
A + models/atm/cam/bld/DefaultSOM_INPARM_Namelist.xml
A + models/atm/cam/bld/csim_inparm.pm
- new files needed for the breakout of the namelists
List all existing files that have been modified, and describe the changes:
M models/ocn/dom/ocn_spmd.F90
- removed #SPMD if def
M models/ocn/dom/sst_data.F90
- removed boudning of sst temp by freezing point of water:
! Bound the sst temp by the freezing point of sea water.
! sst(i,lchnk) = max(sst(i,lchnk),tsice)
the logic is now consistent with that utilized in docn7 dom mode
- also minor cleanup of logic to get code to work with moving the ocn call to after the
atmosphere call
M models/ocn/dom/ocn_types.F90
- removed lwup from ocn_types
M models/ocn/dom/ocn_comp.F90
- incorporated cam-dom specific namelist, time manager and restart file naming utility
M models/ocn/dom/ocn_comp_mct.F90
- introduced new cdata input data structure that his info buf flag capability
- modified domain generation such that it removed maxfrac as an attribute of dom_o
and added aream as a new attribute (area from SCRIP mapping file)
M models/ocn/som/mixed_layer.F90
- removed #ifdef SPMD, removed extra call to time mixed_layer2
- removed "use ice_constants" for tref by adding new ocn_constants module
M models/ocn/som/ocn_spmd.F90
- removed #ifdef SPMD
M models/ocn/som/ocn_types.F90
- removed tsocn(:) (no longer needed)
M models/ocn/som/somint.F90
- removed "use cam_control_mod, only: sstcyc", now obtained from cam-som namelist
- replaced "use time_manager" (cam time manager) with "use ocn_time_manager"
M models/ocn/som/ocn_comp.F90
- incorporated cam-dom specific namelist, time manager and restart file naming utility
M models/ocn/som/ocean_data.F90
- introduced module variable sstcyc which is now set from som specific namelist
M models/ocn/som/somini.F90
- removed "use cam_control_mod, only: sstcyc", now obtained from cam-som namelist
- replaced "use time_manager" (cam time manager) with "use ocn_time_manager"
- removed ifdef SPMD
M models/ocn/som/ocn_comp_mct.F90
- introduced new cdata input data structure that his info buf flag capability
- modified domain generation such that it removed maxfrac as an attribute of dom_o
and added aream as a new attribute (area from SCRIP mapping file)
M models/ice/csim4/ice_spmd.F90
- removed #ifdef SPMD
M models/ice/csim4/ice_data.F90
- removed #ifdef SPMD
- removed "use ocn_comp, only : sst_file_get_id"
- removed reference to aqua_planet
- replaced "use time_manager" (cam time manager) with "use ocn_time_manager"
M models/ice/csim4/ice_dh.F
- added module variables prognostic_icesnow, reset_csim_iceprops that will now be used by
cam-csim namelist
M models/ice/csim4/ice_srf.F90
- removed use statements for cam_history and cam_control_mod
M models/ice/csim4/ice_comp.F90
- incorporated cam-dom specific namelist, time manager and restart file naming utility
- removed module variables asdirice, aldirice, asdifice, aldifice
- moved addfld calls to cam_diagnostics
M models/ice/csim4/ice_diagnostics.F
- removed use of cam time manager and introduced use of ice time manager
M models/ice/csim4/ice_comp_mct.F90
- introduced new cdata input data structure that his info buf flag capability
- modified domain generation such that it removed maxfrac as an attribute of dom_o
and added aream as a new attribute (area from SCRIP mapping file)
M models/atm/cam/test/system/TBL.sh
- mod to diagnostic output message
M models/atm/cam/test/system/TMC.sh
- bug fix to parsing of tracer output for mass conservation test
M models/atm/cam/test/system/TBR.sh
M models/atm/cam/test/system/TSM.sh
M models/atm/cam/test/system/TER.sh
- namelist broken out into individual files for each component; branch test needed
pointers to restart files specified in build-namelist call
M models/atm/cam/test/system/nl_files/ghgrmp
M models/atm/cam/test/system/nl_files/outfrq3s
M models/atm/cam/test/system/nl_files/off2x2.5p
M models/atm/cam/test/system/nl_files/aqpgro
M models/atm/cam/test/system/nl_files/outfrq24h
M models/atm/cam/test/system/nl_files/adia
M models/atm/cam/test/system/nl_files/idphys
M models/atm/cam/test/system/nl_files/fv2d_4tsk
M models/atm/cam/test/system/nl_files/scm_prep
M models/atm/cam/test/system/nl_files/scm_b4b_o1
M models/atm/cam/test/system/nl_files/pghgsul
M models/atm/cam/test/system/nl_files/off2x2.5
M models/atm/cam/test/system/nl_files/no_ttrac
M models/atm/cam/test/system/nl_files/fv2d_8tsk
M models/atm/cam/test/system/nl_files/ttrac_lb1
M models/atm/cam/test/system/nl_files/ttrac_lb2
M models/atm/cam/test/system/nl_files/ttrac
M models/atm/cam/test/system/nl_files/ttrac_lb3
- all namelist options files modified with new pointers to component restart files
M models/atm/cam/bld/config_cam_eul_defaults.xml
M models/atm/cam/bld/configure
M models/atm/cam/bld/DefaultCLM_INPARM_Namelist.xml
M models/atm/cam/bld/config_trop_chem_mozart_defaults.xml
M models/atm/cam/bld/camexp.pm
M models/atm/cam/bld/filter_nl.pm
M models/atm/cam/bld/config_waccm_ghg_defaults.xml
M models/atm/cam/bld/clm_inparm.pm
M models/atm/cam/bld/config_waccm_mozart_defaults.xml
M models/atm/cam/bld/config_cam_fv_defaults.xml
M models/atm/cam/bld/DefaultCTL_NL_Namelist.xml
M models/atm/cam/bld/cam_inparm.pm
M models/atm/cam/bld/NamelistsDescriptions.xml
M models/atm/cam/bld/timemgr_inparm.pm
M models/atm/cam/bld/ctl_nl.pm
M models/atm/cam/bld/config_cam_homme_defaults.xml
M models/atm/cam/bld/CAM_config.pm
M models/atm/cam/bld/clmexp.pm
M models/atm/cam/bld/SeqCCSM_namelist.pm
M models/atm/cam/bld/ccsm_inparm.pm
M models/atm/cam/bld/build-namelist
M models/atm/cam/bld/config_cam_sld_defaults.xml
M models/atm/cam/bld/DefaultFILTER_NL_Namelist.xml
- mods required to account for breakout of the namelists
M models/atm/cam/src/control/runtime_opts.F90
M models/atm/cam/src/control/cam_control_mod.F90
- removed cam namelists for the following variables
sstcyc, icecyc, prognostic_icesnow, reset_csim_iceprops, ice_conschk_frq, som_conschk_frq
M models/atm/cam/src/control/ccsm_msg.F90
- moved addfld for history variables calls to cam_diagnostics.F90 (see below)
M models/atm/cam/src/control/con_cam.F90
- hardwired atm.stdin as namelist for nlfilename argument to cam_init
M models/atm/cam/src/control/atm_comp_mct.F90
- introduced the possibility to do flux averaging of precipitation (as is done in cpl6)
this will be required when cam communicates with the sequential driver only on time steps when
the short wave computation is done
- introduced new cdata input data structure that his info buf flag capability
- moved verification of surface fractions from mrg_x2a_mct.F90 to this module (do not want
to do this verification for dead components)
- modified domain generation such that it removed maxfrac as an attribute of dom_o
and added aream as a new attribute (area from SCRIP mapping file)
M models/atm/cam/src/control/cam_comp.F90
- hard-wired atm_in and argument to NLFileName in call to cam_initial
- moved call to intht to the initialization phase of cam
- moved calls to addfld of FSNSOI, FLNSOI, LHFLXOI and SHFLXOI from atm_comp_mct.F90 to this module
M models/atm/cam/src/utils/spmd_utils.F90
- added public attribute to mpicom module variable when SPMD is not defineda
M models/atm/cam/src/physics/cam1/radiation.F90
- added new routine, radiation_newsw_cday, which returns the calendar day of the
next sw radiation calculation
- added optional input variable, "timestep", to function radiation_do()
M models/atm/cam/src/physics/cam1/cam_diagnostics.F90
- moved the addfld calls from som/ocn_comp.F90 to this module
- moded the addfld calls from dom/ocn_comp.F90 to tihs module
- moved the addfld calls from ccsm_msg.F90 to this module
- moded the addfld calls from ice/csim/ice_comp.F90 to this module
M SVN_EXTERNAL_DIRECTORIES
- externals now include drv code and updated to clm3_expa_92
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista:
all baseline tests fail due to change in answers; ccsm test fails because
latest ccsm tag does not yet have necessary mods...when a ccsm sandbox
was created with ccsm3_1_beta44, clm updated to clm3_expa_91, csm_share
updated to share3_070220, MCT updated to MCT2_3_0_070206, and scripts
updated with small change to cam.template for new namelists, the ccsm
test passes
tempest:
all baseline tests fail due to change in answers
bangkok/lf95:
all baseline tests fail due to change in answers
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: ALL
- what platforms/compilers: ALL
- nature of change: larger than roundoff but same climate
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository): drvord13_cam3_3_43
- platform/compilers: bluevista/mpxlf90_r
- configure commandline:
som run:
$cfgdir/configure -spmd -smp -dyn fv -res 1.9x2.5 -ocn som
dom run:
$cfgdir/configure -spmd -smp -dyn fv -res 1.9x2.5
- build-namelist command (or complete namelist):
som run:
&ccsm_inparm
case_name = 'camrun_som'
start_type = "continue"
/
&timemgr_inparm
atm_cpl_dt = 1800
orb_iyear_ad = 1950
restart_option = 'monthly'
start_ymd = 101
stop_n = 24
stop_option = 'nmonths'
/
&cam_inparm
absems_data = '/fs/cgd/csm/inputdata/atm/cam/rad/abs_ems_factors_fastvx.c030508.nc'
aeroptics = '/fs/cgd/csm/inputdata/atm/cam/rad/AerosolOptics_c050419.nc'
bnd_topo = '/fs/cgd/csm/inputdata/atm/cam/topo/USGS-gtopo30_1.9x2.5_remap_c050602.nc'
bndtvaer = '/fs/cgd/csm/inputdata/atm/cam/rad/AerosolMass_V_1.9x2.5_clim_c040811.nc'
bndtvo = '/fs/cgd/csm/inputdata/atm/cam/ozone/pcmdio3.r8.64x1_L60_clim_c970515.nc'
bndtvs = '/fis/cgd/cms/rneale/cam_mydata/som/cam3_2_59_bvc_0005-0020_qflux.nc'
doisccp = .true.
dtime = 1800
fincl1 = 'CLDST'
isccpdata = '/fs/cgd/csm/inputdata/atm/cam/rad/isccp.tautab_invtau.nc'
ncdata = '/fis/cgd/cms/rneale/cam_mydata/som/cam3_2_59_som0.cam2.i.0050-01-01-00000.nc'
/
&clm_inparm
dtime = 1800
finidat = '/fis/cgd/cms/rneale/cam_mydata/som/cam3_2_59_som0.clm2.r.0050-01-01-00000.nc'
fpftcon = '/fs/cgd/csm/inputdata/lnd/clm2/pftdata/pft-physiology-cn16.c040719'
fsurdat = '/fs/cgd/csm/inputdata/lnd/clm2/srfdata/cam/clms_3.1_1.9x2.5_c050603.nc'
/
dom run:
&ccsm_inparm
case_name = 'camrun_dom'
start_type = "continue"
/
&timemgr_inparm
atm_cpl_dt = 1800
orb_iyear_ad = 1950
restart_option = 'monthly'
start_ymd = 101
stop_n = 24
stop_option = 'nmonths'
/
&cam_inparm
absems_data = '/fs/cgd/csm/inputdata/atm/cam/rad/abs_ems_factors_fastvx.c030508.nc'
aeroptics = '/fs/cgd/csm/inputdata/atm/cam/rad/AerosolOptics_c050419.nc'
bnd_topo = '/fs/cgd/csm/inputdata/atm/cam/topo/USGS-gtopo30_1.9x2.5_remap_c050602.nc'
bndtvaer = '/fs/cgd/csm/inputdata/atm/cam/rad/AerosolMass_V_1.9x2.5_clim_c040811.nc'
bndtvo = '/fs/cgd/csm/inputdata/atm/cam/ozone/pcmdio3.r8.64x1_L60_clim_c970515.nc'
bndtvs = '/fs/cgd/csm/inputdata/atm/cam/sst/sst_HadOIBl_bc_1.9x2.5_clim_c040810.nc'
doisccp = .true.
dtime = 1800
fincl1 = 'CLDST'
isccpdata = '/fs/cgd/csm/inputdata/atm/cam/rad/isccp.tautab_invtau.nc'
ncdata = '/fis/cgd/cms/rneale/cam_mydata/atm/cam3_2_59_bvc.cam2.i.0023-01-01-00000.nc'
/
&clm_inparm
dtime = 1800
finidat = '/fis/cgd/cms/rneale/cam_mydata/lnd/cam3_2_59_bvc.clm2.i.0023-01-01-00000.nc'
fpftcon = '/fs/cgd/csm/inputdata/lnd/clm2/pftdata/pft-physiology-cn16.c040719'
fsurdat = '/fs/cgd/csm/inputdata/lnd/clm2/srfdata/cam/clms_3.1_1.9x2.5_c050603.nc'
/
- MSS location of output:
/MVR/csm/camrun_som
/MVR/csm/camrun_dom
MSS location of control simulations used to validate new climate:
/RNEALE/csm/cam3_2_59_som0/atm/hist
/RNEALE/csm/cam3_2_59_bvc/atm/hist
URL for AMWG diagnostics output used to validate new climate:
MSS:/MVR/csm/camrun_som/camrun_som-cam3_2_59_som0.tar
MSS:/MVR/csm/camrun_dom/camrun_dom-cam3_2_59_bvc.tar
===============================================================
===============================================================
Tag name: cam3_4_00
Originator(s): erik
Date: Thu Feb 22 16:04:59 MST 2007
One-line Summary: Update CLM and csm_share
Purpose of changes: Update CAM to use latest CLM with new hydrology changes in it
Bugs fixed (include bugzilla ID): 307
Describe any changes made to build system: Add SEQ_MCT CPP token for seqCCSM driver
Need to use different CAM_CCSMROOT for CCSM tests see below
Describe any changes made to the namelist: None
List any changes to the defaults for the boundary datasets: New datasets for CLM
Describe any substantial timing or memory changes: None
Code reviewed by: eaton
List all SVN Externals updated:
Update csm_share to share3_070220 -- Has new share freezing constants
Update CLM to clm3_expa_91 --------- New hydrology
Update MCT to MCT2_3_0_070206 ------ New features added to mpi-serial needed for CLM
List all subroutines eliminated:
D models/atm/cam/test/system/scamtest.sh -- no longer used by test_driver.sh
List all subroutines added and what they do:
None, other than updates to new externals.
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/TSC.sh -- leave SCAM files if CAM_RETAIN_FILES is true
M models/atm/cam/bld/DefaultCLM_INPARM_Namelist.xml -- Add new CLM hydrology datasets
M models/atm/cam/bld/Makefile.in -- Add SEQ_MCT define to build
M models/drv/seq_mct/flux_ao.F90 -- rm a shr_const_ value now defined in shr_const_mod.F90
Add CPP Token SEQ_MCT to SEQ_MCT drivers for each model component.
M models/ocn/dom/ocn_comp_mct.F90
M models/ocn/som/ocn_comp_mct.F90
M models/ice/csim4/ice_comp_mct.F90
M models/atm/cam/src/control/atm_comp_mct.F90
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: All PASS except TBL tests -- because answers change
005 bl131 TBL.sh e32c11dh pghgsul 9s ..............................FAIL! rc= 7
008 bl132 TBL.sh e32sdh ghgrmp 9s .................................FAIL! rc= 7
010 bl133 TBL.sh e32pdh aqpgro 3s .................................FAIL! rc= 7
013 bl134 TBL.sh e32dh adia 9s ....................................FAIL! rc= 7
015 bl151 TBL.sh e64h outfrq24h 2d ................................FAIL! rc= 7
019 bl331 TBL.sh f4c11gdh pghgsul 9s ..............................FAIL! rc= 7
022 bl332 TBL.sh f4sdh ghgrmp 9s ..................................FAIL! rc= 7
024 bl333 TBL.sh f4pdh aqpgro 3s ..................................FAIL! rc= 7
027 bl334 TBL.sh f4dh adia 9s .....................................FAIL! rc= 7
030 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7
033 bl337 TBL.sh f4dh fv2d_8tsk 9s ................................FAIL! rc= 7
035 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7
038 bl354 TBL.sh fm2dh outfrq3s 9s ................................FAIL! rc= 7
041 bl355 TBL.sh fmo2dh off2x2.5 9s ...............................FAIL! rc= 7
045 bl531 TBL.sh s32c11dh pghgsul 9s ..............................FAIL! rc= 7
048 bl532 TBL.sh s32sdh ghgrmp 9s .................................FAIL! rc= 7
050 bl533 TBL.sh s32pdh aqpgro 3s .................................FAIL! rc= 7
053 bl534 TBL.sh s32dh adia 9s ....................................FAIL! rc= 7
055 bl551 TBL.sh s64h outfrq24h 2d ................................FAIL! rc= 7
setenv CAM_CCSMROOT /fis/cgd/home/erik/ccsm3_1_beta44+clm91+shr+mct
(this is a CCSM sandbox with ccsm3_1_beta44, clm updated to clm3_expa_91,
csm_share updated to share3_070220 and MCT updated to MCT2_3_0_070206)
tempest: All PASS except TBL tests -- because answers change
004 bl131 TBL.sh e32c11dh pghgsul 9s ..............................FAIL! rc= 7
007 bl132 TBL.sh e32sdh ghgrmp 9s .................................FAIL! rc= 7
009 bl133 TBL.sh e32pdh aqpgro 3s .................................FAIL! rc= 7
012 bl135 TBL.sh e32dh idphys 9s ..................................FAIL! rc= 7
014 bl152 TBL.sh e64o outfrq24h 2d ................................FAIL! rc= 7
017 bl314 TBL.sh wg10dh outfrq3s 9s ...............................FAIL! rc= 7
021 bl331 TBL.sh f4c11gdh pghgsul 9s ..............................FAIL! rc= 7
024 bl332 TBL.sh f4sdh ghgrmp 9s ..................................FAIL! rc= 7
026 bl333 TBL.sh f4pdh aqpgro 3s ..................................FAIL! rc= 7
029 bl335 TBL.sh f4dh idphys 9s ...................................FAIL! rc= 7
031 bl376 TBL.sh f1.9o outfrq24h 2d ...............................FAIL! rc= 7
035 bl531 TBL.sh s32c11dh pghgsul 9s ..............................FAIL! rc= 7
038 bl532 TBL.sh s32sdh ghgrmp 9s .................................FAIL! rc= 7
040 bl533 TBL.sh s32pdh aqpgro 3s .................................FAIL! rc= 7
043 bl535 TBL.sh s32dh idphys 9s ..................................FAIL! rc= 7
045 bl552 TBL.sh s64o outfrq24h 2d ................................FAIL! rc= 7
bangkok/lf95: All PASS except TBL tests -- because answers change
004 bl111 TBL.sh e8c8mdm ttrac 9s .................................FAIL! rc= 5
008 bl112 TBL.sh e8sdm ghgrmp 9s ..................................FAIL! rc= 5
010 bl113 TBL.sh e8pdm aqpgro 3s ..................................FAIL! rc= 5
012 bl153 TBL.sh e64m outfrq24h 2d ................................FAIL! rc= 5
016 bl311 TBL.sh f10c8mdm ttrac 9s ................................FAIL! rc= 5
024 bl312 TBL.sh f10sdm ghgrmp 9s .................................FAIL! rc= 5
026 bl313 TBL.sh f10pdm aqpgro 3s .................................FAIL! rc= 5
029 bl317 TBL.sh f10dm fv2d_4tsk 9s ...............................FAIL! rc= 5
031 bl377 TBL.sh f1.9m outfrq24h 2d ...............................FAIL! rc= 5
035 bl511 TBL.sh s8c8mdm ttrac 9s .................................FAIL! rc= 5
039 bl512 TBL.sh s8sdm ghgrmp 9s ..................................FAIL! rc= 5
041 bl513 TBL.sh s8pdm aqpgro 3s ..................................FAIL! rc= 5
043 bl553 TBL.sh s64m outfrq24h 2d ................................FAIL! rc= 4
Summarize any changes to answers, i.e.,
- what code configurations: All
- what platforms/compilers: All
- nature of change (new CLM climate, and ice freezing point change)
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
In run database...
http://ccsm-rundb.cgd.ucar.edu/case_display.php?case=b31.020ws
MSS location of control simulations used to validate new climate: /CCSM/csm/b31.020ws
Note --this case does NOT include the TKFRZ constant change.
URL for AMWG diagnostics output used to validate new climate:
http://www.cgd.ucar.edu/tss/clm/diagnostics/ccsm2_prod/b31.020ws_atma-b31.002_atma/sets.htm
===============================================================
===============================================================
Tag name: cam3_3_51
Originator(s): mvr
Date: 9 Feb 2007
One-line Summary: Test driver enabled for blueice; bug fix for the use
of env var CAM_ACCOUNT to specify charge account
Purpose of changes: the new ibm machine blueice cam online feb 1; the
recent implementation of the environment variable CAM_ACCOUNT
failed to use the account specified when spawning the CCSM tests
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: myself
List all subroutines eliminated:
none
List all subroutines added and what they do:
A README_EXTERNALS
- documentation for how to work with cam's external directories
A models/atm/cam/test/system/tests_posttag_blueice
- default list of tests to run posttag on blueice
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/TSB.ccsm.sh
M models/atm/cam/test/system/TCT.ccsm.sh
- CCSM test scripts enabled for use on blueice; now uses CAM_ACCOUNT setting
for spawning CCSM tests with user-specified charge account
M models/atm/cam/test/system/tests_posttag_bluesky
M models/atm/cam/test/system/tests_pretag_bluesky
- default set of tests on bluesky now include tests for homme and scam
M models/atm/cam/test/system/test_driver.sh
- top level test script enabled for use on blueice; now exports CAM_ACCOUNT
M models/atm/cam/test/system/CAM_runcmnd.sh
- script enabled to construct run command appropriate for blueice
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: all PASS
tempest: all PASS
bangkok/lf95: all PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_3_50
Originator(s): eaton, mvr, jedwards
Date: Sun Feb 4 15:01:47 MST 2007
One-line Summary: fixes for file archiving & eshr time manager; test suite enhancements
Purpose of changes:
. Fix problems observed in recent NCAR production runs involving failed
transfers to MSS and failed restarts on automatic resubmissions. Part of
the problem appears to be due to the LSF batch system killing msrcp
processes when the CAM processes are finished. The LoadLeveler batch
system did not do this. The fix implemented was designed to restore the
same behavior we had under LoadLeveler by issuing the msrcp commands that
occur on the final timestep syncronously, i.e., don't let CAM finish
until the requested file transfers are successful.
Note that having CAM be responsible for file archiving is undesirable for
several reasons, chief among them being that the implementation is not
portable. We intend to implement a file archiving script for CAM similar
to the file harvester used by CCSM. But the current ability of CAM to
archive files at NCAR will be maintained until a better solution is in
place.
. Fix logic for determining the last timestep in share code for ESMF
time manager. This eliminates the need to use the optional arg nextstep
to get the correct behavior.
. Enhancements to test suite (from mvr) include:
- Allow user to specify a project number on SCD machines via the
CAM_ACCOUNT environment variable.
- Print summary of the cprnc output to the log file, including a list of
fields that contain differences.
Bugs fixed (include bugzilla ID):
. Fix problem in boundarydata.F90 (from jedwards) that was causing seg
fault trying to interpolate unstructured grid cyclic dataset between
December and January.
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
SVN_EXTERNAL_DIRECTORIES
. update csm_share externals to point to csm_share/branch_tags/scmshr_tags/scmshr02
models/csm_share/eshr/eshr_timemgr_mod.F90
. fix logic in eshr_timeMgr_clockisOnLastStep
. change eshr_timemgr_clockAlarmIsOnRes to call
eshr_timeMgr_clockisOnLastStep without optional nextStep=.true. arg
. add method eshr_timemgr_curTimeLEstopTime which returns true when the
current time is <= the stop time.
models/atm/cam/src/control/atm_comp_mct.F90
. remove nextstep=.true. arg from call to eshr_timeMgr_clockIsOnLastStep
that sets nlend.
models/atm/cam/src/control/con_cam.F90
. remove unused reference to eshr_timemgr_clockIsOnLastStep
models/drv/seq_mct_drv/seq_ccsm_drv.F90
. use new method eshr_timeMgr_curTimeLEstopTime in timestep loop
. remove nextstep=.true. from call to eshr_timeMgr_clockisOnLastStep which
sets the logical for removing files
. modify the last msrcp call, which is for the driver restart file, to be
synchronous.
models/atm/cam/src/control/ioFileMod.F90
. modify putfil so that at the end of run when the files are not to be
removed after archiving, the archiving commands are issued
synchronously. This is to avoid having the batch queue kill the
processes that are trying to archive files.
models/atm/cam/src/physics/cam1/boundarydata.F90
. fix allocate statement that was causing seg fault trying to interpolate
unstructured grid cyclic dataset between december and january.
models/atm/cam/bld/DefaultTIMEMGR_INPARM_Namelist.xml
. add default start_ymd for HOMME ne5np8 resolution initial file.
models/atm/cam/bld/DefaultCTL_NL_Namelist.xml
. change statefreq to 60 from 1 (per J Edwards).
models/atm/cam/test/system/TBL.sh
. Remove grepping for non-zero comparisons in cprnc output -- this belongs
in the CAM_compare.sh script
models/atm/cam/test/system/CAM_compare.sh
. Enhance to print summary of cprnc output to the log file, including a
list of fields that contain differences.
models/atm/cam/test/system/test_driver.sh
. Allow user to specify project number on IBMs via CAM_ACCOUNT environment
variable.
. Modify LD_LIBRARY_PATH to include pgi libs even when using lf95 for
testing on CGD clusters. This is because cprnc was built with pgi.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: All PASS except:
057 bl711 TBL.sh h5x8dm adia 9s ...................................FAIL! rc= 7
This failure is expected since I fixed the start_ymd to be consistent with
the initial file, and changed the value of statefreq.
tempest: All PASS
bangkok/lf95: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: bit-for-bit
===============================================================
===============================================================
Tag name: cam3_3_49
Originator(s): Will Sawyer, Lawrence Takacs, Bill Putman
Date: Thu Feb 1 11:41:00 MST 2007
One-line Summary: Merge of GEOS5 and CAM fv-dycores
Purpose of changes:
o Added mass fluxes to the output of the FV dycore (Putman)
o Added cubic vertical interpolation (Takacs) to used optionally
(flag te_method=1) in the remapping.
o More FV constants were cleaned up in a consistent way (Bug #214)
in particular, pi=4*atan(1.0) was replaced by F90 parameter with
sufficient significant digits (cause non-zero differences!)
o Refactoring to remove many CPP #if defined(GEOS_MODE) distinctions
Bugs fixed (include bugzilla ID): #214
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: no difference
Code reviewed by: Sawyer, Takacs, Suarez, Putman
List all subroutines eliminated: none
List all subroutines added and what they do: c2a3d (cubic interpolation
from a C to an A grid) in module dynamics_vars.F90.
List all existing files that have been modified, and describe the changes:
In models/atm/cam/src/dynamics/fv
fill_module.F90: revised definition of PI (R16), no longer 4*atan(1)
dyn_comp.F90: Horizontal mass fluxes (Putman), remapping method
switch (te_method).
mapz_module.F90: Cubic interpolation (Takacs), removed references to
state and grid, in particular clocks.
inital.F90: dyn_init call now has TE_METHOD as an argument
cd_core.F90: bug fix: zero out the first latitude of UC (which
should never be used, but *is* written to the restart.
dynamics_vars.F90: c2a3d (Putman), removed all references to ESMF
(GEOS_MODE), PI passed in as argument
te_map.F90: Horizontal mass fluxes (Putman), optional use
of cubic interpolation (activated by te_method flag)
Isolated magic numbers as F90 parameters.
Partial remapping (Takacs et al.), considerable
code restructuring (zero difference changes)
restart_dynamics.F90: dyn_init call now has TE_METHOD as an argument
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: Fails baseline tests (bl331,332, 333, 334, 336, 337, 354,
355, 375) due to PI roundoff. Passes perturbation growth test.
Fails cs998 (CCSM run -- known problem)
tempest: passes all tests
bangkok/lf95: passes all tests
CAM tag used for the baseline comparison tests if different than previous
tag: cam3_3_48 (previous tag)
Summarize any changes to answers, i.e.,
- what code configurations: WACCM, FV
- what platforms/compilers: Bluevista
- nature of change (roundoff; larger than roundoff but same climate; new
climate): roundoff
If bitwise differences were observed, how did you show they were no worse
than roundoff? perturbation growth test
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_3_48
Originator(s): Stacy Walters, Francis Vitt
Date: 25 Jan 2007
One-line Summary:
Scientific updates for WACCM
Purpose of changes:
- Improved WACCM gravity wave drag
- Enhanced diagnostic outputs
- Added QBO forcing in WACCM physics
- Added NOx and HOx production from solar proton events (SPE)
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
logical :: qbo_use_forcing ! .true. => qbo forcing used; default = .false.
character(len=256) :: qbo_forcing_file ! qbo forcing data file; default = 'NO_QBO_FILE'
! if qbo_use_forcing == .true. this must be set
logical :: qbo_cyclic ! .true. => qbo forcing file is to be used with
! a 28 month periodicity otherwise model time is
! used to access qbo_forcing_file dataset
real(r8) :: t_pert_ubc ! temperature pertubation to apply to the msis
! temperature at the upper boundary in Kelvin
! degrees; default value = 0._r8
real(r8) :: no_xfac_ubc ! nitrogen oxide concentration multiplicative
! factor to be applied to the SNOE no values
! at the upper boundary; default = 1._r8
! waccm solor proton event (SPE) data variables
logical :: spe_remove_file ! true => the offline spe file will be removed
logical :: spe_restart ! true => restart from a previous run with SPEs
character(len=256) :: spe_data_file ! name of file that contains the spe data
character(len=256) :: spe_filenames_list ! file that lists a series of spe files
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Brian Eaton, Stacy Walters, Francis Vitt
List all subroutines eliminated:
List all subroutines added and what they do:
A models/atm/cam/src/physics/waccm/qbo.F90
- new routine to allow for a forced qbo
A models/atm/cam/src/chemistry/waccm/spedata.F90
- new routine to allow for solar proton events
A models/atm/cam/src/chemistry/waccm/spehox.F90
- new routine to allow for hox production from solar proton events
A models/atm/cam/src/chemistry/waccm/mo_solarproton.F90
- new routine to allow for solar proton events
List all existing files that have been modified, and describe the changes:
M models/atm/cam/src/control/runtime_opts.F90
- added the above listed changes to the namelist
M models/atm/cam/src/physics/cam1/check_energy.F90
- changed default value of print_energy_errors from .false. to .true.
(Francis you might want to check with Brian about this)
M models/atm/cam/src/physics/cam1/physics_types.F90
- added real variable uzm, the zonally averaged u wind, for the qbo option
M models/atm/cam/src/physics/cam1/advnce.F90
- added use association and call to qbo_timestep_init for qbo initialization
on a time step basis
M models/atm/cam/src/physics/cam1/tphysac.F90
- added use association and call to qbo_relax for qbo
- added preprocessing tokens to differentiate call to ion_drag depending on
whether the simulation is waccm_mozart or waccm_ghg
M models/atm/cam/src/physics/cam1/nlte_fomichev.F90
- changed constant 1 to 1._r8
M models/atm/cam/src/physics/cam1/nlte_lw.F90
- changed longname for QRLNLTE in addfld call
- add addfld call for QNO; nitrogen oxide cooling rate
M models/atm/cam/src/physics/cam1/ctem.F90
- extensive changes to allow the "transformed eulerian mean" variables to
be computed and output properly with either a one or two dimensional
physical grid decomposition
M models/atm/cam/src/physics/waccm/gw_drag.F90
- added resolution cases for waccm gravity wave drag
M models/atm/cam/src/physics/waccm/tgcm_forcing.F90
- added preprocessing token to limit the diagnotic output
M models/atm/cam/src/chemistry/waccm_mozart/svdcmp.F90
- modification to svd routines to make them fortran90 compliant
M models/atm/cam/src/chemistry/waccm_mozart/mo_setext.F90
- added code to handle insitu production from solar proton events
M models/atm/cam/src/chemistry/waccm_mozart/mo_gas_phase_chemdr.F90
- corrected error in the cloud ice buffer index
M models/atm/cam/src/chemistry/waccm_mozart/mo_tgcm_ubc.F90
- restricted diagnostic output to the masterproc and with preprocessing tokens
M models/atm/cam/src/chemistry/waccm_mozart/upper_bc.F90
- added code to implement t_pert_ubc and no_xfac_ubc the temperature and
nitrogen oxide perturbations at the upper boundary. The resultant temperature
and nitrogen oxide concentration are checked for non-negativity
M models/atm/cam/src/chemistry/waccm_mozart/chem_surfvals.F90
- restricted diagnostic output to the masterproc and with preprocessing tokens
M models/atm/cam/src/chemistry/waccm_mozart/mo_cph.F90
- removed use of molecular nitrogen concentration in chemical heating rates
M models/atm/cam/src/chemistry/waccm_mozart/mo_chemini.F90
- added solar proton event initialization
M models/atm/cam/src/chemistry/waccm_mozart/mo_indprd.F90
- added solar proton event insitu forcing
M models/atm/cam/src/chemistry/waccm_mozart/mo_aurora.F90
- restricted diagnostic output to the masterproc and with preprocessing tokens
M models/atm/cam/src/chemistry/waccm_mozart/mo_imp_sol.F90
- cosmetic changes and removed removed reference to mpi process identifier "iam"
M models/atm/cam/src/chemistry/waccm_mozart/mo_snoe.F90
- restricted diagnostic output to the masterproc and with preprocessing tokens
M models/atm/cam/src/chemistry/waccm_mozart/euvac.F90
- restricted diagnostic output to the masterproc and with preprocessing tokens
M models/atm/cam/src/chemistry/waccm_mozart/chemistry.F90
- added call to advance_spedata to update the solar proton event forcing to present
time step
- added call to addfld for spe related production for ground and excited state
atomic nitrogen and oh
M models/atm/cam/src/chemistry/waccm_mozart/iondrag.F90
- extensive update to waccm_mozart chemistry case wherein ion and electron
concentrations are directly utilized
M models/atm/cam/src/chemistry/waccm_mozart/mo_setrxt.F90
- removed all vector function sections
M models/atm/cam/src/chemistry/waccm_mozart/efield.F90
- added use association for the svd routines
- restricted diagnostic output to the masterproc and with preprocessing tokens
M models/atm/cam/src/chemistry/waccm_mozart/chem_mods.F90
- new module consistent with solar proton event insitu forcing
M models/atm/cam/src/chemistry/waccm_mozart/mo_prod_loss.F90
- added code to handle solar proton insitu forcing
M models/atm/cam/src/chemistry/waccm_mozart/neckel.F90
- restricted diagnostic output to the masterproc and with preprocessing tokens
M models/atm/cam/src/chemistry/waccm_mozart/mo_srf_emissions.F90
- altered flux initialization from gas_pcnst to gas_pcnst-1
M models/atm/cam/src/dynamics/fv/pmgrid.F90
- added decomptype strip2dx for modified ctem.F90 routines
M models/atm/cam/src/dynamics/fv/dp_coupling.F90
- added dynamics block decomposition x and y index ranges to subroutine
definition of d_p_coupling
- modified code to handle both one and two dimensional grid decomposition
- modified code to properly interface with the modified ctem.F90 routines
M models/atm/cam/src/dynamics/fv/spmd_dyn.F90
- created decomptype strip2dx for modified ctem.F90 routines
M models/atm/cam/src/dynamics/fv/dynamics.F90
- created decomptype strip2dx for modified ctem.F90 routines
M models/atm/cam/src/dynamics/fv/stepon.F90
- added dynamics block decomposition x and y index ranges to calling sequence of
d_p_coupling
M models/atm/cam/src/dynamics/fv/restart_dynamics.F90
- added code to output variables needed for restarting with solar proton event
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista:
030 bl336 TBL.sh wm4h outfrq3s 9s .................................FAIL! rc= 7
- expected to fail due to WACCM changes
059 cs998 TCS.ccsm.sh ERS f19_g13 K ...............................FAIL! rc= 2
- this failed in cam3_3_47 and is not expected to pass in this version
tempest:
017 bl314 TBL.sh wg10dh outfrq3s 9s ...............................FAIL! rc= 7
- expected to fail due to WACCM changes
bangkok/lf95: All Pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_3_47
Originator(s): eaton
Date: Tue Jan 23 08:54:59 MST 2007
One-line Summary: Remove non-advected constituents option; misc bug fixes & enhancements.
Purpose of changes:
. Remove the "non-advected" constituents option. This option allowed
constituent transport by the physics parameterizations, but disabled
large-scale advective transports. It was used by the original prognostic
cloud water scheme, but is no longer needed.
. Add configurations for 1/4 and 1/2 degree FV grids. Add
dycore/resolution specific parameters in cldwat and cloud_fraction
modules.
. Fix interpolation code diagnostic that caused cam to quit if all columns
in a process fell into an "extrapolation region" of the input dataset.
The test only makes sense when interpolating on the global grid; not on
sub-domains.
. Fix bug in driver that was preventing the CAM banner from being printed.
. Modify build-namelist defaults for FV, 1.9x2.5. The default landfrac is
now consistent with the coupled model resolution 1.9x2.5_gx1v4.
. Modify build-namelist to fail with informative message if user specifies
ncdata but not start_ymd. This is because build-namelist has default
values for start_ymd which correspond to the default datasets used for
ncdata. If the user overrides the default ncdata, then the default
start_ymd should also be overridden since the default may be inconsistent
with the dataset specified for ncdata.
. add fix to XML::Lite parser code to recognize attributes when there is
optional white space between the equals sign and the value. Also allow
the attribute values to be enclosed in single quotes as well as double
quotes.
. modify build for a homme run
. Add some initializers for index values in FV (from Art Mirin).
. replace deprecated float() intrinsic with real().
. fix for spaces in the name of the spmdstats file (from Pat Worley).
. update mct external to trunk tag MCT2_3_0_070112
Bugs fixed (include bugzilla ID):
. Bugzilla bug 89:
Change to cd_core.F90 required to enable the ifort compiler to compile the
omp directive surrounding "do 4500" loop.
. Bugzilla bug 65:
Remove a informative print statement from inside an CSD region (caused
problem on Cray-X1E).
. add fix for intermittent problem Tony found running in coupled mode on
bassi (NERSC IBM p5). Add initializer for srfflx_state2d%qref.
. fix ctl_nl.pm (a homme namelist) so that it doesn't ignore the resolution
found in the config_cache.xml file. Remove the hardcoded value of "ne"
from Default_CTL_NL_Namelist.xml.
. NOTE -- initial commit and tag contained problem with svn external
definitions. Deleted that tag, fixed definitions, recommitted and
retagged.
Describe any changes made to build system: none
Describe any changes made to the namelist:
. Add checks to build-namelist to fail if ncdata is specified without
specifying start_ymd.
List any changes to the defaults for the boundary datasets:
. New CLM datasets for FV, 1.9x2.5. These datasets are consistent with the
landfraction used for CCSM runs (1.9x2.5_gx1v4).
Describe any substantial timing or memory changes: none
Code reviewed by: myself
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
. update mct external to trunk tag MCT2_3_0_070112
SVN_EXTERNAL_DIRECTORIES
. Mods to add 1/4 and 1/2 degree FV grids
models/atm/cam/bld/DefaultCAM_INPARM_Namelist.xml
models/atm/cam/bld/DefaultCLM_INPARM_Namelist.xml
models/atm/cam/bld/resolution_parameters.xml
models/atm/cam/src/physics/cam1/cldwat.F90
models/atm/cam/src/physics/cam1/cloud_fraction.F90
models/atm/cam/src/physics/cam1/hk_conv.F90
models/atm/cam/src/physics/cam1/zm_conv.F90
models/atm/cam/src/dynamics/fv/dycore.F90
models/atm/cam/src/dynamics/fv/dyn_comp.F90
. modify warning about extrapolating too much of grid
models/atm/cam/src/control/interpolate_data.F90
. bug fix for CAM banner in output log
models/drv/seq_mct/seq_ccsm_drv.F90
. Modify build-namelist defaults for FV, 1.9x2.5.
models/atm/cam/bld/cam_inparm.pm
models/atm/cam/bld/clm_inparm.pm
models/atm/cam/bld/DefaultCAM_INPARM_Namelist.xml
models/atm/cam/bld/DefaultCLM_INPARM_Namelist.xml
. add check to force user to also specify start_ymd if ncdata has been
specified.
models/atm/cam/bld/SeqCCSM_namelist.pm
. add fix to XML::Lite parser code to recognize attributes when there is
optional white space between the equals sign and the value. Also allow
the attribute values to be enclosed in single quotes as well as double
quotes.
models/atm/cam/bld/XML/Lite/Element.pm
. add missing initializer for srfflx_state2d%qref.
models/atm/cam/src/control/camsrfexch_types.F90
. mods for homme build
models/atm/cam/bld/config_cam_homme_defaults.xml
models/atm/cam/bld/ctl_nl.pm
models/atm/cam/bld/DefaultCAM_INPARM_Namelist.xml
models/atm/cam/bld/DefaultCTL_NL_Namelist.xml
models/atm/cam/bld/DefaultTIMEMGR_INPARM_Namelist.xml
. add some initializers for index values in FV (from Art Mirin).
models/atm/cam/src/dynamics/fv/d2a3dijk.F90
models/atm/cam/src/dynamics/fv/d2a3dikj.F90
models/atm/cam/src/dynamics/fv/spmd_dyn.F90
. Replace deprecated float() intrinsic with real():
models/atm/cam/src/physics/cam1/mcshallow.F90
models/atm/cam/src/physics/cam1/convect_ke.F
models/atm/cam/src/physics/cam1/uw_conv.F
. Bugzilla bug 89:
models/atm/cam/src/dynamics/fv/cd_core.F90
. Bugzilla bug 65
models/atm/cam/src/dynamics/fv/fill_module.F90
. Put shr_sys_flush calls inside "#ifndef UNICOSMP" as a workaround for a bug
on the Cray-X1E.
. fix spmdstats filename so that it doesn't contain spaces
models/atm/cam/src/control/cam_comp.F90
. Mods to remove the "non-advected" species option. Remove all occurances
of pnats and ppcnst, and change the cnst_add interface to remove the
'type' argument.
models/ocn/som/mixed_layer.F90
models/utils/pilgrim/mod_comm.F90
models/atm/cam/test/unit/control/file_test.pl
models/atm/cam/test/unit/ocnsice/dom/params.h
models/atm/cam/tools/scam/configure
models/atm/cam/tools/scam/scm_init/forecast.F90
models/atm/cam/tools/scam/obj/Makefile
models/atm/cam/bld/CAM_config.pm
models/atm/cam/src/control/ccsm_msg.F90
models/atm/cam/src/control/wtrc_flxoce.F90
models/atm/cam/src/control/srfxfer.F90
models/atm/cam/src/control/history_scam.F90
models/atm/cam/src/control/cam_history.F90
models/atm/cam/src/control/history_defaults.F90
models/atm/cam/src/control/camsrfexch_types.F90
models/atm/cam/src/control/ncdio_atm.F90
models/atm/cam/src/control/iop.F90
models/atm/cam/src/physics/cam1/tphysidl.F90
models/atm/cam/src/physics/cam1/tphysbc.F90
models/atm/cam/src/physics/cam1/drydep_mod.F90
models/atm/cam/src/physics/cam1/restart_physics.F90
models/atm/cam/src/physics/cam1/hk_conv.F90
models/atm/cam/src/physics/cam1/constituents.F90
models/atm/cam/src/physics/cam1/water_tracers.F90
models/atm/cam/src/physics/cam1/hb_diff.F90
models/atm/cam/src/physics/cam1/convect_shallow.F90
models/atm/cam/src/physics/cam1/convect_deep.F90
models/atm/cam/src/physics/cam1/buffer.F90
models/atm/cam/src/physics/cam1/zm_conv.F90
models/atm/cam/src/physics/cam1/qneg4.F90
models/atm/cam/src/physics/cam1/vertical_diffusion.F90
models/atm/cam/src/physics/cam1/initindx.F90
models/atm/cam/src/physics/cam1/seasalt_intr.F90
models/atm/cam/src/physics/cam1/tracers.F90
models/atm/cam/src/physics/cam1/progseasalts_intr.F90
models/atm/cam/src/physics/cam1/dust_intr.F90
models/atm/cam/src/physics/cam1/stratiform.F90
models/atm/cam/src/physics/cam1/carbon_intr.F90
models/atm/cam/src/physics/cam1/co2_cycle.F90
models/atm/cam/src/physics/cam1/chemistry.F90
models/atm/cam/src/physics/cam1/aerosol_intr.F90
models/atm/cam/src/physics/cam1/sulfur_intr.F90
models/atm/cam/src/physics/cam1/tracers.F90
models/atm/cam/src/physics/cam1/radiation.F90
models/atm/cam/src/physics/cam1/prescribed_aerosols.F90
models/atm/cam/src/physics/cam1/cam_diagnostics.F90
models/atm/cam/src/physics/cam1/upper_bc.F90
models/atm/cam/src/physics/cam1/physics_types.F90
models/atm/cam/src/physics/cam1/physpkg.F90
models/atm/cam/src/physics/cam1/constituent_burden.F90
models/atm/cam/src/physics/cam1/tphysac.F90
models/atm/cam/src/physics/waccm/tracers.F90
models/atm/cam/src/physics/waccm/chemistry.F90
models/atm/cam/src/physics/waccm/gw_drag.F90
models/atm/cam/src/physics/waccm/upper_bc.F90
models/atm/cam/src/physics/waccm/mo_msis_ubc.F90
models/atm/cam/src/chemistry/trop_mozart/chemistry.F90
models/atm/cam/src/chemistry/waccm_mozart/chemistry.F90
models/atm/cam/src/chemistry/waccm_mozart/upper_bc.F90
models/atm/cam/src/chemistry/waccm_mozart/mo_gas_phase_chemdr.F90
models/atm/cam/src/chemistry/waccm_mozart/mo_tgcm_ubc.F90
models/atm/cam/src/chemistry/waccm_mozart/chem_surfvals.F90
models/atm/cam/src/chemistry/waccm_mozart/mo_snoe.F90
models/atm/cam/src/chemistry/waccm_mozart/mo_waccm_hrates.F90
models/atm/cam/src/dynamics/sld/linemsdyn.F90
models/atm/cam/src/dynamics/sld/spegrd.F90
models/atm/cam/src/dynamics/sld/prognostics.F90
models/atm/cam/src/dynamics/sld/scanslt.F90
models/atm/cam/src/dynamics/sld/scan2.F90
models/atm/cam/src/dynamics/sld/restart_dynamics.F90
models/atm/cam/src/dynamics/sld/tfilt_massfix.F90
models/atm/cam/src/dynamics/sld/dp_coupling.F90
models/atm/cam/src/dynamics/sld/inidat.F90
models/atm/cam/src/dynamics/sld/dyn_comp.F90
models/atm/cam/src/dynamics/sld/spmd_dyn.F90
models/atm/cam/src/dynamics/sld/diag_dynvar_ic.F90
models/atm/cam/src/dynamics/eul/linemsdyn.F90
models/atm/cam/src/dynamics/eul/spegrd.F90
models/atm/cam/src/dynamics/eul/scanslt.F90
models/atm/cam/src/dynamics/eul/prognostics.F90
models/atm/cam/src/dynamics/eul/scandyn.F90
models/atm/cam/src/dynamics/eul/restart_dynamics.F90
models/atm/cam/src/dynamics/eul/tfilt_massfix.F90
models/atm/cam/src/dynamics/eul/diag_dynvar_ic.F90
models/atm/cam/src/dynamics/eul/dp_coupling.F90
models/atm/cam/src/dynamics/eul/dyn_comp.F90
models/atm/cam/src/dynamics/eul/inidat.F90
models/atm/cam/src/dynamics/eul/spmd_dyn.F90
models/atm/cam/src/dynamics/fv/stepon.F90
models/atm/cam/src/dynamics/fv/dp_coupling.F90
models/atm/cam/src/dynamics/fv/dyn_comp.F90
models/atm/cam/src/dynamics/fv/spmd_dyn.F90
models/atm/cam/src/dynamics/fv/inital.F90
models/atm/cam/src/dynamics/fv/restart_dynamics.F90
models/atm/cam/src/advection/slt/qmassa.F90
models/atm/cam/src/dynamics/homme/dp_coupling.F90
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: all PASS except:
035 bl375 TBL.sh f1.9h outfrq24h 2d ...............................FAIL! rc= 7
059 cs998 TCS.ccsm.sh ERS f19_g13 K ...............................FAIL! rc= 2
tempest: all PASS except
031 bl376 TBL.sh f1.9o outfrq24h 2d ...............................FAIL! rc= 7
bangkok/lf95: all PASS except:
031 bl377 TBL.sh f1.9m outfrq24h 2d ...............................FAIL! rc= 7
The failure of the FV, 1.9x2.5 baseline comparisons are expected due to a
change in the default landfrac file for that case.
The failure of the CCSM restart test is still under investigation. The
latest CCSM tag in collections (ccsm3_1_beta42) is also failing with
cam3_3_46 which means the failure is not associated with the mods in this
commit.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam3_3_46
Originator(s): mvr,jet,eaton,aconley
Date: 10 Jan 2007
One-line Summary: test_driver mods, consolidation of external
definitions in top directory, various bug fixes
Purpose of changes:
-having external definitions in multiple places led to some confusion
-previously had no pretag tests at fv 1.9x2.5, an important production res
-wanted some posttag testing to occur on calgary as alternative linux cluster
-testing of fv 2d decomposition needed bolstering and multiple platforms
-needed test of an extended run (several months) to catch certain run-time bugs
-added lightning test environment setting to override that of users
-branch test again branches from an earlier restart file than the last created
(broken since cam3_3_15)
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: myself,eaton
List all subroutines eliminated:
D models/utils/SVN_EXTERNAL_DIRECTORIES
D models/SVN_EXTERNAL_DIRECTORIES
- files no longer needed with consolidation of external defs to top dir
D models/atm/cam/test/system/config_files/f2h
D models/atm/cam/test/system/config_files/f2m
D models/atm/cam/test/system/config_files/f2o
- fv tests at "production" settings moved from fv2x2.5 to fv1.9x2.5
D models/atm/cam/test/system/tests_posttag_bangkok
- posttag testing of linux cluster moved from bangkok to calgary
D models/atm/cam/test/system/TRX.sh
- restart test while toggling fv 2d decomp settings merged with TER.sh
List all subroutines added and what they do:
A models/atm/cam/test/system/config_files/f4h
- new configuration options file for performance/extended-run test
A models/atm/cam/test/system/config_files/f1.9o
- new configuration options file for "production" setting on tempest
A models/atm/cam/test/system/nl_files/fv2d_4tsk
A models/atm/cam/test/system/nl_files/fv2d_8tsk
- new namelist options files for testing of fv 2d decomposition
A models/atm/cam/test/system/tests_posttag_calgary
- posttag testing of linux cluster moved from bangkok to calgary
A SVN_EXTERNAL_DIRECTORIES
- new, consolidated list of all cam external definitions
List all existing files that have been modified, and describe the changes:
M .
- set externals property on top level dir
M models
M models/utils
- deleted externals properties where they previously existed
M models/atm/cam/test/system/TBR.sh
M models/atm/cam/test/system/TSM.sh
- modified to handle new syntax for run length
M models/atm/cam/test/system/TER.sh
- modified to handle new syntax for run length; restart now toggles resources
appropriately if running fv 2d decompostion
M models/atm/cam/test/system/tests_pretag_tempest
- now running "production" test at fv 1.9x2.5
M models/atm/cam/test/system/tests_pretag_bangkok
M models/atm/cam/test/system/tests_posttag_lightning
- added test for fv 2d decomp and now running "production" test at fv 1.9x2.5
M models/atm/cam/test/system/tests_posttag_bluesky
M models/atm/cam/test/system/tests_pretag_bluesky
M models/atm/cam/test/system/tests_pretag_bluevista
- changed test for fv 2d decomp and now running "production" test at fv 1.9x2.5
M models/atm/cam/test/system/tests_posttag_bluevista
- added extended run (two months) and moved the performance test from 10d to 2m
M models/atm/cam/test/system/test_driver.sh
- modified settings for calgary; bug fix for lightning environment
M models/atm/cam/test/system/nl_files/pghgsul
- added namelist settings for freq of restart writes
M models/atm/cam/test/system/input_tests_master
- modified syntax for run length of tests - needed to specify something other
than timesteps or days; addition and deletion of tests mentioned
M models/atm/cam/test/system/CAM_runcmnd.sh
- modified setting for calgary
M models/atm/cam/src/control/getnetcdfdata.F90
- SCM bug fix for setlatlonidx routine when determining closest model column to
given lat/lon pair
M models/atm/cam/src/physics/cam1/aerosol_radiation_interface.F90
- fix to memory bug when calling radforce with omp enabled
M models/atm/cam/src/physics/cam1/phys_gmean.F90
- fix to memory leak introduced in cam3_3_35; whitespace cleanup
M models/atm/cam/src/physics/cam1/prescribed_aerosols.F90
- fixed bug in reading PS data from Match. Bug introduced in cam3_3_43
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista:
- all baseline tests failed due to the changes in the testing scripts and in
some cases the fixing of bugs - when a baseline was set up with the same
script changes and bug fixes, all baseline tests passed
tempest:
- all baseline tests failed due to the changes in the testing scripts and in
some cases the fixing of bugs - when a baseline was set up with the same
script changes and bug fixes, all baseline tests passed
bangkok/lf95:
- all baseline tests failed due to the changes in the testing scripts and in
some cases the fixing of bugs - when a baseline was set up with the same
script changes and bug fixes, all baseline tests passed
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_3_45
Originator(s): Dani Bundy Coleman, Jerry Olson
Date: 8 Jan 2007
One-line Summary: Alternative convection & PBL schemes, forecast mode
Purpose of changes: Allow users to run the UW PBL scheme or the Emanuel
deep convection scheme. Add rough code to run in forecast mode.
The UW PBL scheme requires:
configuration option: nlev = 30
namelist options: shallow_scheme = 'UW'
eddy_scheme = 'diag TKE'
The Emanuel deep convection requires:
namelist options: deep_scheme = 'KE'
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist:
character(len=16) :: shallow_scheme ! 'Hack' (default),'UW','off'
! 'Hack' = Hack shallow convection (default)
! 'UW' = original McCaa UW pbl scheme, modified by Sungsu Park
! 'off' = no shallow convection
!
character(len=16) :: deep_scheme ! 'ZM' (default), 'KE', 'off'
! 'ZM' = Zhang-McFarlane (default)
! 'KE' = Kerry Emanuel's scheme
! 'off' no deep convection
!
character(len=16) :: eddy_scheme ! 'HB' (default), 'diag TKE'
! 'HB' = Holtslag and Boville (default)
! 'diag TKE' = diagnostic tke version of Grenier and Bretherton
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: Brian Eaton
List all subroutines eliminated: none
List all subroutines added and what they do:
A models/atm/cam/src/utils/error_function.F90
provides an erfc function for the UW PBL scheme
A models/atm/cam/src/physics/cam1/phys_control.F90
Control interface to CAM physics packages
A models/atm/cam/src/physics/cam1/mcshallow.F90
UW PBL shallow convection package
A models/atm/cam/src/physics/cam1/convect_ke_intr.F90
Interface to Kerry Emanuel's deep convection scheme
A models/atm/cam/src/physics/cam1/convect_ke.F
Kerry Emanuel's deep convection scheme
A models/atm/cam/src/physics/cam1/zm_conv_intr.F90
Interface to Zhang-McFarlane deep convection scheme
(used to be convect_deep.F90)
A models/atm/cam/src/physics/cam1/eddy_diff.F90
UW PBL eddy diffusivity package
A models/atm/cam/src/physics/cam1/uw_conv.F90
UW PBL shallow convection package
List all existing files that have been modified, and describe the changes:
Forecast mode changes:
M models/atm/cam/src/control/cam_history.F90
M models/atm/cam/src/physics/cam1/check_energy.F90
M models/atm/cam/src/physics/cam1/cam_diagnostics.F90
M models/atm/cam/src/dynamics/sld/spegrd.F90
M models/atm/cam/src/dynamics/sld/linemsdyn.F90
M models/atm/cam/src/dynamics/sld/scanslt.F90
M models/atm/cam/src/dynamics/sld/dyn_comp.F90
M models/atm/cam/src/dynamics/sld/inidat.F90
M models/atm/cam/src/dynamics/sld/grmult.F90
M models/atm/cam/src/dynamics/sld/tfilt_massfix.F90
M models/atm/cam/src/dynamics/eul/linemsdyn.F90
M models/atm/cam/src/dynamics/eul/inidat.F90
M models/atm/cam/src/dynamics/eul/dyn_comp.F90
M models/atm/cam/src/dynamics/eul/tfilt_massfix.F90
M models/atm/cam/src/dynamics/fv/initcom.F90
M models/atm/cam/src/dynamics/fv/inidat.F90
M models/atm/cam/src/dynamics/fv/dyn_comp.F90
M models/atm/cam/src/dynamics/fv/commap.F90
M models/atm/cam/src/dynamics/fv/diag_dynvar_ic.F90
M models/atm/cam/src/dynamics/fv/uv3s_update.F90
M models/atm/cam/src/dynamics/fv/stepon.F90
Convection option changes:
M models/atm/cam/src/control/runtime_opts.F90
M models/atm/cam/src/physics/cam1/initindx.F90
M models/atm/cam/src/physics/cam1/param_cldoptics.F90
M models/atm/cam/src/physics/cam1/physpkg.F90
M models/atm/cam/src/physics/cam1/cloud_fraction.F90
M models/atm/cam/src/physics/cam1/gw_drag.F90
M models/atm/cam/src/physics/cam1/phys_buffer.F90
M models/atm/cam/src/physics/cam1/zm_conv.F90
M models/atm/cam/src/physics/cam1/cldwat.F90
M models/atm/cam/src/physics/cam1/vertical_diffusion.F90
M models/atm/cam/src/physics/cam1/convect_shallow.F90
This is now the interface to all shallow convection schemes
M models/atm/cam/src/physics/cam1/convect_deep.F90
This is now the interface to all deep convection schemes
Convection and Forecast-mode changes:
M models/atm/cam/src/physics/cam1/tphysbc.F90
M models/atm/cam/src/physics/cam1/stratiform.F90
M models/atm/cam/src/physics/cam1/constituents.F90
M models/atm/cam/src/physics/cam1/tphysac.F90
Miscellaneous changes:
M models/atm/cam/test/system/TBL.sh
Add more information to output
M models/atm/cam/test/system/test_driver.sh
Change bluevista gmake option from "-j" to "-j8"
M models/atm/cam/src/physics/cam1/acbnd.F90
Check that oxidant file has correct vertical dimensions
M models/atm/cam/src/physics/cam1/radiation.F90
M models/atm/cam/src/physics/cam1/radsw.F90
M models/atm/cam/src/physics/cam1/radlw.F90
Calculate clearsky solor & longwave heating rates
New output fields: QRSC, QRSC_RF, QRLC
M models/atm/cam/src/physics/cam1/prescribed_aerosols.F90
stdout
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista:
Some code was moved from timestepping routines to initialization
routines. In cldwat.F90, this causes a roundoff change
on bluevista that causes the following baseline tests to fail:
005 bl131 TBL.sh e32c11dh pghgsul 9
008 bl132 TBL.sh e32sdh ghgrmp 9
015 bl151 TBL.sh e64h outfrq24h -2
019 bl331 TBL.sh f4c11gdh pghgsul 9
022 bl332 TBL.sh f4sdh ghgrmp 9
030 bl336 TBL.sh wm4h outfrq3s 9
034 bl351 TBL.sh f2h outfrq24h -2
037 bl354 TBL.sh fm2dh outfrq3s 9
040 bl355 TBL.sh fmo2dh off2x2.5 9
054 bl551 TBL.sh s64h outfrq24h -2
tempest:
The roundoff difference as seen on bluevista seems to
show up in a waccm test that fails:
017 bl314 TBL.sh wg10dh outfrq3s 9
The baseline model field DTCOND included the effects of the energy
fixer (which is actually a fix to the nonconservation of the
dynamics). We moved the call to diag_conv_tend_ini from its position
above the energy fixer to below it, which makes the field different
and allows the heat budget to work out. This causes the following tests to
fail (only DTCOND is different):
021 bl331 TBL.sh f4c11gdh pghgsul 9
024 bl332 TBL.sh f4sdh ghgrmp 9
031 bl352 TBL.sh f2o outfrq24h -2
bangkok/lf95:
Failed because of fixed DTCOND (only DTCOND is different)
016 bl311 TBL.sh f10c8mdm ttrac 9
024 bl312 TBL.sh f10sdm ghgrmp 9
028 bl353 TBL.sh f2m outfrq24h -2
CAM tag used for the baseline comparison tests if different than previous
tag: cam3_3_44
Summarize any changes to answers, i.e.,
- what code configurations:
Any configuration on bluevista, although the sulfur cycle is particularly sensitive.
WACCM on tempest was also sensitive to the roundoff change
- what platforms/compilers:
bluevista, tempest
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
roundoff
If bitwise differences were observed, how did you show they were no worse
than roundoff?
pergro test, tracking roundoff change to culprit code
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_3_44
Originator(s): Jim Edwards
Date: 12-13-2006
One-line Summary: Update of homme dynamics, update of tests and timing utils
Purpose of changes: Merge homme dynamics development branch.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Added a default homme core build
M 2606 models/atm/cam/bld/config_cam_homme_defaults.xml
M 2606 models/atm/cam/bld/DefaultCAM_INPARM_Namelist.xml
M 2606 models/atm/cam/bld/DefaultCTL_NL_Namelist.xml
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: myself Brian Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
updated for new timings tag:
models/utils/SVN_EXTERNAL_DIRECTORIES
added a homme test and updated the CCSM test:
A 0 models/atm/cam/test/system/config_files/h5x8dm
M 2606 models/atm/cam/test/system/TCT.ccsm.sh
M 2606 models/atm/cam/test/system/tests_pretag_bluevista
M 2606 models/atm/cam/test/system/input_tests_master
ccsm tests will now pull in share code from tester's sandbox by default...
added environment variable (CAM_USE_SHR) to control behavior
Added support for homme grid in energy calculation
M 2606 models/atm/cam/src/physics/cam1/phys_gmean.F90
Replaced some data statements with parameter declarations
M 2606 models/atm/cam/src/physics/cam1/radae.F90
Removed some debugging code
M 2606 models/atm/cam/src/physics/cam1/physpkg.F90
M 2606 models/atm/cam/src/physics/cam1/qneg3.F90
Removed some timers that were causing a problem on bluevista -
this is a hack, it looks like there is a memory problem associated
with fmo2dh - I will continue to investigate.
models/atm/cam/src/dynamics/fv/dyn_comp.F90
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: 056 bl711 TBL.sh h5x8dm adia 9 FAIL (no existing baseline)
058 cs998 TCS.ccsm.sh ERS f19_g13 K
040 bl355 TBL.sh fmo2dh off2x2.5 9
FAILED against cam3_3_43 PASS against cam3_3_42 this is due to
problems introduced in 3_3_42
tempest: none
bangkok/lf95: none
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_3_43
Originator(s): John Truesdale
Date: 4 Dec 2006
One-line Summary: Added single column mode to CAM
Purpose of changes:
This commit adds a single column mode (SCM) to CAM. The scam
GUI is being refactored and has been removed from the
trunk. The CAM single column mode is controlled via the
standard cam namelist. Additional namelist parameters allow
cam to be forced using a standard IOP dataset. Future commits
will add back much of the original scam and column radiation
capablity as well as an updated GUI.
To run cam in single column mode you need to invoke the standard
cam configure script with the -scam option and compile the model
as usual. Although single column mode only runs with the eularian
dynamics most other options for configuring cam (resolution, external
libraries, etc) will work. Next invoking the build-namelist
script will provide a namelist template for running with the
default ARM IOP. The build-namelist procedure requires the user
to supply namelist values for the latitude and longitude of the
IOP column using the namelist variables scmlat and scmlon.
In practice though the user must also specify the starting date
and timesteps to run as well as an appropriate initial condition.
The following parameters to the build-namelist script may be used for
running with the various IOP datasets. The initial data being read
in and used as a background for each of the IOPs is the standard
SEP 1 initial condition. It is suggested that the user generate an
initial condition closer to the starting date of the IOP.
Once the namelist is generated the user may run cam and analyze the
standard history tape produced for the IOP column with any netcdf
visualization tool.
The relavent namelist parameters for use with the single column mode
of cam are:
single_column = .true.
scmlat = xxx ! scm latitude to use for the model run
scmlon = xxx ! scm longitude to use for the model run
iopfile = 'xxx' ! iop boundary condition file
build-namelist script params for the ARM IOP
build-namelist -namelist \
"&camexp \
mss_irt=0 \
scmlat=36.6 \
scmlon=262.5 \
start_ymd=19950718 \
start_tod=19800 \
stop_n=1259 \
stop_option='nsteps' \
iopfile='/fs/cgd/csm/inputdata/atm/cam/scam/iop/arm0795v1.2.nc' \
ncdata ='/fs/cgd/csm/inputdata/atm/cam/inic/gaus/cami_0000-09-01_64x128_L26_c030918.nc' \
mfilt=1400 \
nhtfrq=1/"
build-namelist script params for the Gate IOP
build-namelist -namelist \
"&camexp \
mss_irt=0 \
scmlat=9. \
scmlon=336. \
start_ymd=19740830 \
start_tod=0 \
stop_n=1440 \
stop_option='nsteps' \
iopfile='/fs/cgd/csm/inputdata/atm/cam/scam/iop/gate0874v1.2.nc' \
ncdata ='/fs/cgd/csm/inputdata/atm/cam/inic/gaus/cami_0000-09-01_64x128_L26_c030918.nc' \
mfilt=1500 \
nhtfrq=1/"
build-namelist script params for GCSS IOP
build-namelist -namelist \
"&camexp \
mss_irt=0 \
scmlat=-2. \
scmlon=155. \
start_ymd=19921220 \
start_tod=0 \
stop_n=505 \
stop_option='nsteps' \
iopfile='/fs/cgd/csm/inputdata/atm/cam/scam/iop/gcss1292v1.2.nc' \
ncdata ='/fs/cgd/csm/inputdata/atm/cam/inic/gaus/cami_0000-09-01_64x128_L26_c030918.nc' \
mfilt=1500 \
nhtfrq=1/"
In addition to the standard IOP datasets the single column version
of CAM may be forced with an intial condition file and a "pseudo
IOP" history file generated from a previous global CAM run. This
mode is useful for getting a feel for how new physics modules
behave given forcing from an actual CAM run. To generate an
initial condition and IOP style forcing data from a standard cam
run configure the model using the new -camiop option of the
configure script. The build-namelist script has been modified to
generate a namelist which will create an initial condition for a
scm model run as well as an auxillary history tape containing the
IOP fields needed to drive a single column version of CAM. Both
the initial conditions file and the IOP forcing history tape are 3
dimensional. When running with these datasets in SCM mode the
user can choose which column will be used via the namelist
parameters scmlat and scmlon.
The relavent namelist parameters for generating an CAMIOP are
fincl2 = 'Ps','u','v','t','q','omega','phis','Prec','lhflx','shflx',\
'Tsair','Tg','divq3d','divT3d','dcldice','dcldliq','beta',\
'fixmas','afixq','afixcldl','afixcldi','dqfxq','dqfxcldl',\
'dqfxcldi','CLAT','T','Q'
inithist = 'CAMIOP'
mfilt = 1,10
nhtfrq = 0,1
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: Added -scam and -camiop options
to configure script. The -scam option will configure cam for
single column mode. The -camiop option will configure cam to produce
an iop initial condition file and a boundary forcing history tape
to be used in single column mode.
Describe any changes made to the namelist: Added namelist parameters for
single column mode. Since all surface models need single column
control information the following namelist variables were added
to ccsm_inparm
! enable single_column mode
logical :: single_column ! default false set to true to
! enable single column mode
! single column longitude
real(r8) :: scmlon ! longitude setting used by cam in
! scm mode
! single column latitude
real(r8) :: scmlat ! latitude setting used by cam in
! scm mode
The following namelist variable was added to the cam_inparm namelist
! iop boundary file
integer, parameter :: max_chars = 128
character(len=max_chars) iopfile !IOP Boundary condition file
Added the CAMIOP option to the inithist namelist variable. This
option will produce an initial condition for use with a cam
generated iop forcing history tape.
! inithist Generate initial dataset as auxillary history file
! can be set to '6-HOURLY', 'DAILY', 'MONTHLY',
! 'YEARLY', 'CAMIOP', or 'NONE'.
! default: 'YEARLY'
List any changes to the defaults for the boundary datasets:
Added a namelist option to specify an IOP file for SCM mode.
Describe any substantial timing or memory changes:
Code reviewed by: Brian Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
configuration/namelist/script files for testing of scm cam mode:
models/atm/cam/test/system/config_files/scm64bfbiop
models/atm/cam/test/system/config_files/e64bfbiop
models/atm/cam/test/system/nl_files/scm_prep
models/atm/cam/test/system/nl_files/scm_b4b_o1
models/atm/cam/test/system/TSC.sh
Files adding scm capability to cam:
models/atm/cam/src/control/scamMod.F90
models/atm/cam/src/control/getnetcdfdata.F90
models/atm/cam/src/dynamics/eul/forecast.F90
models/atm/cam/src/dynamics/eul/getinterpnetcdfdata.F90
models/lnd/clm2/src/main/getnetcdfdata.F90
List all existing files that have been modified, and describe the changes:
The majority of the modifications had to do with removing the
#define SCAM sections of code. There is still some duplication of
functionality that will be cleaned up in future commits (reading of
netcdf data). The pieces of independent single column mode code
are delineated now by a run time 'if (single_column)' test. The single_column
logical is defined in the ccsm_inparm namelist and is available to all component
models. Files in the test and bld subdirectories were modified to allow
bit for bit testing of scam against a control run of the 3d model as well
as adding functionality to the build-namelist and configure scripts to make
running in single column mode easier.
models/ocn/dom/sst_data.F90
models/ice/csim4/ice_data.F90
models/atm/cam/test/system/tests_pretag_bangkok
models/atm/cam/test/system/TPF.sh
TPF.sh: the script for testing performance problems was modified to
reflect the move to a single timing file introduced in cam3_3_42
models/atm/cam/test/system/tests_pretag_bluevista
models/atm/cam/test/system/nl_files/scm_prep
models/atm/cam/test/system/nl_files/scm_b4b_o1
models/atm/cam/test/system/input_tests_master
models/atm/cam/test/system/TSC.sh
models/atm/cam/test/system/TCT.sh
TCT.sh: all ccsm tests through cam's test suite will now use the
csm_share and clm code from the cam sandboxes rather than that
from the latest ccsm tag (up to now it used just the cam code)
models/atm/cam/test/system/tests_posttag_lightning
models/atm/cam/bld/configure
models/atm/cam/bld/DefaultCAM_INPARM_Namelist.xml
models/atm/cam/bld/cam_inparm.pm
models/atm/cam/bld/NamelistsDescriptions.xml
models/atm/cam/bld/resolution_parameters.xml
models/atm/cam/bld/CAM_config.pm
models/atm/cam/bld/ccsm_inparm.pm
models/atm/cam/src/control/readinitial.F90
models/atm/cam/src/control/history_defaults.F90
models/atm/cam/src/control/runtime_opts.F90
models/atm/cam/src/control/ncdio_atm.F90
models/atm/cam/src/control/rgrid.F90
models/atm/cam/src/control/interpolate_data.F90
models/atm/cam/src/control/con_cam.F90
models/atm/cam/src/control/atm_comp_mct.F90
models/atm/cam/src/control/startup_initialconds.F90
models/atm/cam/src/control/history_scam.F90
models/atm/cam/src/control/cam_comp.F90
models/atm/cam/src/control/cam_history.F90
models/atm/cam/src/physics/cam1/initindx.F90
models/atm/cam/src/physics/cam1/radiation.F90
models/atm/cam/src/physics/cam1/radsw.F90
models/atm/cam/src/physics/cam1/cam_diagnostics.F90
models/atm/cam/src/physics/cam1/physics_types.F90
models/atm/cam/src/physics/cam1/radlw.F90
models/atm/cam/src/physics/cam1/physpkg.F90
models/atm/cam/src/physics/cam1/prescribed_aerosols.F90
models/atm/cam/src/physics/cam1/aerosol_index.F90
models/atm/cam/src/physics/cam1/gw_drag.F90
models/atm/cam/src/chemistry/trop_mozart/mz_aerosols_intr.F90
models/atm/cam/src/dynamics/sld/dyn_comp.F90
models/atm/cam/src/dynamics/eul/sphdep.F90
models/atm/cam/src/dynamics/eul/initcom.F90
models/atm/cam/src/dynamics/eul/scanslt.F90
models/atm/cam/src/dynamics/eul/inidat.F90
models/atm/cam/src/dynamics/eul/getinterpnetcdfdata.F90
models/atm/cam/src/dynamics/eul/inital.F90
models/atm/cam/src/dynamics/eul/dycore.F90
models/atm/cam/src/dynamics/eul/iop.F90
models/atm/cam/src/dynamics/eul/dynpkg.F90
models/atm/cam/src/dynamics/eul/diag_dynvar_ic.F90
models/atm/cam/src/dynamics/eul/scan2.F90
models/atm/cam/src/dynamics/eul/stepon.F90
models/atm/cam/src/dynamics/eul/restart_dynamics.F90
models/atm/cam/src/dynamics/eul/tfilt_massfix.F90
models/SVN_EXTERNAL_DIRECTORIES
models/drv/seq_mct_drv/seq_ccsm_drv.F90
models/drv/seq_mct/mrg_x2a_mct.F90
models/csm_share/shr/shr_inputinfo_mod.F90
models/lnd/clm2/src/main/inicFileMod.F90
models/lnd/clm2/src/main/driver.F90
models/lnd/clm2/src/main/ncdio.F90
models/lnd/clm2/src/main/initializeMod.F90
models/lnd/clm2/src/main/iniTimeConst.F90
models/lnd/clm2/src/main/restFileMod.F90
models/lnd/clm2/src/main/controlMod.F90
models/lnd/clm2/src/main/lnd_comp_mct.F90
models/lnd/clm2/src/main/surfrdMod.F90
models/lnd/clm2/src/biogeochem/STATICEcosysDynMod.F90
models/lnd/clm2/src//main/clm_varctl.F90
The following files were deleted. As stated above
the GUI version of SCAM is being refactored. An
impoved GUI will be added in the future.
scam
scam/configure
scam/crm
scam/GNUmakefile
scam/.scam_defaults
scam/README
scam/tools
scam/tools/sdev
scam/tools/sdev/sdev
scam/tools/sdev/sdev.C
scam/tools/sdev/Makefile
scam/tools/sdev/README
scam/tools/ncmult
scam/tools/ncmult/ncmult.C
scam/tools/ncmult/Makefile
scam/tools/ncmult/README
scam/tools/nctrans
scam/tools/nctrans/nctrans
scam/tools/nctrans/nctrans.C
scam/tools/nctrans/Makefile
scam/tools/nctrans/README
scam/tools/diurnal_ave
scam/tools/diurnal_ave/diurnal_ave
scam/tools/diurnal_ave/diurnal_ave.C
scam/tools/diurnal_ave/Makefile
scam/tools/diurnal_ave/README
scam/tools/intercomparison-post-processing
scam/tools/intercomparison-post-processing/ncfile.C
scam/tools/intercomparison-post-processing/field.C
scam/tools/intercomparison-post-processing/post
scam/tools/intercomparison-post-processing/ncfile.H
scam/tools/intercomparison-post-processing/post.C
scam/tools/intercomparison-post-processing/field.H
scam/tools/intercomparison-post-processing/intercompare.scm
scam/tools/intercomparison-post-processing/Makefile
scam/tools/intercomparison-post-processing/process
scam/tools/ncadd
scam/tools/ncadd/ncadd
scam/tools/ncadd/ncadd.C
scam/tools/ncadd/Makefile
scam/tools/ncadd/README
scam/tools/pdf
scam/tools/pdf/ferret.jnl
scam/tools/pdf/pdf
scam/tools/pdf/pdf.C
scam/tools/pdf/Makefile
scam/tools/pdf/README
scam/tools/ccm2iop
scam/tools/ccm2iop/ncfile.C
scam/tools/ccm2iop/ncfile.cpp
scam/tools/ccm2iop/ioerr.h
scam/tools/ccm2iop/ncfile.h
scam/tools/ccm2iop/Makefile
scam/tools/ccm2iop/ccm2iop.C
scam/tools/ccm2iop/test.nc
scam/getlatlon.pl
scam/testscript
scam/html
scam/html/userguide.html
scam/html/tdiff-cam2.gif
scam/html/gif
scam/html/gif/tdiff-cam2.gif
scam/html/gif/plot.gif
scam/html/gif/tdiff3.2.gif
scam/html/gif/tdiff3.6.gif
scam/html/gif/post.gif
scam/html/gif/formula1.gif
scam/html/gif/formula2.gif
scam/html/gif/iop.gif
scam/html/gif/formula3.gif
scam/html/gif/formula4.gif
scam/html/gif/qdiff.gif
scam/html/gif/formula5.gif
scam/html/gif/formula6.gif
scam/html/gif/tdiff.gif
scam/html/gif/mainwnd.gif
scam/html/gif/qls.gif
scam/html/gif/global.gif
scam/html/gif/sls.gif
scam/html/gif/mainwin.gif
scam/html/gif/uls.gif
scam/html/gif/vls.gif
scam/html/gif/w.gif
scam/html/gif/option.gif
scam/html/gif/g2.gif
scam/html/gif/uphys.gif
scam/html/gif/selectdata.gif
scam/html/gif/vphys.gif
scam/html/tdiff3.6.gif
scam/html/index.html
scam/scm_init
scam/scm_init/init_model.F90
scam/scm_init/misc.h
scam/scm_init/scam_srfdata.F90
scam/scm_init/drand48_.c
scam/scm_init/get_levels.F90
scam/scm_init/scam_srfdata_MCT.F90
scam/scm_init/runtype.h
scam/scm_init/scam_rpc.x
scam/scm_init/calcdate.F90
scam/scm_init/comfrc.h
scam/scm_init/fortran.h
scam/scm_init/ipc.h
scam/scm_init/setlatlon.F90
scam/scm_init/forecast.F90
scam/scm_init/scam_run.F90
scam/scm_init/myhandler.F
scam/scm_init/realtype.h
scam/scm_init/inital.F90
scam/scm_init/readpressdata.F90
scam/scm_init/c_outfld.c
scam/scm_init/scamMod.F90
scam/scm_init/c_outfld.h
scam/scm_init/interplevs.F90
scam/scm_init/getnetcdfdata.F90
scam/scm_init/preproc.h
scam/scm_init/setiopupdate.F90
scam/scm_init/scam.c
scam/scm_init/scam_fifo.c
scam/scm_init/getinterpncdata.F90
scam/scm_init/readiopdata.F90
scam/scm_init/readsaveinit.F90
scam/scm_init/max.h
scam/scm_init/scam_fifo.h
scam/INSTALL
scam/mymods
scam/mymods/README
scam/userdata
scam/userdata/crmtest18.in
scam/userdata/crmtest26.out
scam/userdata/crmdarwintest26.out
scam/userdata/crmtest200.in
scam/userdata/README
scam/userdata/crmtest26.in
scam/data
scam/data/README
scam/obj
scam/obj/makdep.linux
scam/obj/makdep
scam/obj/Makefile
scam/ui
scam/ui/timeconvert.cpp
scam/ui/PostPlottingDlg.h
scam/ui/OptionsDlg.ui.h
scam/ui/LoadData.ui
scam/ui/utils.h
scam/ui/history.h
scam/ui/IOPSelectDateDlg.ui
scam/ui/PlistDlgImpl.h
scam/ui/rpcmodel.h
scam/ui/defaults.h
scam/ui/map.cpp
scam/ui/PlistDlg.ui
scam/ui/datasetselector.cw
scam/ui/plot.cpp
scam/ui/manager.h
scam/ui/map.xbm
scam/ui/msgdlg.h
scam/ui/field.cpp
scam/ui/globalmap.cw
scam/ui/crm.h
scam/ui/fieldlistbox.cpp
scam/ui/dataset.h
scam/ui/Platform.Notes
scam/ui/ncarg.cpp
scam/ui/observer.h
scam/ui/IOPSelectDateDlg.ui.h
scam/ui/PostPlottingDlg.cpp
scam/ui/IOPSelectDateDlgImpl.cpp
scam/ui/ncfile.cpp
scam/ui/PlotDlg.ui
scam/ui/LoadDataImpl.h
scam/ui/history.cpp
scam/ui/ascii_dataset.h
scam/ui/globalmap.h
scam/ui/fifomodel.cpp
scam/ui/rpcmodel.cpp
scam/ui/plot.cw
scam/ui/GNUmakefile
scam/ui/defaults.cpp
scam/ui/ncarg
scam/ui/ncarg/c.h
scam/ui/SelectGlobalDataDlg.ui
scam/ui/msgdlg.cpp
scam/ui/sicfile.cpp
scam/ui/list.h
scam/ui/crm.cpp
scam/ui/dbgmodel.h
scam/ui/plot.h
scam/ui/PostPlottingDlgImpl.h
scam/ui/dataset.cpp
scam/ui/MainWnd.ui
scam/ui/world.xbm
scam/ui/LoadDataImpl.cpp
scam/ui/PlotDlgImpl.h
scam/ui/fieldlistbox.h
scam/ui/ncarg.h
scam/ui/model.cpp
scam/ui/configure
scam/ui/IOPSelectDateDlgImpl.h
scam/ui/map.cw
scam/ui/OptionsDlgImpl.cpp
scam/ui/ChangeAxisScaleDlg.ui
scam/ui/ncfile.h
scam/ui/SelectGlobalDataDlgImpl.h
scam/ui/fifomodel.h
scam/ui/ncarg_stubs.cpp
scam/ui/fieldplotwidget.cw
scam/ui/dbgmodel.cpp
scam/ui/PostPlottingDlgImpl.cpp
scam/ui/fortran.h
scam/ui/numlined.h
scam/ui/sicfile.h
scam/ui/images
scam/ui/images/editcopy
scam/ui/images/editcut
scam/ui/images/fileopen
scam/ui/images/editpaste
scam/ui/images/textbold
scam/ui/images/undo
scam/ui/images/searchfind
scam/ui/images/filenew
scam/ui/images/redo
scam/ui/images/print
scam/ui/images/filesave
scam/ui/main.cpp
scam/ui/MainWndImpl.h
scam/ui/ioerr.h
scam/ui/PlotDlgImpl.cpp
scam/ui/ld.man
scam/ui/ChangeAxisScaleDlgImpl.h
scam/ui/utils.cpp
scam/ui/SelectGlobalDataDlgImpl.cpp
scam/ui/configure.works
scam/ui/PlistDlgImpl.cpp
scam/ui/timeconvert.h
scam/ui/model.h
scam/ui/OptionsDlgImpl.h
scam/ui/manager.cpp
scam/ui/OptionsDlg.ui
scam/ui/fieldlistbox.cw
scam/ui/MainWndImpl.cpp
scam/ui/c.h
scam/ui/field.h
scam/ui/sst
scam/ui/max.h
scam/ui/ChangeAxisScaleDlgImpl.cpp
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista:
The following test failed as part of the cam3_3_42 commit and also fail as part
of this commit.
038 sm355 TSM.sh fmo2dh off2x2.5 9 ................................FAIL! rc= 8
039 er355 TER.sh fmo2dh off2x2.5 4+5 ..............................FAIL! rc= 5
040 bl355 TBL.sh fmo2dh off2x2.5 9 ................................FAIL! rc= 4
All ccsm tests through cam's test suite will now use the
csm_share and clm code from the cam sandboxes rather than that
from the latest ccsm tag (up to now it used just the cam code)
tempest:
bangkok/lf95:
CAM tag used for the baseline comparison tests if different than previous
tag:cam3_3_42
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
NONE
If bitwise differences were observed, how did you show they were no worse
than roundoff?
NA
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
NA
MSS location of control simulations used to validate new climate:
NA
URL for AMWG diagnostics output used to validate new climate:
===============================================================
Tag name: cam3_3_42
Originator(s): Patrick Worley
Date: 24 Nov 2006
One-line Summary: refactored timer instrumentation and support for single timing output file
Purpose of changes:
To update timing instrumentation to reflect recent model changes,
and to improve 'coverage' by adding timers for all top level events.
To improve usability on large numbers of processes by decreasing number
of process count-dependent files and lines of output:
- added option to generate single timing and spmdstats files
(containing all of the information that had been spread between
multiple files before)
- added option to restrict reporting of many runtime options and
memory usage statistics to process 0
These new options are enabled by default
Bugs fixed (include bugzilla ID):
'debugging' shr_sys_flush call in seq_domain_mct.F90 erroneously left in
during earlier check-in caused error on Cray X1E
Describe any changes made to build system:
Added new CPP flags:
- MULTIPLE_PERF_FILES: if defined, timing and output files will be
created for each process (the previous default)
- WRAP_MPI_TIMING: if defined, timing events are defined for each MPI
call in wrap_mpi.F90 (the previous default)
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Patrick Worley
List all subroutines eliminated:
List all subroutines added and what they do:
t_pr_onef in utils/timing/f_wrappers.c:
variant of t_prf that write all timing data in one file
t_enablef in utils/timing/f_wrappers.c,
GPTL_enable in utils/timing/gptl.c:
runtime timing routine enabler
t_disablef in utils/timing/f_wrappers.c,
GPTL_disable in utils/timing/gptl.c:
runtime timing routine disabler
List all existing files that have been modified, and describe the changes:
SVN_EXTERNAL_DIRECTORIES
using clm3_expa_79
utils/SVN_EXTERNAL_DIRECTORIES
using timing_061124
atm/cam/src/control/atm_comp_mct.F90
atm/cam/src/control/cam_comp.F90
atm/cam/src/control/wrap_mpi.F90
atm/cam/src/dynamics/eul/dp_coupling.F90
atm/cam/src/dynamics/eul/dyndrv.F90
atm/cam/src/dynamics/eul/dynpkg.F90
atm/cam/src/dynamics/eul/scan2.F90
atm/cam/src/dynamics/eul/scandyn.F90
atm/cam/src/dynamics/eul/scanslt.F90
atm/cam/src/dynamics/eul/stepon.F90
atm/cam/src/dynamics/fv/cd_core.F90
atm/cam/src/dynamics/fv/dp_coupling.F90
atm/cam/src/dynamics/fv/dyn_comp.F90
atm/cam/src/dynamics/fv/mapz_module.F90
atm/cam/src/dynamics/fv/stepon.F90
atm/cam/src/dynamics/fv/trac2d.F90
atm/cam/src/dynamics/sld/dp_coupling.F90
atm/cam/src/dynamics/sld/dyndrv.F90
atm/cam/src/dynamics/sld/dynpkg.F90
atm/cam/src/dynamics/sld/scan2.F90
atm/cam/src/dynamics/sld/scandyn.F90
atm/cam/src/dynamics/sld/scanslt.F90
atm/cam/src/dynamics/sld/stepon.F90
atm/cam/src/physics/cam1/convect_deep.F90
atm/cam/src/physics/cam1/convect_shallow.F90
atm/cam/src/physics/cam1/diffusion_solver.F90
atm/cam/src/physics/cam1/physpkg.F90
atm/cam/src/physics/cam1/radiation.F90
atm/cam/src/physics/cam1/tphysac.F90
atm/cam/src/physics/cam1/tphysbc.F90
atm/cam/src/physics/cam1/vertical_diffusion.F90
atm/cam/src/physics/cam1/zm_conv.F90
drv/seq_mct_drv/seq_ccsm_drv.F90
ice/csim4/ice_comp.F9
ice/csim4/ice_comp_mct.F90
ice/csim4/ice_srf.F90
lnd/clm2/src/main/clm_comp.F90
lnd/clm2/src/main/driver.F90
lnd/clm2/src/main/lnd_comp_mct.F90
ocn/dom/ocn_comp.F90
ocn/dom/ocn_comp_mct.F90
ocn/som/ocn_comp.F90
ocn/som/ocn_comp_mct.F90
modified performance timer logic
atm/cam/src/control/cam_comp.F90
single spmdstats output file option
atm/cam/src/control/cam_history.F90
process 0-only reporting (eliminating unnecessary output)
atm/cam/src/control/print_memusage.F90
process 0-only reporting memory usage. Old default restored
by defining DEBUG.
atm/cam/src/dynamics/eul/spmd_dyn.F90
atm/cam/src/dynamics/sld/spmd_dyn.F90
process 0-only reporting of SPMD buffer allocation.
atm/cam/src/physics/cam1/chem_surfvals.F90
atm/cam/src/physics/cam1/prescribed_aerosols.F90
process 0-only reporting of runtime options
drv/seq_mct/seq_domain_mct.F90
deleted 'call shr_sys_flush(25+iam)'
drv/seq_mct_drv/seq_ccsm_drv.F90
disabled CATAMOUNT-specific call to usetlbuf; I/O buffering is now
the system default, and usetlbuf is unecessary
added option to create a single timing output file
lnd/clm2/src/main/restFileMod.F90
(ask T. Craig or M. Vertenstein)
utils/timing/f_wrappers.c
added t_pr_onef, t_enablef, t_disablef
utils/timing/gptl.c
utils/timing/gptl.h
added GPTL_enable, GPTL_disable
modified GPTLpr to create either one timer output file,
or one file per process
process 0-only reporting timing options. Old default restored
by defining DEBUG.
utils/timing/gptl_papi.c
modified error message (mod. from J. Edwards)
utils/timing/private.h
increased timing event name length
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: All PASSED
tempest: All PASSED
bangkok/lf95: All PASSED
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
Tag name: cam3_3_41
Originator(s): Francis Vitt
Date: 15 Nov 2006
One-line Summary: Added functionality and flexibility to the trop_mozart chemistry.
Purpose of changes:
To implement flexibility in the chemical mechanism used in trop_mozart
chemistry package. This flexibility is provided
via:
- flexible namelist variables
- helper functions which use data provided by the chemical preprocessor
Use cam implemented prognostic sea salt and dust with the trop_mozart
chemistry package.
Use cam implemented rain out of trop_mozart aerosols.
Provide the ability to use cam sulfur chemistry routines with trop_mozart
chemistry package.
Provide the ability to use prescribed chemical constituents and sources
in trop_mozart chemistry package.
Fix bug in the offline driver on bluevista.
Bugs fixed (include bugzilla ID):
#239 fv offline driver fails in cam on bluevista
Describe any changes made to build system:
Describe any changes made to the namelist:
Added several namelist variables for trop_mozart chemistry package
for the flexable chemical mechanism implementation. These are:
! trop_mozart surface emissions:
character(len=256) :: srf_emis_specifier(pcnst) ! string array where each
! string contains the constituent name and filepath separated by "->", e.g.
! srf_emis_specifier = 'CO -> /path/CO_emis_file.nc','NO -> /path/NO_emis_file.nc',...
character(len=8) :: emis_type ! 'CYCLICAL' | 'SERIAL' | 'FIXED'
integer :: emis_date ! yyyymmdd for SERIAL(yyyy) or FIXED
integer :: emis_yr_offset ! yrs added to current model year for CYCLICAL
! trop_mozart chem sources(sinks):
character(len=256) :: ext_frc_specifier(pcnst) ! string array where each
! string contains the constituent name and filepath separated by "->", e.g.
! ext_frc_specifier = 'CO -> /path/CO_extfrc_file.nc','NO -> /path/NO_extfrc_file.nc',...
character(len=8) :: ext_frc_type ! 'CYCLICAL' | 'SERIAL' | 'FIXED'
integer :: ext_frc_date ! yyyymmdd for SERIAL(yyyy) or FIXED
integer :: ext_frc_yr_offset ! yrs added to current model year for CYCLICAL
! trop_mozart fixed lower boundary:
character(len=16) :: flbc_list(pcnst) ! list of constituents
character(len=256) :: flbc_file ! lower boundary data file
character(len=8) :: flbc_type ! 'CYCLICAL' | 'SERIAL' | 'FIXED'
integer :: flbc_date ! yyyymmdd for SERIAL(yyyy) or FIXED
integer :: flbc_yr_offset ! yrs added to current yr for CYCLICAL
! trop_mozart constituents that have dry deposition
character(len=16) :: drydep_list(pcnst)
! trop_mozart constituents that are constrained in the stratosphere
character(len=16) :: fstrat_list(pcnst) ! list of constituents
character(len=256) :: fstrat_file ! data file for fixed stratosphere constituents
character(len=256) :: trop_pause_file
! trop_mozart aerosol constituents that have wet deposition
character(len=16) :: aer_wetdep_list(pcnst)
! trop_mozart use CAM sulfur chemistry driver rather than mo_setsox
logical :: use_cam_sulfchem
! trop_mozart use CAM prescribed ozone for radiation rather than trop_moz prescribed data
logical :: use_cam_ozone_data
! trop_mozart use CAM prescribed green house gases for radiation rather than trop_moz prescribed data
logical :: use_cam_ghg_data
! trop_mozart prescribed constituents
character(len=256) :: tracer_cnst_file ! prescribed data file
character(len=256) :: tracer_cnst_filelist ! list of prescribed data files (series of files)
character(len=256) :: tracer_cnst_specifier(pcnst) ! string array where each
! string contains the prescribed constituent name and field name in the data file separated by ":", e.g.
! ext_frc_specifier = 'CO : CO_VMR','OH : OH_VMR',...
logical :: tracer_cnst_rmfile ! remove data file from local disk (default .false.)
integer :: tracer_cnst_ymd ! yyyymmdd - start date of the prescribed data (default is current model date)
integer :: tracer_cnst_tod ! start time of day (seconds) of the prescribed data (default is current model time)
! trop_mozart prescribed constituent sourrces/sinks
character(len=256) :: tracer_srcs_file ! prescribed data file
character(len=256) :: tracer_srcs_filelist ! list of prescribed data files (series of files)
character(len=256) :: tracer_srcs_specifier(pcnst) ! string array where each
! string contains the constituent source name and field name in the data file separated by ":", e.g.
! ext_frc_specifier = 'CO : CO_SRC','OH : OH_SRC',...
logical :: tracer_srcs_rmfile ! remove data file from local disk (default .false.)
integer :: tracer_srcs_ymd ! yyyymmdd - start date of the prescribed data (default is current model date)
integer :: tracer_srcs_tod ! start time of day (seconds) of the prescribed data (default is current model time)
Replaced the namelist variables for the individual species emissions
files with the srf_emis_specifier variable.
Replaced trop_mozart ub and lb variables for individual species with
corresponding specifier variables as described above.
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Brian Eaton
List all subroutines eliminated:
atm/cam/src/chemistry/trop_mozart/mo_ch4_lbc.F90
atm/cam/src/chemistry/trop_mozart/mo_lb_vals.F90
atm/cam/src/chemistry/trop_mozart/mo_ub_vals.F90
atm/cam/src/chemistry/trop_mozart/mo_wetdep.F90
List all subroutines added and what they do:
atm/cam/src/chemistry/trop_mozart/mo_chem_utls.F90
provides helper functons to allow easier changes in chemical mechanisms
atm/cam/src/chemistry/trop_mozart/mo_extfrc.F90
provides the ability to include external (or prescribed) sources (or sinks)
atm/cam/src/chemistry/trop_mozart/mo_flbc.F90
applies fixed lower boundary conditions to constituents
atm/cam/src/chemistry/trop_mozart/mo_fstrat.F90
applies constraints to constituents in the stratosphere
atm/cam/src/chemistry/trop_mozart/mo_sethet.F90
replaces mo_wetdep module -- set heterogeneous reaction rates
atm/cam/src/chemistry/trop_mozart/mo_synoz.F90
implements synthetic ozone scheme in the stratosphere
atm/cam/src/chemistry/trop_mozart/mz_aerosols_intr.F90
handles the wet depostion of trop_mozart aerososl as well as the
operation of cam implemented sulfur chemistry routines on trop_mozart
constituents
atm/cam/src/chemistry/trop_mozart/tracer_cnst.F90
atm/cam/src/chemistry/trop_mozart/tracer_data.F90
atm/cam/src/chemistry/trop_mozart/tracer_srcs.F90
These provide the ability to use prescribed constituents as well as
sources for the trop_mozart chemical mechanism.
atm/cam/src/chemistry/trop_mozart/rad_cnst_data_interface.F90
Overrides the cam routines that set radiative constituents to prescribed
data to use trop_mozart prescribed data.
atm/cam/src/physics/cam1/rad_cnst_data_interface.F90
provides the chemistry packages the opportunity to override procedures
that set the radiative constituents from prescribed data
atm/cam/src/dynamics/fv/advect_tend.F90
Handles the computation and history output of advection tendency.
List all existing files that have been modified, and describe the changes:
atm/cam/bld/DefaultCAM_INPARM_Namelist.xml
Change defualt upper boundary data files and dust input files.
atm/cam/bld/cam_inparm.pm
Made changes to accommodate the namelist changes for the trop_mozart
chemistry package.
Set default aero_feedback=.false. since a special aerosol optics
file is required for dust.
atm/cam/bld/config_trop_chem_mozart_defaults.xml
set prog_aero="dust,seasalt" and nadv=98 for default trop_mozart chemistry
atm/cam/bld/configure
- allow prog_aero seasalt without dust for trop_mozart chemistry
- set cppdef DUST for seasalt or dust when trop_mozart is used
- disallow prog_aero sulfur and caer (carbon aerosols) when
trop_mozart is used
atm/cam/src/chemistry/trop_mozart/chem_mods.F90
atm/cam/src/chemistry/trop_mozart/chemistry.F90
atm/cam/src/chemistry/trop_mozart/m_het_id.F90
atm/cam/src/chemistry/trop_mozart/m_spc_id.F90
atm/cam/src/chemistry/trop_mozart/mo_aerosols.F90
atm/cam/src/chemistry/trop_mozart/mo_airplane.F90
atm/cam/src/chemistry/trop_mozart/mo_chemini.F90
atm/cam/src/chemistry/trop_mozart/mo_constants.F90
atm/cam/src/chemistry/trop_mozart/mo_drydep.F90
atm/cam/src/chemistry/trop_mozart/mo_exp_sol.F90
atm/cam/src/chemistry/trop_mozart/mo_gas_phase_chemdr.F90
atm/cam/src/chemistry/trop_mozart/mo_imp_sol.F90
atm/cam/src/chemistry/trop_mozart/mo_indprd.F90
atm/cam/src/chemistry/trop_mozart/mo_jlong.F90
atm/cam/src/chemistry/trop_mozart/mo_jshort.F90
atm/cam/src/chemistry/trop_mozart/mo_lightning.F90
atm/cam/src/chemistry/trop_mozart/mo_lin_matrix.F90
atm/cam/src/chemistry/trop_mozart/mo_lu_factor.F90
atm/cam/src/chemistry/trop_mozart/mo_lu_solve.F90
atm/cam/src/chemistry/trop_mozart/mo_mass_xforms.F90
atm/cam/src/chemistry/trop_mozart/mo_nln_matrix.F90
atm/cam/src/chemistry/trop_mozart/mo_photo.F90
atm/cam/src/chemistry/trop_mozart/mo_photoin.F90
atm/cam/src/chemistry/trop_mozart/mo_prod_loss.F90
atm/cam/src/chemistry/trop_mozart/mo_rtlink.F90
atm/cam/src/chemistry/trop_mozart/mo_seasalt.F90
atm/cam/src/chemistry/trop_mozart/mo_setaer.F90
atm/cam/src/chemistry/trop_mozart/mo_setext.F90
atm/cam/src/chemistry/trop_mozart/mo_setinv.F90
atm/cam/src/chemistry/trop_mozart/mo_setrxt.F90
atm/cam/src/chemistry/trop_mozart/mo_setsoa.F90
atm/cam/src/chemistry/trop_mozart/mo_setsox.F90
atm/cam/src/chemistry/trop_mozart/mo_sim_dat.F90
atm/cam/src/chemistry/trop_mozart/mo_srf_emissions.F90
atm/cam/src/chemistry/trop_mozart/mo_strato_sad.F90
atm/cam/src/chemistry/trop_mozart/mo_tropopause.F90
atm/cam/src/chemistry/trop_mozart/mo_usrrxt.F90
Changes in the trop_mozart chemistry are mostly for:
- the addition to SYNOZ/O3RAD ozone formulation and NH4/H2SO4 chemistry
- flexible namelist variables
- addition of helper functions to allow flexibility in terms of
chemical mechanisms facilitated by the chemical preprocessor
atm/cam/src/chemistry/waccm_mozart/chemistry.F90
interface changes that correspond to the interface changes of
trop_mozart/chemistry.F90
atm/cam/src/control/atm_comp_mct.F90
Added a check to insure that dust constituents are in the
simulation before setting the dust surface fluxes.
atm/cam/src/control/runtime_opts.F90
Added several namelist variables for trop_mozart chemistry package
for the flexible chemical mechanism implementation (see above).
Added a check to not allow trace_gas = .true. when trop_mozart
chemistry package is used.
atm/cam/src/dynamics/fv/dyn_comp.F90
Changed the actual argument "dyn_state%grid" to "grid" to
subroutines metdata_dyn_init, get_met_fields and adjust_press.
atm/cam/src/dynamics/fv/metdata.F90
Removed the unnecessary ghosting of US and VS in the get_us_vs
subroutine.
atm/cam/src/dynamics/fv/stepon.F90
Added calls to compute and output advection tendencies.
atm/cam/src/physics/cam1/aerosol_intr.F90
Moved the call to hub2atm_sepopts to aerosol_register_cnst
from the dust_intr module -- this is to allow one to have
sea salt constituents without dust constituents.
Disabled the register call for dust and sea salts when
trop_mozart chemistry package is used.
Added the invocation of wet deposition of trop_mozart aerosols.
atm/cam/src/physics/cam1/aerosol_radiation_interface.F90
added optical depth history outputs of carbon and dust
atm/cam/src/physics/cam1/chemistry.F90
interface changes that correspond to the interface changes of
trop_mozart/chemistry.F90
atm/cam/src/physics/cam1/convect_deep.F90
add prognostics outputs of convective mass fluxes
atm/cam/src/physics/cam1/drydep_mod.F90
don't allow surface friction velocity returned from calcram
to be equal to zero -- zero fv causes a float point exception
in the dry deposition calculation of dust and sea salt
atm/cam/src/physics/cam1/dust_intr.F90
changes dust names
a lot of code cleanup
added the public interfaces:
dust_set_idx, dust_names, and dust_has_wet_dep
atm/cam/src/physics/cam1/progseasalts_intr.F90
changes sea salt names
a lot of code cleanup
added the public interfaces:
progseasalts_set_idx, progseasalt_names, and progseasalt_has_wet_dep
atm/cam/src/physics/cam1/rad_constituents.F90
added string variables to be used for radiative constituent names
atm/cam/src/physics/cam1/radiation.F90
replaced hard wired radiative constituent names with string
variables defined on the rad_constituents module
atm/cam/src/physics/cam1/restart_physics.F90
added a call to write and read chemistry data to restart file
atm/cam/src/physics/cam1/sulchem.F90
created a sulfur chemistry driver seperate from the wet deposition
-- this gives the trop_mozart chemistry the opportunity to use
the "cam" sulfur chemistry routines without the wet deposition.
atm/cam/src/physics/cam1/sulfur_intr.F90
moved some of the addfld calls to sulchem.F90 -- where the
fields are output to history
atm/cam/src/physics/waccm/chemistry.F90
interface changes that correspond to the interface changes of
trop_mozart/chemistry.F90
drv/seq_flds/dust/seq_flds_mod.F90
corrected fortran syntax errors and made changes for ibm compile
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista:
037 bl354 TBL.sh fm2dh outfrq3s 9 .................................FAIL! rc= 7
040 bl355 TBL.sh fmo2dh off2x2.5 9 ................................FAIL! rc= 5
These failures are expected due to changes in trop_mozart chemistry package.
tempest: All Pass
bangkok/lf95: All Pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_3_40
Originator(s): eaton
Date: Tue Nov 7 10:19:16 MST 2006
One-line Summary: Misc updates
Purpose of changes:
. Replace all occurances of 273.16 either by SHR_CONST_TKFRZ or by tmelt
which is the CAM specific constant that's set to SHR_CONST_TKFRZ in the
physconst module.
. Remove the cfort.h file -- put the Fortran name mangling CPP macro into
CAM and SCAM's configure.
. Remove cfort.h from the timing library. This required creating a new
trunk_tag for the timing library (timing_061028) and updating CAM's
external property to use it.
. Update MCT to trunk_tag MCT2_2_3_061018.
. Update CLM to trunk_tag clm3_expa_76.
. Remove excessive info/debug print statements (there is more to do).
. Fix bug (introduced at cam3_3_34) to enable running w/ HOMME dycore.
Bugs fixed (include bugzilla ID):
. bugID 224 -- fix for SHR_CONST_TKFRZ
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes:
. possible performance improvements on certain platforms due to update of
MCT and CLM externals
Code reviewed by: eaton
List all subroutines eliminated:
atm/cam/src/utils/cfort.h
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
. SHR_CONST_TKFRZ mods in:
models/atm/cam/src/physics/cam1/stratiform.F90
models/atm/cam/src/physics/cam1/wetdep.F90
models/atm/cam/src/physics/cam1/sulchem.F90
models/atm/cam/src/physics/cam1/water_isotopes.F90
models/atm/cam/src/chemistry/trop_mozart/mo_drydep.F90
. mods to remove cfort.h in:
models/atm/cam/bld/configure
models/atm/cam/bld/Makefile.in
models/atm/cam/tools/scam/configure
models/atm/cam/src/control/get_memusage.c
models/atm/cam/src/control/linebuf_stdout.c
models/atm/cam/src/control/uselbuf.c
models/utils/timing/f_wrappers.c # create trunk_tag timing_061028
. update external for timing library, MCT, and CLM
models/SVN_EXTERNAL_DIRECTORIES
models/utils/SVN_EXTERNAL_DIRECTORIES
. remove excessive info print statements
models/atm/cam/src/physics/cam1/phys_grid.F90
models/atm/cam/src/control/cam_history.F90
. Replace deprecated float intrinsic by real
models/atm/cam/src/control/interpolate_data.F90
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista: All PASS except expected failures of offline trop_mozart tests:
038 sm355 TSM.sh fmo2dh off2x2.5 9 ................................FAIL! rc= 8
039 er355 TER.sh fmo2dh off2x2.5 4+5 ..............................FAIL! rc= 5
040 bl355 TBL.sh fmo2dh off2x2.5 9 ................................FAIL! rc= 4
tempest: All PASS.
bangkok/lf95: All PASS.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: BFB
===============================================================
===============================================================
Tag name: cam3_3_39
Originator(s): mvertens
Date: Fri Nov 3 16:40:37 MST 2006
One-line Summary: Top level driver now invokes cam as one run phase
Purpose of changes: To have CAM called as one run phase from the top level driver.
This is necessary if the surface components are to couple to cam at frequencies
other than the cam time step (a feature that will be required for a sequential CCSM).
Bugs fixed (include bugzilla ID): #255
Describe any changes made to build system: None
Describe any changes made to the namelist: None
List any changes to the defaults for the boundary datasets: None
Describe any substantial timing or memory changes: None
Code reviewed by: eaton
List all subroutines eliminated: None
List all subroutines added and what they do: None
List all existing files that have been modified, and describe the changes:
M atm/cam/tools/scam/scm_init/init_model.F90
M atm/cam/tools/scam/scm_init/scam_run.F90
M atm/cam/src/control/atm_comp_mct.F90
M drv/seq_mct_drv/seq_ccsm_drv.F90
- Seq_ccsm_drv.F90 and scam_run.F90 will now call only one run phase of cam.
The basic idea is that the first time the cam run phase is called outside
of the time loop, it will only call tphysbc. Within the run loop the cam
mct levelrun phase will call tphysac, update the cam time and then call tphysbc.
M drv/seq_mct/seq_domain_mct.F90
- removed extra writes
M drv/seq_mct/seq_flux_mct.F90
- fixed bug #255
M drv/seq_flds/dust/seq_flds_mod.F90
M drv/seq_flds/default/seq_flds_mod.F90
M drv/seq_flds/gensom/seq_flds_mod.F90
M drv/seq_flds_indices/seq_flds_indices.F90
- Added Foxx_salt to drv->ocn fields
Note needed now, but will be needed when pop is used for active ocn
M drv/seq_mct/mrg_x2o_mct.F90
- Fixed bugs for fields needed in merging when active ocean will be
utilized
M ice/csim4/ice_data.F90
M ice/csim4/ice_dh.F
M ice/csim4/ice_srf.F90
M ice/csim4/ice_comp.F90
M atm/cam/src/control/runtime_opts.F90
M atm/cam/src/control/cam_control_mod.F90
- moved input namelist variables prognostic_icesnow and reset_csim_iceprops
from ice_dh to cam_control_mod.F90. This will make it possible to replace
cam-csim with cice4.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista:
019 bl331 TBL.sh f4c11gdh pghgsul 9 ...............................FAIL! rc= 7
- this test failed due to a sign change implemented for the latent/sensible
oi gensom fluxes (this was a bug introduced in cam3_3_37 that did not get picked
up in the testing.
038 sm355 TSM.sh fmo2dh off2x2.5 9 ................................FAIL! rc= 8
039 er355 TER.sh fmo2dh off2x2.5 4+5 ..............................FAIL! rc= 5
040 bl355 TBL.sh fmo2dh off2x2.5 9 ................................FAIL! rc= 4
- these tests began failing with the upgrade of the compilers on bluevista
around 060918...a retest of a previous tag (cam3_3_28) where these tests
had passed, showed these tests now fail as well...it fails run-time as it
tries to read in the met data...see bug #239
tempest: All tests passed except for
021 bl331 TBL.sh f4c11gdh pghgsul 9 ...............................FAIL! rc= 7
- this tested failed due to the same gensom bug fix as appeared in bluevista
test 19
bangkok/lf95: All tests passed
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: None, answers are bfb except for gensom oi diagnostics
===============================================================
===============================================================
Tag name: cam3_3_38
Originator(s): mvr
Date: 20061027
One-line Summary: mods to resolve naming conflict when cam is used in
a ccsm single-executable; bluevista replaces bluesky as required pretag
test platform; mods to conform to changes in ccsm's test scripts
Purpose of changes: one more naming conflict in ccsm single-executable
that was overlooked; bluesky is scheduled for decommission by the end
of the year, so bluevista will now be the platform required for testing
cam prior to commits; ccsm changed its default behavior for cleanup
of files after successful tests - a change was made to specifically
turn off this cleanup and let cam's test scripts handle the cleanup
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: myself
List all subroutines eliminated:
D models/atm/cam/src/control/restart.F90
- just a rename required to clear up name conflict in ccsm single-executable
List all subroutines added and what they do:
A + models/atm/cam/src/control/cam_restart.F90
- newly named file and mods to clear up name conflict in ccsm single-executable
A models/atm/cam/test/system/tests_pretag_bluevista
- new file with test ids to run by default for pretag testing on bluevista
List all existing files that have been modified, and describe the changes:
M models/ocn/dom/ocn_comp.F90
M models/ocn/som/ocn_comp.F90
M models/ice/csim4/ice_comp.F90
M models/atm/cam/src/control/runtime_opts.F90
M models/atm/cam/src/control/ccsm_msg.F90
M models/atm/cam/src/control/cam_comp.F90
M models/drv/seq_mct/mrg_x2a_mct.F90
- files modified to reflect name change: restart -> cam_restart
M ChangeLog_template
M models/atm/cam/test/system/TSB.ccsm.sh
M models/atm/cam/test/system/TCT.ccsm.sh
M models/atm/cam/test/system/tests_posttag_bluesky
M models/atm/cam/test/system/test_driver.sh
M models/atm/cam/test/system/tests_posttag_bluevista
- mods to test scripts to have bluevista replace bluesky as the required
platform for pretag testing, utilize mods in ccsm scripts for the bluevista16
machine definition, and turned off ccsm's cleanup functionality
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluevista:
038 sm355 TSM.sh fmo2dh off2x2.5 9 ................................FAIL! rc= 8
039 er355 TER.sh fmo2dh off2x2.5 4+5 ..............................FAIL! rc= 5
040 bl355 TBL.sh fmo2dh off2x2.5 9 ................................FAIL! rc= 4
- these tests began failing with the upgrade of the compilers on bluevista
around 060918...a retest of a previous tag (cam3_3_28) where these tests
had passed, showed these tests now fail as well...it fails run-time as it
tries to read in the met data...see bug #239
tempest:
all pass
bangkok/lf95:
all pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_3_37
Originator(s): mvertens
Date: Tue Oct 24 22:58:04 MDT 2006
One-line Summary: use cpl6 ocn/atm flux calculation and surface flux sign convention
Purpose of changes: To bring the current sequential system in line with the
concurrent system atm/ocn flux calculation, ocn and ice and merge functionalities,
and ccsm surface flux sign convention. In addition, to remove the need for the ice
and ocean components to know pass any landfrac information to the top level driver.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none (to the best of my knowledge)
Code reviewed by: mvertens, rjacob, tcraig
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
M ocn/dom/ocn_comp.F90
- Removed all references to lwup
M ocn/dom/ocn_comp_mct.F90
- Removed latr and lonr from dom_o - conversion is now done in seq_flux_mct.F90
- Removed o2x_o(index_o2x_Faoo_lwup,ig) - (now done in atm/ocn flux calc)
- Removed setting of maxfrac to 1.-landfrac in dom_o
M ocn/som/ocn_types.F90
- Removed lwup from ocn_out type
M ocn/som/ocn_comp.F90
- Removed computation of lwup (now done in atm/ocn flux calc)
- Removed lwup from restart file
M ocn/som/ocn_comp_mct.F90
- Removed latr and lonr from dom_o - conversion is now done in seq_flux_mct.F90
- Removed o2x_o(index_o2x_Faoo_lwup,ig) - (now done in atm/ocn flux calc)
- Removed setting of maxfrac to 1.-landfrac in dom_o
- Set sign convention to fluxes are positive downwards
- Moved multiplication of fswabs, shflx, lhflx, lwdn and lwup by (1.-aice)
to mrg_x2o_mct.F90 (this is consistent now with ocn merging in cpl6)
M ice/csim4/ice_types.F90
- Made cflx component of ice_out one-dimensional
- Made qbot component of ice_in one-dimensional
- Removed reference to pcnst
M ice/csim4/ice_srf.F90
- Fluxes are now positive downwards for lwup, shflx, lhflx, taux, tauy, qflx
- Removed setting of non-water constituent fluxes to zero
- Arguemnts qbot and qflx no longer depend on pcnst+pnats
M ice/csim4/ice_diagnostics.F
- Made qbot component of ice_in one-dimensional
M ice/csim4/ice_comp.F90
- Sign convention for lwup is now that fluxes are positive downwards
- Removed references to aqua-planet
M ice/csim4/ice_comp_mct.F90
- Removed i2x_i%rAttr(index_i2x_Si_aice,ig) and set i2x_i%rAttr(index_i2x_Si_ifrac,ig)
to now correspond to fraction of ice with respect to ocean part of gridcell
M atm/cam/test/system/test_driver.sh
- Updates to lightning and bluevista tests
- Replaced x2a_a%rAttr(index_x2a_Faoo_lwup,n) with x2a_a%rAttr(index_x2a_Faox_lwup,n)
M atm/cam/src/control/atm_comp_mct.F90
- Sign convention is that all fluxes from driver are positive downwards
M drv/seq_flds/dust/seq_flds_mod.F90
M drv/seq_flds/default/seq_flds_mod.F90
M drv/seq_flds/gensom/seq_flds_mod.F90
M drv/seq_flds_indices/seq_flds_indices.F90
- Added fields and indices to get closer to cpl_fields_mod. This involved adding:
'So_u' ! velocity, zonal DEF
'So_v' ! velocity, meridional DEF
'So_s' ! salinity DEF
'So_dhdx' ! surface slope, zonal DEF
'So_dhdy' ! surface slope, meridional DEF
'Si_ifrac' ! state: ice fraction wrt ocean DEF
'Sa_pslv' ! state: sea level pressure DEF
'Foxx_lwup' ! heat flux: long-wave up DEF
'Foxx_prec' ! water flux: rain+snow DEF
'Foxx_snow' ! water flux: snow DEF
'Foxx_rain' ! water flux: rain DEF
'Foxx_evap' ! water flux: evap DEF
'Foxx_meltw' ! water flux: melt DEF
'Forr_roff' ! water flux: runoff DEF
'Faoo_lwup' ! shortwave: net absorbed DEF
'Faox_lwup' ! upward longwave heat flux DEF
and removed Si_aice and changed the previous functionality of Si_ifrac
Some of the above fields are not currently used - but are place holders for
when the ccsm components will be put in.
M atm/cam/tools/scam/scm_init/init_model.F90
- Added domain information to mapping initialization
M drv/seq_mct_drv/seq_ccsm_drv.F90
- Added domain information to mapping initialization
M drv/seq_mct/mrg_x2a_mct.F90
- Replaced index_o2x_Faoo_lwup with index_xao_Faox_lwup
M drv/seq_mct/mrg_x2o_mct.F90
- Merging functionality for input to ocn component is now same as cpl6
M drv/seq_mct/seq_domain_mct.F90
- maxfrac_o = 1.- maxfrac_l is set for points where maxfrac_l > 0 and
maxfrac_l < 1. The same is true for maxfrac_i. As a result, the ocean
and ice components need not know the land fraction in the gridcell.
Furthermore, the ice fraction need only be scaled as that relative to
the total gridcell for merging functionality from drv->atm. The ice fraction
is only needed relative to the ocean part of the gridcell for merging
functionality from drv->ocn.
M drv/seq_mct/seq_flux_mct.F90
- xao_o%rAttr(index_xao_Faox_lwup,n) is now used (instead of lwup
being calculated by the ocean components)
- flux_albo is not identical to cpl6 code
M drv/seq_mct/flux_ao.F90
- This code is now the same as the flux_atmOcn routine in the cpl6 flux_mod.F90.
Both this routine and flux_atmOcn will be migrated to a csm_share/shr/shr_flux_mod.F90
code base in upcoming tags.
M drv/seq_mct/seq_mct_mod.F90
- Added use aliases for mct_aVectimport Rattr and mct_aVect_exportRattr
M drv/seq_mct/map_atmice_mct.F90
- Obtain maxfrac_ice on atmosphere decomp, needed for merging to atmosphere grid
when all components are on same grid and when partial land cover in ice gridcell.
- Added domain information to initialization phase
M drv/seq_mct/map_atmlnd_mct.F90
M drv/seq_mct/map_iceocn_mct.F90
M drv/seq_mct/map_atmocn_mct.F90
- Added domain information to initialization phases
SVN_EXTERNALS has been updated to clm3_expa_75
M lnd/clm2/src/main/clm_time_manager.F90
- Removed write statement to fort.30
M lnd/clm2/src/main/lnd_comp_mct.F90
- Changed sign convention of fluxes sent to driver to be consistent
with those used by concurrent ccsm
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: all tests passed except for
005 bl131 TBL.sh e32c11dh pghgsul 9 ...............................FAIL! rc= 7
015 bl151 TBL.sh e64h outfrq24h -2 ................................FAIL! rc= 7
019 bl331 TBL.sh f4c11gdh pghgsul 9 ...............................FAIL! rc= 7
030 bl336 TBL.sh wm4h outfrq3s 9 ..................................FAIL! rc= 7
034 bl351 TBL.sh f2h outfrq24h -2 .................................FAIL! rc= 7
037 bl354 TBL.sh fm2dh outfrq3s 9 .................................FAIL! rc= 7
044 bl531 TBL.sh s32c11dh pghgsul 9 ...............................FAIL! rc= 7
047 bl532 TBL.sh s32sdh ghgrmp 9 ..................................FAIL! rc= 7
054 bl551 TBL.sh s64h outfrq24h -2 ................................FAIL! rc= 5
the above were bfb except for the TREFHT output (which is not totally different)
008 bl132 TBL.sh e32sdh ghgrmp 9 ..................................FAIL! rc= 7
022 bl332 TBL.sh f4sdh ghgrmp 9 ...................................FAIL! rc= 7
040 bl355 TBL.sh fmo2dh off2x2.5 9 ................................FAIL! rc= 7
the above represent round-off level diffs for som tests
tempest: all tests passed except for
004 bl131 TBL.sh e32c11dh pghgsul 9 ...............................FAIL! rc= 7
014 bl152 TBL.sh e64o outfrq24h -2 ................................FAIL! rc= 7
017 bl314 TBL.sh wg10dh outfrq3s 9 ................................FAIL! rc= 7
021 bl331 TBL.sh f4c11gdh pghgsul 9 ...............................FAIL! rc= 7
031 bl352 TBL.sh f2o outfrq24h -2 .................................FAIL! rc= 7
035 bl531 TBL.sh s32c11dh pghgsul 9 ...............................FAIL! rc= 7
045 bl552 TBL.sh s64o outfrq24h -2 ................................FAIL! rc= 7
the above were bfb except for the TREFHT output (which is not totally different)
007 bl132 TBL.sh e32sdh ghgrmp 9 ..................................FAIL! rc= 7
024 bl332 TBL.sh f4sdh ghgrmp 9 ...................................FAIL! rc= 7
038 bl532 TBL.sh s32sdh ghgrmp 9 ..................................FAIL! rc= 7
the above represent round-off level diffs for som tests
bangkok/lf95: all tests passed except for
004 bl111 TBL.sh e8c8mdm ttrac 9 ..................................FAIL! rc= 7
016 bl311 TBL.sh f10c8mdm ttrac 9 .................................FAIL! rc= 7
024 bl312 TBL.sh f10sdm ghgrmp 9 ..................................FAIL! rc= 7
032 bl511 TBL.sh s8c8mdm ttrac 9 ..................................FAIL! rc= 7
the above were bfb except for the TREFHT output (which is not totally different)
012 bl153 TBL.sh e64m outfrq24h -2 ................................FAIL! rc= 7
028 bl353 TBL.sh f2m outfrq24h -2 .................................FAIL! rc= 7
040 bl553 TBL.sh s64m outfrq24h -2 ................................FAIL! rc= 7
the above represent round-off level diffs due to higher optimization levels
than the 9 timestep tests
008 bl112 TBL.sh e8sdm ghgrmp 9 ...................................FAIL! rc= 7
036 bl512 TBL.sh s8sdm ghgrmp 9 ...................................FAIL! rc= 7
the above represent round-off level diffs for som tests
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- nature of change (larger than roundoff but same climate)
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
cam3_3_37
- platform/compilers:
ibm (bluesky)
- configure commandline:
configure -ocn dom -dyn fv -spmd -smp -test -res 4x5 -gensom
- build-namelist command (or complete namelist):
&ccsm_inparm
case_name = 'cam3_3_34_nonbfb_fv4x5dom'
start_type = "startup"
/
&timemgr_inparm
atm_cpl_dt = 1800
orb_iyear_ad = 1950
restart_option= 'monthly'
start_ymd = 10101
stop_n = 4000
stop_option = 'ndays'
/
&cam_inparm
absems_data = '/fis/cgd/cseg/csm/inputdata/atm/cam/rad/abs_ems_factors_fastvx.c030508.nc'
aeroptics = '/fis/cgd/cseg/csm/inputdata/atm/cam/rad/AerosolOptics_c050419.nc'
bnd_topo = '/fis/cgd/cseg/csm/inputdata/atm/cam/topo/USGS-gtopo30_4x5_remap_c050520.nc'
bndtvaer = '/fis/cgd/cseg/csm/inputdata/atm/cam/rad/AerosolMass_V_4x5_clim_c031022.nc'
bndtvo = '/fis/cgd/cseg/csm/inputdata/atm/cam/ozone/pcmdio3.r8.64x1_L60_clim_c970515.nc'
bndtvs = '/fis/cgd/cseg/csm/inputdata/atm/cam/sst/sst_HadOIBl_bc_4x5_clim_c030228.nc'
dtime = 1800
ncdata = '/fis/cgd/cseg/csm/inputdata/atm/cam/inic/fv/cami_0001-01-01_4x5_L26_c060608.nc'
/
&clm_inparm
dtime = 1800
fpftcon = '/fis/cgd/cseg/csm/inputdata/lnd/clm2/pftdata/pft-physiology-cn16.c040719'
fsurdat = '/fis/cgd/cseg/csm/inputdata/lnd/clm2/srfdata/cam/clms_3.1_4x5_c050523.nc'
/
MSS location of output:
/MVERTENS/csm/cam3_3_34_nonbfb_fv4x5dom
MSS location of control simulations used to validate new climate:
/MVERTENS/csm/cam3_3_34_nonbfb_fv4x5dom_0
URL for AMWG diagnostics output used to validate new climate:
http://www.cgd.ucar.edu/cms/rneale/runs/cam/cam3_3_34_nonbfb_fv4x5dom-cam3_3_34_nonbfb_fv4x5dom_0/
===============================================================
===============================================================
Tag name: cam3_3_36
Originator(s): mvertens
Date: Tue Oct 24 21:58:40 MDT 2006
One-line Summary: non-bfb changes corresponding to cam3_3_37
Purpose of changes: to create the minimal code base that generates the
greater than round-off changes in cam3_3_37. Since cam3_3_37 will
contain number round-off level changes for som, it is useful to isolate
the few pieces of code that produced the greater than round-off differences.a
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: mvertens
M models/ocn/dom/ocn_comp.F90
- Computation of lwup is occuring before temperature is updated. This is
now consistent with obtaining lwup from atm/ocn flux computation. This
will ensure that all fluxes in atm/ocn flux computation (including lwup)
are based on the same surface temperature.
M models/ocn/som/ocn_comp.F90
- No longer multiplying onf by 1.-aice. Now doing this to each component of
onf in ocn_comp_mct.F90
M models/ocn/som/ocn_comp_mct.F90
- Now multiplying Foxx_melth, Foxx_swnet, Foxx_sen, Foxx_lat and Foxx_lwdn,
and lwup by 1.-aice.
M models/atm/cam/src/control/atm_comp_mct.F90
- At first time step determine the longwave up flux from the surface
temperature. This is identical to what is being done in ccsm_msg.F90
M models/drv/seq_mct/seq_flux_mct.F90
- Put in computation to no longer require lat/lon to be passed in radians
from the ocean domain. Ocean domain will now pass lat/lon in degrees and
will convert to radians.
- Set ocean albedos to 1. when coszen < 0.
M models/drv/seq_mct/flux_ao.F90
- Set umin to 0.5_r8
This is the only answer changing modification between this code and the
atm/ocn flux calcuation code in cpl6. The goal is for both the concurrent
and sequential ccsm to use the same atm/ocn flux calculation code base.
Note that the diagnostic TREF will be totally different when the cpl6
atm/ocn flux calculation is utilized
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: all passed except for following baseline comparisons
005 bl131 TBL.sh e32c11dh pghgsul 9 ...............................FAIL! rc= 7
008 bl132 TBL.sh e32sdh ghgrmp 9 ..................................FAIL! rc= 7
010 bl133 TBL.sh e32pdh aqpgro 3 ..................................FAIL! rc= 7
015 bl151 TBL.sh e64h outfrq24h -2 ................................FAIL! rc= 7
019 bl331 TBL.sh f4c11gdh pghgsul 9 ...............................FAIL! rc= 7
022 bl332 TBL.sh f4sdh ghgrmp 9 ...................................FAIL! rc= 7
024 bl333 TBL.sh f4pdh aqpgro 3 ...................................FAIL! rc= 7
030 bl336 TBL.sh wm4h outfrq3s 9 ..................................FAIL! rc= 7
034 bl351 TBL.sh f2h outfrq24h -2 .................................FAIL! rc= 7
037 bl354 TBL.sh fm2dh outfrq3s 9 .................................FAIL! rc= 7
040 bl355 TBL.sh fmo2dh off2x2.5 9 ................................FAIL! rc= 7
044 bl531 TBL.sh s32c11dh pghgsul 9 ...............................FAIL! rc= 7
047 bl532 TBL.sh s32sdh ghgrmp 9 ..................................FAIL! rc= 7
049 bl533 TBL.sh s32pdh aqpgro 3 ..................................FAIL! rc= 7
054 bl551 TBL.sh s64h outfrq24h -2 ................................FAIL! rc= 7
tempest: all passed except for following baseline comparisons
004 bl131 TBL.sh e32c11dh pghgsul 9 ...............................FAIL! rc= 5
007 bl132 TBL.sh e32sdh ghgrmp 9 ..................................FAIL! rc= 5
009 bl133 TBL.sh e32pdh aqpgro 3 ..................................FAIL! rc= 5
012 bl135 TBL.sh e32dh idphys 9 ...................................FAIL! rc= 5
014 bl152 TBL.sh e64o outfrq24h -2 ................................FAIL! rc= 5
017 bl314 TBL.sh wg10dh outfrq3s 9 ................................FAIL! rc= 5
021 bl331 TBL.sh f4c11gdh pghgsul 9 ...............................FAIL! rc= 5
024 bl332 TBL.sh f4sdh ghgrmp 9 ...................................FAIL! rc= 5
026 bl333 TBL.sh f4pdh aqpgro 3 ...................................FAIL! rc= 5
029 bl335 TBL.sh f4dh idphys 9 ....................................FAIL! rc= 5
031 bl352 TBL.sh f2o outfrq24h -2 .................................FAIL! rc= 5
035 bl531 TBL.sh s32c11dh pghgsul 9 ...............................FAIL! rc= 5
038 bl532 TBL.sh s32sdh ghgrmp 9 ..................................FAIL! rc= 5
040 bl533 TBL.sh s32pdh aqpgro 3 ..................................FAIL! rc= 5
043 bl535 TBL.sh s32dh idphys 9 ...................................FAIL! rc= 5
045 bl552 TBL.sh s64o outfrq24h -2 ................................FAIL! rc= 5
bangkok/lf95: all passed except for following baseline comparisons
004 bl111 TBL.sh e8c8mdm ttrac 9 ..................................FAIL! rc= 7
008 bl112 TBL.sh e8sdm ghgrmp 9 ...................................FAIL! rc= 7
010 bl113 TBL.sh e8pdm aqpgro 3 ...................................FAIL! rc= 7
012 bl153 TBL.sh e64m outfrq24h -2 ................................FAIL! rc= 7
016 bl311 TBL.sh f10c8mdm ttrac 9 .................................FAIL! rc= 7
024 bl312 TBL.sh f10sdm ghgrmp 9 ..................................FAIL! rc= 7
026 bl313 TBL.sh f10pdm aqpgro 3 ..................................FAIL! rc= 7
028 bl353 TBL.sh f2m outfrq24h -2 .................................FAIL! rc= 7
032 bl511 TBL.sh s8c8mdm ttrac 9 ..................................FAIL! rc= 7
036 bl512 TBL.sh s8sdm ghgrmp 9 ...................................FAIL! rc= 7
038 bl513 TBL.sh s8pdm aqpgro 3 ...................................FAIL! rc= 7
040 bl553 TBL.sh s64m outfrq24h -2 ................................FAIL! rc= 7
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- nature of change (larger than roundoff but same climate)
see complete documentation of runs to evaluate changes in cam3_3_37 tag info
===============================================================
===============================================================
Tag name: cam3_3_35
Originator(s): Jim Edwards
Date: 10-11-2006
One-line Summary: homme dycore development, replacement of cam common blocks
Purpose of changes: development, code clean-up
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: refined build support for target_os=bgl
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes:
Code reviewed by: self
List all subroutines eliminated:
D 2094 models/atm/cam/src/control/comtfc.h
D 2094 models/atm/cam/src/control/comfft.h
D 2094 models/atm/cam/src/control/comctl.h
D 2094 models/atm/cam/src/control/comqfl.h
D 2094 models/atm/cam/src/control/perturb.h
D 2094 models/atm/cam/src/physics/cam1/comsol.h
D 2094 models/atm/cam/src/physics/cam1/comtsc.h
D 2094 models/atm/cam/src/physics/cam1/comadj.h
List all subroutines added and what they do:
These are data modules which replace the common blocks eliminated above,
because the fv_control_mod contains namelist variables it needs to be compiled
regardless of dycore - thus it's location in the control directory.
A + - models/atm/cam/src/control/cam_control_mod.F90
A + - models/atm/cam/src/control/fv_control_mod.F90
A + - models/atm/cam/src/dynamics/sld/sld_control_mod.F90
A + - models/atm/cam/src/dynamics/eul/eul_control_mod.F90
List all existing files that have been modified, and describe the changes:
Changes related to the continuing development of the homme dycore:
M 2094 models/atm/cam/src/dynamics/homme/external/forcing_mod.F90
M 2094 models/atm/cam/src/dynamics/homme/external/parallel_mod.F90
M 2094 models/atm/cam/src/dynamics/homme/external/prim_state_mod.F90
M 2094 models/atm/cam/src/dynamics/homme/external/element_mod.F90
M 2094 models/atm/cam/src/dynamics/homme/external/prim_advection_mod.F90
M 2094 models/atm/cam/src/dynamics/homme/external/dof_mod.F90
M 2094 models/atm/cam/src/dynamics/homme/external/bndry_mod.F90
M 2094 models/atm/cam/src/dynamics/homme/external/diffusion_mod.F90
M 2094 models/atm/cam/src/dynamics/homme/external/prim_advance_mod.F90
M 2094 models/atm/cam/src/dynamics/homme/external/baroclinic_inst_mod.F90
M 2094 models/atm/cam/src/dynamics/homme/external/edge_mod.F90
M 2094 models/atm/cam/src/dynamics/homme/README
M 2094 models/atm/cam/src/dynamics/homme/dp_coupling.F90
M 2094 models/atm/cam/src/dynamics/homme/stepon.F90
M 2094 models/atm/cam/src/dynamics/homme/inidat.F90
M 2094 models/atm/cam/src/dynamics/homme/dyn_comp.F90
M 2094 models/atm/cam/src/dynamics/homme/restart_dynamics.F90
M 2094 models/atm/cam/src/dynamics/homme/inital.F90
C 2094 models/ocn/dom/sst_data.F90
M 2094 models/atm/cam/tools/interpic_new/README
M 2094 models/atm/cam/bld/configure
M 2094 models/atm/cam/bld/Makefile.in
M 2094 models/atm/cam/src/control/cam_history.F90
M 2094 models/atm/cam/src/utils/abortutils.F90
M 2094 models/atm/cam/src/physics/cam1/tphysidl.F90
Changes related to the elemination of common blocks (note: there is some overlap in these lists):
C 2094 models/ocn/dom/sst_data.F90
M 2094 models/ocn/dom/ocn_comp.F90
M 2094 models/ocn/som/somint.F90
M 2094 models/ocn/som/ocn_comp.F90
M 2094 models/ocn/som/somini.F90
M 2094 models/ice/csim4/ice_data.F90
C 2094 models/ice/csim4/ice_comp.F90
M 2094 models/atm/cam/tools/scam/scm_init/init_model.F90
M 2094 models/atm/cam/tools/scam/scm_init/forecast.F90
M 2094 models/atm/cam/tools/scam/scm_init/inital.F90
M 2094 models/atm/cam/tools/scam/scm_init/readiopdata.F90
M 2094 models/atm/cam/src/control/readinitial.F90
M 2094 models/atm/cam/src/control/history_defaults.F90
M 2094 models/atm/cam/src/control/runtime_opts.F90
M 2094 models/atm/cam/src/control/ncdio_atm.F90
M 2094 models/atm/cam/src/control/srfxfer.F90
M 2094 models/atm/cam/src/control/restart.F90
M 2094 models/atm/cam/src/control/startup_initialconds.F90
M 2094 models/atm/cam/src/control/history_scam.F90
C 2094 models/atm/cam/src/control/cam_comp.F90
M 2094 models/atm/cam/src/control/cam_history.F90
M 2094 models/atm/cam/src/physics/cam1/initindx.F90
M 2094 models/atm/cam/src/physics/cam1/zenith.F90
M 2094 models/atm/cam/src/physics/cam1/radiation.F90
M 2094 models/atm/cam/src/physics/cam1/tphysbc.F90
M 2094 models/atm/cam/src/physics/cam1/check_energy.F90
M 2094 models/atm/cam/src/physics/cam1/restart_physics.F90
M 2094 models/atm/cam/src/physics/cam1/cam_diagnostics.F90
M 2094 models/atm/cam/src/physics/cam1/phys_gmean.F90
M 2094 models/atm/cam/src/physics/cam1/stratiform.F90
M 2094 models/atm/cam/src/physics/cam1/physics_types.F90
M 2094 models/atm/cam/src/physics/cam1/radlw.F90
M 2094 models/atm/cam/src/physics/cam1/tsinti.F90
M 2094 models/atm/cam/src/physics/cam1/co2_data_flux.F90
M 2094 models/atm/cam/src/physics/cam1/physpkg.F90
M 2094 models/atm/cam/src/physics/cam1/prescribed_aerosols.F90
M 2094 models/atm/cam/src/physics/cam1/dadadj.F90
M 2094 models/atm/cam/src/physics/cam1/chemistry.F90
M 2094 models/atm/cam/src/physics/cam1/phys_buffer.F90
M 2094 models/atm/cam/src/physics/cam1/ramp_scon.F90
M 2094 models/atm/cam/src/physics/cam1/phys_grid.F90
M 2094 models/atm/cam/src/physics/cam1/advnce.F90
M 2094 models/atm/cam/src/physics/cam1/tphysac.F90
M 2094 models/atm/cam/src/physics/cam1/zm_conv.F90
M 2094 models/atm/cam/src/physics/cam1/cldwat.F90
M 2094 models/atm/cam/src/physics/waccm/nlte_fomichev.F90
M 2094 models/atm/cam/src/physics/waccm/chemistry.F90
M 2094 models/atm/cam/src/chemistry/waccm_mozart/mo_gas_phase_chemdr.F90
M 2094 models/atm/cam/src/chemistry/waccm_mozart/solvar_interface.F90
M 2094 models/atm/cam/src/chemistry/waccm_mozart/mo_apex.F90
M 2094 models/atm/cam/src/chemistry/waccm_mozart/efield.F90
M 2094 models/atm/cam/src/chemistry/waccm_mozart/mo_waccm_hrates.F90
M 2094 models/atm/cam/src/chemistry/waccm_mozart/exbdrift.F90
M 2094 models/atm/cam/src/dynamics/sld/spegrd.F90
M 2094 models/atm/cam/src/dynamics/sld/linemsdyn.F90
M 2094 models/atm/cam/src/dynamics/sld/initcom.F90
M 2094 models/atm/cam/src/dynamics/sld/scanslt.F90
M 2094 models/atm/cam/src/dynamics/sld/inidat.F90
M 2094 models/atm/cam/src/dynamics/sld/dyn_comp.F90
M 2094 models/atm/cam/src/dynamics/sld/hordif.F90
M 2094 models/atm/cam/src/dynamics/sld/spetru.F90
M 2094 models/atm/cam/src/dynamics/sld/scan2.F90
M 2094 models/atm/cam/src/dynamics/sld/restart_dynamics.F90
M 2094 models/atm/cam/src/dynamics/sld/tfilt_massfix.F90
M 2094 models/atm/cam/src/dynamics/eul/spegrd.F90
M 2094 models/atm/cam/src/dynamics/eul/initcom.F90
M 2094 models/atm/cam/src/dynamics/eul/linemsdyn.F90
M 2094 models/atm/cam/src/dynamics/eul/scanslt.F90
M 2094 models/atm/cam/src/dynamics/eul/inidat.F90
M 2094 models/atm/cam/src/dynamics/eul/dyn_comp.F90
M 2094 models/atm/cam/src/dynamics/eul/hordif.F90
M 2094 models/atm/cam/src/dynamics/eul/spetru.F90
M 2094 models/atm/cam/src/dynamics/eul/scan2.F90
M 2094 models/atm/cam/src/dynamics/eul/restart_dynamics.F90
M 2094 models/atm/cam/src/dynamics/eul/tfilt_massfix.F90
M 2094 models/atm/cam/src/dynamics/fv/pmgrid.F90
M 2094 models/atm/cam/src/dynamics/fv/initcom.F90
M 2094 models/atm/cam/src/dynamics/fv/dryairm.F90
M 2094 models/atm/cam/src/dynamics/fv/inidat.F90
M 2094 models/atm/cam/src/dynamics/fv/dyn_comp.F90
M 2094 models/atm/cam/src/dynamics/fv/inital.F90
M 2094 models/atm/cam/src/dynamics/fv/stepon.F90
M 2094 models/atm/cam/src/dynamics/fv/restart_dynamics.F90
M 2094 models/atm/cam/src/dynamics/fv/metdata.F90
M 2094 models/atm/cam/src/advection/slt/difcor.F90
C 2094 models/drv/seq_mct/mrg_x2l_mct.F90
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: none
tempest: none
bangkok/lf95: none
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: - what platforms/compilers: - nature
of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_3_34
Originator(s): Will Sawyer
Date: Wed Oct 11 03:05:07 MDT 2006
One-line Summary: Reduce init time (#213), clean up FV constants (#214), allocate status (#166)
Purpose of changes:
o PILGRIM initialization time has been reduced (Bug #213). This has
been achieved by removing the communications for XZY index and replacing
them by XYZ communication patterns. These are assembled by generating
an XY decomposition comm. pattern, then extending it to 3D in the
(undistributed) Z direction. This reduces initialization time, drastically
in the case of high resolution. Note that this change required extensive
restructuring of cam_history.
o The status of allocation of large 3-D arrays in FV is checked (Bug #166)
o FV constants were cleaned up in a consistent way (Bug #214), e.g.,
real(r8), parameter :: D0_0 = 0.0_r8
real(r8), parameter :: D0_5 = 0.5_r8
real(r8), parameter :: D1_0 = 1.0_r8
real(r8), parameter :: D1_5 = 1.5_r8
Bugs fixed (include bugzilla ID): #166, #213, #214
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: FV initialization time is
now much faster, especially so for high resolutions.
Code reviewed by: myself
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/src/control/cam_history.F90: Extensive changes to use
XYZ indexing to gather and scatter arrays. Now only one type of
PILGRIM gather/scatter operation (fv_gather_??, fv_scatter_??)
is used. Addition of routines to copy the *local* portion of the
array from XYZ -> XZY (loc_xyz_to_xzy) and XZY -> XYZ (loc_xzy_to_xyz)
decompositions. Performing this local give a minor performance
improvement. Overall this reduces the length and complexity of the
module. Treats the writing of the staggered U-winds field (which
has 1 less latitude than other fields) in a cleaner way. Now
writes all fields to restart consistently in XYZ indexing.
models/atm/cam/src/dynamics/fv/benergy.F90: new treatment of constants
models/atm/cam/src/dynamics/fv/cd_core.F90: new treatment of constants
models/atm/cam/src/dynamics/fv/d2a3dijk.F90: new treatment of constants
models/atm/cam/src/dynamics/fv/d2a3dikj.F90: new treatment of constants
models/atm/cam/src/dynamics/fv/dp_coupling.F90: new treatment of constants
models/atm/cam/src/dynamics/fv/dyn_grid.F90: new treatment of constants
models/atm/cam/src/dynamics/fv/dynconst.F90: new treatment of constants
models/atm/cam/src/dynamics/fv/epvd.F90: new treatment of constants
models/atm/cam/src/dynamics/fv/fill_module.F90: new treatment of constants
models/atm/cam/src/dynamics/fv/fv_prints.F90: new treatment of constants
models/atm/cam/src/dynamics/fv/geopk.F90: new treatment of constants
models/atm/cam/src/dynamics/fv/inital.F90: new treatment of constants
models/atm/cam/src/dynamics/fv/initcom.F90: new treatment of constants
models/atm/cam/src/dynamics/fv/mapz_module.F90: new treatment of constants
models/atm/cam/src/dynamics/fv/mean_module.F90: new treatment of constants, e.g.,
models/atm/cam/src/dynamics/fv/metdata.F90: new treatment of constants
models/atm/cam/src/dynamics/fv/p_d_adjust.F90: new treatment of constants
models/atm/cam/src/dynamics/fv/par_vecsum.F90: new treatment of constants
models/atm/cam/src/dynamics/fv/pfixer.F90: new treatment of constants
models/atm/cam/src/dynamics/fv/pft_module.F90: new treatment of constants
models/atm/cam/src/dynamics/fv/stepon.F90: new treatment of constants
models/atm/cam/src/dynamics/fv/sw_core.F90: new treatment of constants
models/atm/cam/src/dynamics/fv/te_map.F90: new treatment of constants
models/atm/cam/src/dynamics/fv/trac2d.F90: new treatment of constants
models/atm/cam/src/dynamics/fv/tp_core.F90: new treatment of constants, cleaning
models/atm/cam/src/dynamics/fv/xpavg.F90: new treatment of constants
models/atm/cam/src/dynamics/fv/dyn_comp.F90: new treatment of constants, test return
status of allocations
models/atm/cam/src/dynamics/fv/inidat.F90: new treatment of constants, test return
status of allocations, use new naming scheme for scatter/gather comm. patterns
models/atm/cam/src/dynamics/fv/restart_dynamics.F90: new treatment of constants;
use new naming scheme for scatter/gather comm. patterns
models/atm/cam/src/dynamics/fv/io_dist.F90: extensive cleaning. Removed old
methods for writing and reading; scheme is now consistent with
fv_read/write_r4/r8/i4 and fv_scatter/gather_r4/r8/i4 routines.
models/atm/cam/src/dynamics/fv/dynamics_vars.F90: removed (unused) ESMF code
revised naming scheme for gather/scatter comm. patterns now consistent,
added PI argument to dynamics_init, dynpkg_init; XYZ patterns generated
by XY pattern with extension in Z; new treatment of constants
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: Failed on tests 030 (bl336), 034 (bl351), 055 (cs998), probably
because of new treatment of constants, specifically removal of
4.0*atan(1.0) as definition of PI. Perturbation growth tests by
Brian Eaton indicate that these are roundoff differences. Check-in
OK'ed by Mathew Rothstein.
tempest: passes all tests
bangkok/lf95: passes all tests
CAM tag used for the baseline comparison tests if different than previous
tag: cam3_3_33 (previous tag)
Summarize any changes to answers, i.e.,
- what code configurations: WACCM, FV
- what platforms/compilers: Bluesky
- nature of change (roundoff; larger than roundoff but same climate; new
climate): roundoff
If bitwise differences were observed, how did you show they were no worse
than roundoff? perturbation growth test (by Eaton)
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_3_33
Originator(s): eaton
Date: Fri Oct 6 16:22:01 MDT 2006
One-line Summary: Enable linking to external ESMF library
Purpose of changes:
. Changes to configure and Makefile.in to allow linking to ESMF library
This can be done by:
1. setting the new -esmf_libdir option,
or
2. by setting the environment variable ESMF_LIBDIR.
Either of these methods will cause configure to check that the
files libesmf.a or libesmf.so and esmf.mk are present in the specified
directory. The esmf.mk file is produced when the ESMF library is built
and contains Makefile macros that CAM's Makefile will reference when
compiling and linking to the external library.
. Remove interactive mode from configure.
. Cleanup of Makefile.in. Remove untested SUN and OSF1 sections.
Bugs fixed (include bugzilla ID):
. Fix bug in fv/inidat.F90 applying perturbation to initial temperature
field.
. Fixed SCAM build on bangkok.
Describe any changes made to build system:
. Add ability to link to an external ESMF library. The default remains to use
the WRF_ESMF time manager code.
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/configure
. Add capability to specify that an external ESMF library should be used.
. Add test for linking to the ESMF library to test suite invoked by
configure's -test option.
. Add ability to try running a successfully built test. This helps to
identify problems related shared libraries not being found.
. Remove interactive mode.
. Simplify verbose mode; now set by "-v" option without a numeric argument.
models/atm/cam/bld/Makefile.in
. Include the esmf.mk file if ESMF_LIBDIR is defined.
. use ESMF_F90COMPILEPATHS from esmf.mk to set search path for esmf module file .
. Use ESMF_F90LINKPATHS and ESMF_F90ESMFLINKLIBS from esmf.mk to set search
path and library name for linking to external esmf lib.
. Remove 32-bit addressing option for AIX. If this is needed somewhere it
should be re-implemented in configure rather than by requiring editing of the
Makefile.
. Remove unused SUN and OSF1 sections.
. Substantial cleanup. Tried to eliminate duplicated code from the
architecture specific sections of the Makefile. Tried to organize the
architecture specific sections to more or less follow the same order of
setting macros.
models/atm/cam/src/dynamics/fv/inidat.F90
. fix bug in how perturbation was applied to initial temperature field.
models/atm/cam/test/system/test_driver.sh
. increase to bluevista time limit from 2:28 to 4:28 to account for
slowdown in compilation times
. update pathscale compiler version to 2.4 from 2.2.1 (on lightning)
models/atm/cam/tools/scam/ui/configure
. add -lpgc to the LINK_LIBS macro in the Linux section.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: All PASS.
tempest: All PASS.
bangkok/lf95: All PASS.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_3_32
Originator(s): mvertens
Date: Mon Oct 2 21:03:25 MDT 2006
One-line Summary: Have SCAM driver determine the fractional
land cover in any gridcell.
Purpose of changes: remove SCAM if-defs in lnd_comp_mct.F90
and put the logic in SCAM driver.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: erik, eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
M ocn/dom/ocn_comp_mct.F90
- removed SCAM if-def
M ocn/som/ocn_comp_mct.F90
- removed SCAM if-def
M atm/cam/tools/scam/scm_init/init_model.F90
M atm/cam/tools/scam/scm_init/scamMod.F90
- changes enable SCAM to read in landfrac for the gridcell and determine
if the land model should be called
M SVN_EXTERNAL_DIRECTORIES
- updated to clm_exp_73
M drv/seq_mct_drv/seq_ccsm_drv.F90
- formatting change
M lnd/clm2/src/main/lnd_comp_mct.F90
- removed SCAM if-defs
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: ALL TESTS PASSED
tempest: ALL TESTS PASSED
bangkok/lf95: ALL TESTS PASSED
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: NONE
===============================================================
===============================================================
Tag name: cam3_3_31
Originator(s): mvertens
Date: Tue Sep 26 19:48:56 MDT 2006
One-line Summary: made surface components mpi utilities and clm time manager independent of cam
Purpose of changes: cam surface components no longer "use" cam communicator group settings
and clm no longer uses cam time manager when in COUP_CAM mode
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist:
M atm/cam/bld/clm_inparm.pm
M atm/cam/bld/SeqCCSM_namelist.pm
- added dtime to clm namelist input
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton, tcraig, robj
List all subroutines eliminated: none
List all subroutines added and what they do:
A ocn/dom/ocn_spmd.F90
- determine communicator group and masterproc, iam, npes for ocn
A ocn/som/ocn_spmd.F90
- determine communicator group and masterproc, iam, npes for ocn
A ice/csim4/ice_spmd.F90
- determine communicator group and masterproc, iam, npes for ice
A drv/seq_mct/seq_communicator.F90
- determine communicator group and masterproc, iam, npes for drv routines
List all existing files that have been modified, and describe the changes:
M ocn/dom/sst_data.F90
- replace "use mpishorthand" with "use ocn_utils"
- made explicit calls to mpi_bcast rather than the interfaces in mpishorthand
M ocn/dom/ocn_comp.F90
- replace "use spmd_utils" with "use ocn_utils"
- ocn_mpicom now passed in as argument and used to initialize ocn communicator
group, masterproc, iam and npes
M ocn/dom/ocn_comp_mct.F90
- ocn_mpicom now passed in as argument
- replaced call to module routine seq_init_SetgsMap with call to mct routine
mct_gsMap_init in order to initialize gsmap
M ocn/som/mixed_layer.F90
- replace "use mpishorthand" with "use ocn_utils"
- made explicit calls to mpi_bcast rather than the interfaces in mpishorthand
M ocn/som/ocn_comp.F90
- replace "use spmd_utils" with "use ocn_utils"
- ocn_mpicom now passed in as argument and used to initialize ocn communicator
group, masterproc, iam and npes
- fixed bug in conservation check equation (this is turned off in all tests)
ocn_in(c)%netfocn replaced with ocn_in(c)%melth
M ocn/som/somini.F90
- replace "use mpishorthand" with "use ocn_utils"
- made explicit calls to mpi_bcast rather than the interfaces in mpishorthand
M ocn/som/ocn_comp_mct.F90
- ocn_mpicom now passed in as argument
- replaced call to module routine seq_init_SetgsMap with call to mct routine
mct_gsMap_init in order to initialize gsmap
M ice/csim4/ice_data.F90
- replace "use spmd_utils" with "use ice_utils"
- made explicit calls to mpi_bcast rather than the interfaces in mpishorthand
M ice/csim4/print_coverage.F90
- replace "use spmd_utils" with "use ice_utils"
M ice/csim4/ice_comp.F90
- replace "use spmd_utils" with "use ice_utils"
- made explicit calls to mpi_bcast rather than the interfaces in mpishorthand
M ice/csim4/ice_comp_mct.F90
- added ICEID and mpicom_ice to input arguments
- replaced call to module routine seq_init_SetgsMap with call to mct routine
mct_gsMap_init in order to initialize gsmap
M atm/cam/tools/scam/scm_init/init_model.F90
- put in changes analogous to those in seq_ccsm_drv.F90
M atm/cam/tools/scam/scm_init/scam_run.F90
- separated calls to ocn_run_mct and atm/ocn flux calculation into two separate blocks
M atm/cam/tools/scam/scm_init/scamMod.F90
- added dtime_out as argument to scam_clm_default_opts (see clm notes below
M atm/cam/src/control/con_cam.F90
- removed call to cam_init2 (call to cam_init1 now is call to cam_init)
- removed call to spmdinit
- mpicom now passed as argument to cam_init
M atm/cam/src/control/atm_comp_mct.F90
M atm/cam/src/control/cam_comp.F90
- removed cam_init2 and replaced cam_init1 with cam_init
- call to intht is now done on first call of cam_comp
M atm/cam/src/utils/spmd_utils.F90
- no longer calls mpi_init when COUP_CSM ins not defined
- removed code-block for if-defined IRIX (no longer needed)
- mpicom_atm passed as now as argument
M atm/cam/src/utils/time_manager.F90
- removed module variables perpetual_ymd and perpetual_run (no longer needed)
M SVN_EXTERNAL_DIRECTORIES
- now using clm3_expa_73
- clm3_expa_73 no longer uses the cam time manager and as a result dtime needs to be
specified explicitly in the clm namelist
- clm3_expa_73 also no longer obtains irad from the call to the cam routine radiation_get
as a result bugzilla #228 has been added
M drv/seq_mct_drv/seq_ccsm_drv.F90
- removed call to atm_init2
- mpi now initialized in seq_ccsm_drv.F90 rather than in the cam routine spmdinit
- currently it is assumed that cam and all the surface components will utilize the global
communicator group
- mpicom_glob now passed as an argument to cam and the surface components
- replaced calls to mrg_x2*_alloc_mct with mrg_x2*_init_mct, and removed the original calls
to mrg_x2*
M drv/seq_mct/seq_init_mct.F90
- now calls mpi_init (rather than in cam routine spmdinit)
- removed routine seq_init_setgsmap (see above)
M drv/seq_mct/seq_domain_mct.F90
- replaced "use mpishorthand" with "use seq_comunicator"
- removed explicit setting of mpicom=1 when SPMD is not defined
M drv/seq_mct/seq_flux_mct.F90
- replaced "use mpishorthand" with "use seq_comunicator"
- removed explicit setting of mpicom=1 when SPMD is not defined
M drv/seq_mct/mrg_x2a_mct.F90
- replaced "use mpishorthand" with "use seq_comunicator"
- replaced hard-wired copies for each field with mct call to mct_aVect_copy.
M drv/seq_mct/mrg_x2i_mct.F90
- replaced "use mpishorthand" with "use seq_comunicator"
all common attribute vector components from one attribute vector to another.
- replaced hard-wired copies for each field with mct call to mct_aVect_copy (only
rain and snow are now hard-wired in the copy command). mct_aVect_copy will copy
- removed explicit setting of mpicom=1 when SPMD is not defined
M drv/seq_mct/mrg_x2l_mct.F90
- replaced "use mpishorthand" with "use seq_comunicator"
- removed explicit setting of mpicom=1 when SPMD is not defined
- replaced hard-wired copies for each field with mct call to mct_aVect_copy.
M drv/seq_mct/mrg_x2o_mct.F90
- replaced "use mpishorthand" with "use seq_comunicator"
- removed explicit setting of mpicom=1 when SPMD is not defined
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: ALL PASS
tempest: ALL PASS
bangkok/lf95: ALL PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: NONE
===============================================================
===============================================================
Tag name: cam3_3_30
Originator(s): Jim Edwards
Pat Worley (phys_grid)
Date: 09-21-2006
One-line Summary: homme dycore development
Purpose of changes: homme dycore development
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: added build for bluegene
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: Edwards, Eaton
List all subroutines eliminated:
List all subroutines added and what they do: added dynamics/homme/external
List all existing files that have been modified, and describe the changes:
Improved documentation for creating initial and boundary files
for the homme dycore
M models/atm/cam/tools/interpic_new/README
Added support for bluegene system at NCAR (frost)
and improved support for homme dycore.
M models/atm/cam/bld/configure
M models/atm/cam/bld/Makefile.in
M models/atm/cam/bld/namelist.pm
A + models/atm/cam/bld/filter_nl.pm
M models/atm/cam/bld/DefaultCAM_INPARM_Namelist.xml
M models/atm/cam/bld/system_defaults.xml
A + models/atm/cam/bld/DefaultCTL_NL_Namelist.xml
M models/atm/cam/bld/cam_inparm.pm
M models/atm/cam/bld/resolution_parameters.xml
A + models/atm/cam/bld/ctl_nl.pm
M models/atm/cam/bld/config_cam_homme_defaults.xml
M models/atm/cam/bld/SeqCCSM_namelist.pm
A + models/atm/cam/bld/DefaultFILTER_NL_Namelist.xml
Improved the interface that allows reading multiple namelists from multiple source code files.
Fixed a problem with the dyn_grid -> phys_grid mapping apparent in the homme dycore.
M models/atm/cam/src/control/runtime_opts.F90
M models/atm/cam/src/control/con_cam.F90
M models/atm/cam/src/control/atm_comp_mct.F90
M models/atm/cam/src/control/cam_comp.F90
M models/atm/cam/src/control/cam_history.F90
M models/atm/cam/src/physics/cam1/boundarydata.F90
M models/atm/cam/src/physics/cam1/physpkg.F90
M models/atm/cam/src/physics/cam1/phys_grid.F90
M models/atm/cam/src/physics/cam1/cldwat.F90
M models/atm/cam/src/dynamics/fv/inital.F90
Continued development of the homme dycore: the external directory represents a
mirror of the files in repository
https://svn-homme-model.cgd.ucar.edu/trunk/src/share
A + models/atm/cam/src/dynamics/homme/external
A + models/atm/cam/src/dynamics/homme/external/flops_mod.F90
A + models/atm/cam/src/dynamics/homme/external/filter_mod.F90
A + models/atm/cam/src/dynamics/homme/external/math_constants.F90
A + models/atm/cam/src/dynamics/homme/external/forcing_mod.F90
A + models/atm/cam/src/dynamics/homme/external/stats.h
A + models/atm/cam/src/dynamics/homme/external/preq_init_mod.F90
A + models/atm/cam/src/dynamics/homme/external/interface.h
A + models/atm/cam/src/dynamics/homme/external/schedule_mod.F90
A + models/atm/cam/src/dynamics/homme/external/reduction_mod.F90
A + models/atm/cam/src/dynamics/homme/external/parallel_mod.F90
A + models/atm/cam/src/dynamics/homme/external/spacecurve_mod.F90
A + models/atm/cam/src/dynamics/homme/external/utils_mod.F90
A + models/atm/cam/src/dynamics/homme/external/derivative_mod.F90
A + models/atm/cam/src/dynamics/homme/external/prim_state_mod.F90
A + models/atm/cam/src/dynamics/homme/external/dimensions_mod.F90
A + models/atm/cam/src/dynamics/homme/external/locate.F90
A + models/atm/cam/src/dynamics/homme/external/prim_restart_mod.F90
A + models/atm/cam/src/dynamics/homme/external/torus_mod.F90
A + models/atm/cam/src/dynamics/homme/external/quadrature_mod.F90
A + models/atm/cam/src/dynamics/homme/external/coordinate_systems_mod.F90
A + models/atm/cam/src/dynamics/homme/external/global_norms_mod.F90
A + models/atm/cam/src/dynamics/homme/external/control_mod.F90
A + models/atm/cam/src/dynamics/homme/external/element_mod.F90
A + models/atm/cam/src/dynamics/homme/external/time_mod.F90
A + models/atm/cam/src/dynamics/homme/external/prim_si_ref_mod.F90
A + models/atm/cam/src/dynamics/homme/external/stats.F90
A + models/atm/cam/src/dynamics/homme/external/hybvcoord_mod.F90
A + models/atm/cam/src/dynamics/homme/external/checksum_mod.F90
A + models/atm/cam/src/dynamics/homme/external/prim_advection_mod.F90
A + models/atm/cam/src/dynamics/homme/external/directions.F90
A + models/atm/cam/src/dynamics/homme/external/thread_mod.F90
A + models/atm/cam/src/dynamics/homme/external/dof_mod.F90
A + models/atm/cam/src/dynamics/homme/external/physical_constants.F90
A + models/atm/cam/src/dynamics/homme/external/kinds.F90
A + models/atm/cam/src/dynamics/homme/external/solver_mod.F90
A + models/atm/cam/src/dynamics/homme/external/restart_io_mod.F90
A + models/atm/cam/src/dynamics/homme/external/namelist_mod.F90
A + models/atm/cam/src/dynamics/homme/external/perfmodel_mod.F90
A + models/atm/cam/src/dynamics/homme/external/bndry_mod.F90
A + models/atm/cam/src/dynamics/homme/external/metis_mod.F90
A + models/atm/cam/src/dynamics/homme/external/params_mod.F90
A + models/atm/cam/src/dynamics/homme/external/field_mod.F90
A + models/atm/cam/src/dynamics/homme/external/diffusion_mod.F90
A + models/atm/cam/src/dynamics/homme/external/mass_matrix_mod.F90
A + models/atm/cam/src/dynamics/homme/external/prim_advance_mod.F90
A + models/atm/cam/src/dynamics/homme/external/cg_mod.F90
A + models/atm/cam/src/dynamics/homme/external/prim_si_mod.F90
A + models/atm/cam/src/dynamics/homme/external/types_mod.F90
A + models/atm/cam/src/dynamics/homme/external/domain_mod.F90
A + models/atm/cam/src/dynamics/homme/external/gridgraph_mod.F90
A + models/atm/cam/src/dynamics/homme/external/metagraph_mod.F90
A + models/atm/cam/src/dynamics/homme/external/generic_list.F90
A + models/atm/cam/src/dynamics/homme/external/ref_state_mod.F90
A + models/atm/cam/src/dynamics/homme/external/vertex_mod.F90
A + models/atm/cam/src/dynamics/homme/external/cube_mod.F90
A + models/atm/cam/src/dynamics/homme/external/physics_mod.F90
A + models/atm/cam/src/dynamics/homme/external/rotation_init.F90
A + models/atm/cam/src/dynamics/homme/external/interpolate_mod.F90
A + models/atm/cam/src/dynamics/homme/external/linear_algebra_mod.F90
A + models/atm/cam/src/dynamics/homme/external/baroclinic_inst_mod.F90
A + models/atm/cam/src/dynamics/homme/external/edge_mod.F90
A + models/atm/cam/src/dynamics/homme/external/ll_mod.F90
A + models/atm/cam/src/dynamics/homme/external/vertical_mod.F90
A + models/atm/cam/src/dynamics/homme/external/timer.h
A + models/atm/cam/src/dynamics/homme/external/timer_mod.F90
A + models/atm/cam/src/dynamics/homme/external/state_mod.F90
A + models/atm/cam/src/dynamics/homme/external/hybrid_mod.F90
A + models/atm/cam/src/dynamics/homme/io_dist.F90
M models/atm/cam/src/dynamics/homme/dyn_grid.F90
M models/atm/cam/src/dynamics/homme/dp_coupling.F90
M models/atm/cam/src/dynamics/homme/dyn_comp.F90
M models/atm/cam/src/dynamics/homme/inidat.F90
M models/atm/cam/src/dynamics/homme/stepon.F90
M models/atm/cam/src/dynamics/homme/spmd_dyn.F90
M models/atm/cam/src/dynamics/homme/inital.F90
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: passed
tempest: passed
bangkok/lf95: passed
CAM tag used for the baseline comparison tests if different than previous
tag: cam3_2_28
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
Tag name: cam3_3_29
Originator(s): jwolfe, eaton
Date: Wed Sep 20 17:08:08 MDT 2006
One-line Summary: fix for single executable CCSM
Purpose of changes: fix for single executable CCSM
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
In con_cam.F90 change this:
#ifdef SINGLE_EXEC
call MPH_get_argument("THREADS", nthreads, "atm")
call OMP_SET_NUM_THREADS(nthreads)
#endif
to this:
#ifdef SINGLE_EXEC
call MPH_get_argument("THREADS", nThreads, "atm")
#ifdef _OPENMP
call OMP_SET_NUM_THREADS(nThreads)
#endif
#endif
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: not done
tempest: not done
bangkok/lf95: not done
Tests not done because this change is inside an ifdef that isn't turned on
for any of CAM's current regression tests.
===============================================================
===============================================================
Tag name: cam3_3_28
Originator(s): mvr, jwolfe
Date: 060914
One-line Summary: implemented last remaining items needed for cam to
be included in a ccsm single-executable; bug fix to enable scam to compile
Purpose of changes: ccsm single-executable required a few more mods,
including one name conflict that was overlooked; a compile-time bug was
introduced to the scam build in cam3_3_27
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: myself, eaton, mvertens
List all subroutines eliminated:
D models/atm/cam/src/physics/cam1/diagnostics.F90
- just a rename required to clear up name conflict in ccsm single-executable
List all subroutines added and what they do:
A + models/atm/cam/src/physics/cam1/cam_diagnostics.F90
- newly named file and mods to clear up name conflict in ccsm single-executable
List all existing files that have been modified, and describe the changes:
M models/ice/csim4/ice_comp.F90
M models/atm/cam/src/control/runtime_opts.F90
M models/atm/cam/src/control/con_cam.F90
M models/atm/cam/src/physics/cam1/tphysbc.F90
M models/atm/cam/src/physics/cam1/physpkg.F90
- files modified to reflect name change: diagnostics -> cam_diagnostics
M models/atm/cam/tools/scam/scm_init/init_model.F90
- bug fix to enable scam to compile
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky:
ALL PASS
tempest:
ALL PASS
bangkok/lf95:
ALL PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_3_27
Originator(s): mvertens
Date: Mon Sep 11 21:44:10 MDT 2006
One-line Summary: moved atm/ocn flux calculation to top level
Purpose of changes:
Moved the atmosphere/ocean flux calculation out of the ocn (dom/som)
models and up to the top level driver. This is a necessary stop in constructing
a sequential ccsm system. It will also enable the use of the cpl6 atm/ocn flux
calculation to be utilized in both the sequential and concurrent system.
Bugs fixed (include bugzilla ID): None
Describe any changes made to build system: None
Describe any changes made to the namelist: None
List any changes to the defaults for the boundary datasets: None
Describe any substantial timing or memory changes: Probably none (this should be checked)
Code reviewed by: eaton, erik (briefly)
List all subroutines eliminated:
D ocn/dom/albocean.F90
- calculation now done in seq_flux_mct.F90 (see below)
D ocn/dom/print_coverage.F90
- not used by any code in dom/
D ocn/dom/parpbl.h
- only used in flxoce.F90 - which is no longer utilized
D ocn/dom/albedo.h
- not used by any code in dom/
D ocn/dom/wtrc_flxoce.F90
- moved to atm/cam/src/control
D ocn/dom/flxoce.F90
- replaced by flux_ao routine (see below)
D ocn/dom/srfoce.F90
- see below
D ocn/som/albocean.F90
- calculation now done in seq_flux_mct.F90 (see below)
D ocn/som/mixed_layer_globalcalcs.F90
D ocn/som/ocn_srf.F90
- ocn_srf.F90 and mixed_layer_globalcalcs.F90 merged into mixed_layer.F90
D ocn/som/parpbl.h
- only used in flxoce.F90 - which is no longer utilized
D ocn/som/flxoce.F90
- replaced by flux_ao routine (see below)
List all subroutines added and what they do:
A ocn/som/mixed_layer.F90
- merge of mixed_layer_globalcalcs.F90 and ocn_srf.F90
A atm/cam/src/control/wtrc_flxoce.F90
- moved computation of water tracer exchange from ocean into atmospheric code base
to reduce number of fields that need to be exchanged between ocean and atmosphere
A drv/seq_mct/seq_flux_mct.F90
A drv/seq_mct/flux_ao.F90
- performs calculation previously done in dom/flxocn.F90 and som/flxocn.F90
- also calculates ocean albedo and net shortwace absorbed by surface
List all existing files that have been modified, and describe the changes:
M ocn/dom/sst_data.F90
M ocn/dom/ocn_types.F90
M ocn/dom/ocn_comp.F90
M ocn/dom/ocn_comp_mct.F90
- see below
M ocn/som/ocn_types.F90
M ocn/som/ocn_comp.F90
M ocn/som/ocn_comp_mct.F90
- see below
M atm/cam/tools/scam/scm_init/init_model.F90
M atm/cam/tools/scam/scm_init/scam_srfdata_MCT.F90
M atm/cam/tools/scam/scm_init/scam_run.F90
- put in changes to incorporate xao_o and xao_a along with calls to ocean/atmosphere
flux calculation from top level driver
M atm/cam/src/control/ccsm_msg.F90
M atm/cam/src/control/camsrfexch_types.F90
- added "rho, netsw, ustar, re, ssq" as cam_out components in camsrfexch_types.F90
M atm/cam/src/control/srfxfer.F90
- removed rho and netsw as module variables in ccsm_msg.F90
M atm/cam/src/control/atm_comp_mct.F90
- introduced call to wtrc_flxoce if trace_water is true
- replace call to atm_init1_mct with atm_init_mct
- added ustar, re and ssq to x1a_a input
M drv/seq_flds/dust/seq_flds_mod.F90
M drv/seq_flds/default/seq_flds_mod.F90
M drv/seq_flds/gensom/seq_flds_mod.F90
- see below
M drv/seq_mct_drv/seq_ccsm_drv.F90
M drv/seq_mct/mrg_x2a_mct.F90
- replaced o2x_a components that are now calculated in seq_flux_mct with xao_a components
M drv/seq_mct/mrg_x2i_mct.F90
- removed index_o2x_So_tsocn from o2x_i and x2i_i
M drv/seq_mct/mrg_x2o_mct.F90
- removed unnecessary a2x_o (this is now utilized in atm/ocn flux calculatin in seq_flux_mct)
M drv/seq_flds_indices/seq_flds_indices.F90
- put in changes necessary for new atm/ocn flux computation at top level (see below)
Summary: New attribute vectors xao_o and xao_a were introduced at the top level
to account for the calculation of the atm/ocn flux calculation.
The routines dom/srfoce.F90, dom/flxoce.F90 and som/flxoce.F90 were replaced with
drv/seq_mct/flux_ao.F90 and drv/seq_mct/seq_flux_mct.F90.
The current form of flx_ao.F90 is such that it can be replaced with
the equivalent coupler code once sign conventions are addressed. (Note,
to increase clarity som/ocn_srf.F90 and som/mixed_layer_globalcalcs.F90 were
merged to create som/mixed_layer.F90).
The ocean albedo calculation was also moved out of the ocean code and
put into seq_flux_mct.F90. As a result, ocn/som/albocean.F90 and
ocn/dom/albedo.h are no longer needed). The new routine to calculate ocean albedos,
seq_flux_albo_mct, is now contained in seq_flux_mct.F90.
Ocean input information was significantly altered as a result of bringing the atm/ocn
flux calculation out of the ocean code base. The CAM DOM code now does not need any
input information. The CAM SOM code also requires significantly less input information
(only netsw from the atmosphere).
It is important to note that lwup is still being computed in the ocean code. The
impact on SOM needs to be evaluated when moving this calculation to the
atm/ocean flux routine (as is done in the cpl6 code).
The routine, wtrc_flxoce.F90 (isotope calculation) was moved out of the ocean model
and into atm/cam/src/control. The extra fields necessary to do this
computation (ustar, re, ssq) were added to the x2a attribute vector.
Note that this feature is currently not tested.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: ALL PASSED
tempest: ALL PASSED
bangkok/lf95: ALL PASSED except for
012 bl153 TBL.sh e64m outfrq24h -2 ................................FAIL! rc= 7
028 bl353 TBL.sh f2m outfrq24h -2 .................................FAIL! rc= 7
040 bl553 TBL.sh s64m outfrq24h -2 ................................FAIL! rc= 7
- the above difference appear to be roundoff errors that set in very early
- note that all other tests (including on all other platforms) produced bfb
baseline comparisons
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: NONE
===============================================================
===============================================================
Tag name: cam3_3_26
Originator(s): mvertens
Date: Sun Sep 10 20:22:42 MDT 2006
One-line Summary: Changes necessary for moving atm/ocn flux calc to top level
Purpose of changes:
Incorporated changes necessary to move the atm/ocn flux calculation out of the dom/som
ocean code and into the top level application driver.
Bugs fixed (include bugzilla ID): None
Describe any changes made to build system: None
Describe any changes made to the namelist: None
List any changes to the defaults for the boundary datasets: None
Describe any substantial timing or memory changes: None
Code reviewed by: eaton
List all subroutines eliminated: None
List all subroutines added and what they do: None
List all existing files that have been modified, and describe the changes:
M ocn/dom/ocn_comp.F90
- Moved update of ocean temperature to after calculation of ocean surface/fluxes
This was done to be consistent with SOM update and to be able to move the atm/ocean
surface flux calculation to the top level driver without introducint cpp ifdefs.
This change also required writing the surface temperature to the ocean restart file.
This modification will result in greater than round-off level differences with cam3_3_25.
- removed the call to sstan and made the conversion from degees K to C explicit
M ice/csim4/ice_types.F90
M ice/csim4/ice_comp.F90
M ice/csim4/ice_diagnostics.F
- Removed "precsc, precsl, precc, precl" as ice_in types and replaced them with "snow"
since only snow is needed as input to csim. This change should only result in round-off
level differences with cam3_3_25.
M ice/csim4/ice_comp_mct.F90
- In addition to only using "snow" as input from the application driver, also removed
x2i_i%rAttr(index_x2i_So_tsocn,g) as input and instead used
x2i_i%rAttr(index_x2i_So_t,g)-SHR_CONST_TKFRZ. This change should only result in
round-off level differences with cam3_3_25.
M drv/seq_flds/dust/seq_flds_mod.F90
M drv/seq_flds/default/seq_flds_mod.F90
M drv/seq_flds/gensom/seq_flds_mod.F90
M drv/seq_mct/mrg_x2i_mct.F90
M drv/seq_flds_indices/seq_flds_indices.F90
- replaced 'Faxa_rainc', Faxa_rainl','Faxa_snowc','Faxa_snowl' as input to ice
with 'Faxa_rain' and :Faxa_snow'
M drv/seq_mct/mrg_x2a_mct.F90
- Removed calculation of ts for non-fractional grid box
This makes the merge consistent with that done in cpl6.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: All tests passed (no baseline compare was done)
tempest: All tests passed (no baseline compare was done)
bangkok/lf95: All tests passed (no baseline compare was done)
Summarize any changes to answers, i.e.,
- what code configurations: see above
- what platforms/compilers: effects all platforms
- nature of change: larger than roundoff but same climate
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository): cam3_3_26
- platform/compilers: bluesky
- configure commandline:
configure -ocn dom -dyn fv -spmd -smp -test -res 4x5
- build-namelist command (or complete namelist):
&ccsm_inparm
case_name = 'cam3_3_24_fluxao_fv4x5dom'
start_type = "continue"
/
&timemgr_inparm
atm_cpl_dt = 1800
orb_iyear_ad = 1950
restart_option = 'monthly'
start_ymd = 10101
stop_n = 720
stop_option = 'ndays'
/
&cam_inparm
dtime = 1800
absems_data = '/fis/cgd/cseg/csm/inputdata/atm/cam/rad/abs_ems_factors_fastvx.c030508.nc'
aeroptics = '/fis/cgd/cseg/csm/inputdata/atm/cam/rad/AerosolOptics_c050419.nc'
bnd_topo = '/fis/cgd/cseg/csm/inputdata/atm/cam/topo/USGS-gtopo30_4x5_remap_c050520.nc'
bndtvaer = '/fis/cgd/cseg/csm/inputdata/atm/cam/rad/AerosolMass_V_4x5_clim_c031022.nc'
bndtvo = '/fis/cgd/cseg/csm/inputdata/atm/cam/ozone/pcmdio3.r8.64x1_L60_clim_c970515.nc'
bndtvs = '/fis/cgd/cseg/csm/inputdata/atm/cam/sst/sst_HadOIBl_bc_4x5_clim_c030228.nc'
ncdata = '/fis/cgd/cseg/csm/inputdata/atm/cam/inic/fv/cami_0001-01-01_4x5_L26_c060608.nc'
/
&clm_inparm
fpftcon = '/fis/cgd/cseg/csm/inputdata/lnd/clm2/pftdata/pft-physiology-cn16.c040719'
fsurdat = '/fis/cgd/cseg/csm/inputdata/lnd/clm2/srfdata/cam/clms_3.1_4x5_c050523.nc'
/
- MSS location of output:
/MVERTENS/csm/cam3_3_24_fluxao_fv4x5dom/hist (cam3_3_24+ current mods)
MSS location of control simulations used to validate new climate:
/MVERTENS/csm/cam3_3_24_fluxao_fv4x5dom_0/hist (cam3_3_24)
URL for AMWG diagnostics output used to validate new climate:
http://www.cgd.ucar.edu/cms/rneale/runs/cam/cam3_3_24_fluxao_fv4x5dom-cam3_3_24_fluxao_fv4x5dom_0/
===============================================================
===============================================================
Tag name: cam3_3_25
Originator(s): eaton
Date: Tue Sep 5 08:53:01 MDT 2006
One-line Summary: add spun-up CLM initial files
Purpose of changes:
. add spun-up CLM initial files for FV 1.9x2.5. There are files for both
0000-01-01 and 0000-09-01
. Fixed syntax of some attributes in the default XML files. XML requires
that attribute values are quoted, and there were unquoted values in the
files.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets:
. add spun-up CLM initial files for FV 1.9x2.5. There are files for both
0000-01-01 and 0000-09-01
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
. models/atm/cam/bld/DefaultCAM_INPARM_Namelist.xml
. models/atm/cam/bld/DefaultCLM_INPARM_Namelist.xml
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: All PASS.
tempest: All PASS.
bangkok/lf95: All PASS.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: NONE
===============================================================
===============================================================
Tag name: cam3_3_24
Originator(s): mvr, jwolfe
Date: 060830
One-line Summary: mods to eliminate naming conflicts in the ccsm
single executable; increased max fields for history tapes;
minor test driver enhancements
Purpose of changes: the move to a single executable for running
cam as part of ccsm introduced some naming conflicts that needed
to be resolved; running cam in ccsm and turning on the chemistry
and co2 blew out the maximum allowed fields on a history tape
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: myself, eaton
List all subroutines eliminated:
D models/atm/cam/src/control/history.F90
- renamed due to conflict in the ccsm single executable
List all subroutines added and what they do:
A + models/atm/cam/src/control/cam_history.F90
- newly named to work with the ccsm single executable
- also modified the maximum number of fields allowed on history tapes
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/TSB.ccsm.sh
- modified to use resource settings returned from ccsm scripts
- ccsm tests now use default queue on phoenix
M models/atm/cam/test/system/test_driver.sh
- machine calgary now supported by test driver;
- increased wall clock limit for default testing on bluevista
M models/atm/cam/test/system/CAM_runcmnd.sh
- added code in support of machine calgary
M models/ocn/dom/ocn_comp.F90
M models/ocn/som/ocn_srf.F90
M models/ocn/som/ocn_comp.F90
M models/ice/csim4/ice_srf.F90
M models/ice/csim4/ice_comp.F90
M models/atm/cam/tools/scam/scm_init/init_model.F90
M models/atm/cam/tools/scam/scm_init/forecast.F90
M models/atm/cam/src/control/history_defaults.F90
M models/atm/cam/src/control/runtime_opts.F90
M models/atm/cam/src/control/ccsm_msg.F90
M models/atm/cam/src/control/srfxfer.F90
M models/atm/cam/src/control/restart.F90
M models/atm/cam/src/control/atm_comp_mct.F90
M models/atm/cam/src/control/filenames.F90
M models/atm/cam/src/control/history_scam.F90
M models/atm/cam/src/control/cam_comp.F90
M models/atm/cam/src/utils/time_manager.F90
M models/atm/cam/src/physics/cam1/seasalt_intr.F90
M models/atm/cam/src/physics/cam1/tphysidl.F90
M models/atm/cam/src/physics/cam1/progseasalts_intr.F90
M models/atm/cam/src/physics/cam1/tracers.F90
M models/atm/cam/src/physics/cam1/param_cldoptics.F90
M models/atm/cam/src/physics/cam1/diagnostics.F90
M models/atm/cam/src/physics/cam1/radiation.F90
M models/atm/cam/src/physics/cam1/dust_intr.F90
M models/atm/cam/src/physics/cam1/tphysbc.F90
M models/atm/cam/src/physics/cam1/aerosol_radiation_interface.F90
M models/atm/cam/src/physics/cam1/radsw.F90
M models/atm/cam/src/physics/cam1/check_energy.F90
M models/atm/cam/src/physics/cam1/stratiform.F90
M models/atm/cam/src/physics/cam1/water_tracers.F90
M models/atm/cam/src/physics/cam1/radlw.F90
M models/atm/cam/src/physics/cam1/ozone_data.F90
M models/atm/cam/src/physics/cam1/physpkg.F90
M models/atm/cam/src/physics/cam1/prescribed_aerosols.F90
M models/atm/cam/src/physics/cam1/convect_shallow.F90
M models/atm/cam/src/physics/cam1/convect_deep.F90
M models/atm/cam/src/physics/cam1/carbon_intr.F90
M models/atm/cam/src/physics/cam1/sulchem.F90
M models/atm/cam/src/physics/cam1/co2_cycle.F90
M models/atm/cam/src/physics/cam1/chemistry.F90
M models/atm/cam/src/physics/cam1/gw_drag.F90
M models/atm/cam/src/physics/cam1/sulemis.F90
M models/atm/cam/src/physics/cam1/aerosol_intr.F90
M models/atm/cam/src/physics/cam1/cloudsimulator.F90
M models/atm/cam/src/physics/cam1/zm_conv.F90
M models/atm/cam/src/physics/cam1/sulfur_intr.F90
M models/atm/cam/src/physics/cam1/constituent_burden.F90
M models/atm/cam/src/physics/cam1/vertical_diffusion.F90
M models/atm/cam/src/physics/waccm/iondrag.F90
M models/atm/cam/src/physics/waccm/tracers.F90
M models/atm/cam/src/physics/waccm/nlte_lw.F90
M models/atm/cam/src/physics/waccm/chemistry.F90
M models/atm/cam/src/physics/waccm/ctem.F90
M models/atm/cam/src/physics/waccm/radheat.F90
M models/atm/cam/src/physics/waccm/gw_drag.F90
M models/atm/cam/src/physics/waccm/tgcm_forcing.F90
M models/atm/cam/src/chemistry/trop_mozart/mo_gas_phase_chemdr.F90
M models/atm/cam/src/chemistry/trop_mozart/mo_lightning.F90
M models/atm/cam/src/chemistry/trop_mozart/mo_chemini.F90
M models/atm/cam/src/chemistry/trop_mozart/mo_aerosols.F90
M models/atm/cam/src/chemistry/trop_mozart/chemistry.F90
M models/atm/cam/src/chemistry/trop_mozart/mo_seasalt.F90
M models/atm/cam/src/chemistry/waccm_mozart/mo_aero_settling.F90
M models/atm/cam/src/chemistry/waccm_mozart/mo_airglow.F90
M models/atm/cam/src/chemistry/waccm_mozart/mo_jlong.F90
M models/atm/cam/src/chemistry/waccm_mozart/mo_setext.F90
M models/atm/cam/src/chemistry/waccm_mozart/mo_gas_phase_chemdr.F90
M models/atm/cam/src/chemistry/waccm_mozart/mo_waccm_photo.F90
M models/atm/cam/src/chemistry/waccm_mozart/mo_jshort.F90
M models/atm/cam/src/chemistry/waccm_mozart/mo_cph.F90
M models/atm/cam/src/chemistry/waccm_mozart/mo_chm_diags.F90
M models/atm/cam/src/chemistry/waccm_mozart/mo_aurora.F90
M models/atm/cam/src/chemistry/waccm_mozart/mo_apex.F90
M models/atm/cam/src/chemistry/waccm_mozart/chemistry.F90
M models/atm/cam/src/chemistry/waccm_mozart/iondrag.F90
M models/atm/cam/src/chemistry/waccm_mozart/mo_waccm_hrates.F90
M models/atm/cam/src/chemistry/waccm_mozart/mo_jeuv.F90
M models/atm/cam/src/chemistry/waccm_mozart/exbdrift.F90
M models/atm/cam/src/dynamics/sld/spegrd.F90
M models/atm/cam/src/dynamics/sld/linemsdyn.F90
M models/atm/cam/src/dynamics/sld/dyn_comp.F90
M models/atm/cam/src/dynamics/sld/diag_dynvar_ic.F90
M models/atm/cam/src/dynamics/sld/tfilt_massfix.F90
M models/atm/cam/src/dynamics/eul/spegrd.F90
M models/atm/cam/src/dynamics/eul/linemsdyn.F90
M models/atm/cam/src/dynamics/eul/dyn_comp.F90
M models/atm/cam/src/dynamics/eul/diag_dynvar_ic.F90
M models/atm/cam/src/dynamics/eul/tfilt_massfix.F90
M models/atm/cam/src/dynamics/fv/inidat.F90
M models/atm/cam/src/dynamics/fv/dyn_comp.F90
M models/atm/cam/src/dynamics/fv/diag_dynvar_ic.F90
M models/atm/cam/src/dynamics/fv/uv3s_update.F90
- mods to eliminate naming conflicts when used in the ccsm single executable:
(mods included 'use' statements, documentation, and whitespace formatting)
subroutine write_restart -> cam_write_restart
subroutine read_restart -> cam_read_restart
module history -> cam_history
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky:
ALL PASS
tempest:
ALL PASS
bangkok/lf95:
ALL PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_3_23
Originator(s): eaton
Date: Sun Aug 27 14:17:37 MDT 2006
One-line Summary: fix some intent attributes in various high level interfaces
Purpose of changes:
. Some of the intents for derived types that contain pointer components are
wrong in the FV interface routines. In this situation the intent refers to
the association status of the pointer, not to the target of the pointer.
Fix these intents in fv/stepon.F90 and fv/dyn_comp.F90.
. cam_run1 has args cam_in and cam_out declared intent(inout). Most of the
components of these user defined types are arrays of length pcols, but
we're adding components that are pointers, and probably all the
components should be pointers. Since for pointer components the
intent(inout) doesn't make sense, and for non-pointer components it's the
default, this intent has been removed. Also switched the order of these
args to match the convention of listing input args first.
. replace the variables surface_state2d and srfflx_state2d by cam_out and
cam_in respectively in physpkg. There are alot of other places where
this still needs to happen.
. utils/timing/f_wrappers.c was missing the header file string.h.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/utils/timing/f_wrappers.c
. add missing include of string.h
models/atm/cam/src/dynamics/fv/stepon.F90
. remove the intent(out) attributes of dummy args dyn_in and dyn_out in
both stepon_init and stepon_run1. That intent
implies that the pointer components should be associated in this
routine. But the pointers were previously associated in
dyn_create_interface and the targets are what is being set here.
models/atm/cam/src/dynamics/fv/dyn_comp.F90
. remove the intent(inout) attributes of dummy args dyn_in and dyn_out in
dyn_run
models/atm/cam/src/control/cam_comp.F90
. remove the intent(inout) attribute of args cam_in and cam_out in
cam_run1 interface. Also switched the order of these
args to match the convention of listing input args first.
. order actual args (.., cam_in, cam_out) in call to phys_run1
models/atm/cam/src/control/con_cam.F90
. switch order of actual args in call to cam_run1
models/atm/cam/src/control/atm_comp_mct.F90
. switch order of actual args in call to cam_run1
models/atm/cam/src/physics/cam1/physpkg.F90
. replace surface_state2d by cam_out and replace srfflx_state2d by cam_in
. order dummy args (.., cam_in, cam_out) in definition of phys_run1
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: All PASS.
tempest: All PASS.
bangkok/lf95: All PASS.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: NONE
===============================================================
===============================================================
Tag name: cam3_3_22
Originator(s): erik
Date: Fri Aug 18 10:18:32 MDT 2006
One-line Summary: fix restart_option=end, get_perp_date bugs
Purpose of changes:
Bugs fixed (include bugzilla ID): 107, 185
Describe any changes made to build system: None
Describe any changes made to the namelist: None
List any changes to the defaults for the boundary datasets: None
Describe any substantial timing or memory changes: None
Code reviewed by: self
List all subroutines eliminated: None
List all subroutines added and what they do: None
List all existing files that have been modified, and describe the changes:
M cam/test/unit/control/run_time_test
Add tests for perpetual case and other restart_options
M atm/cam/test/unit/control/test_time.F90
Add tests for perpetual case and other restart_options
M atm/cam/src/utils/time_manager.F90
Add ability to use offset optional argument for get_perp_date
Externals updated:
csm_share to share3_060814
clm2 to clm3_expa_69
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: All PASS
tempest: All PASS
bangkok/lf95: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: bit-for-bit
===============================================================
===============================================================
Tag name: cam3_3_21
Originator(s): eaton, klindsay, mirin
Date: 2006-08-17
One-line Summary: Update icesst tools; add column burden diags; FV fixes/cleanup.
Purpose of changes:
. Update the icesst tool to recognize the new datasets being provided by
Dennis Shea. These have a CF-conforming time coordinate and the date
information in YYYYMMDD format is moved to a new date variable.
Also needed to change some fixed sizes in bcgen/solver.f90 to allow for
the new datasets starting from 1870 rather than 1949.
. Add hist_fld_active query function to history.F90. Returns whether or
not a field is active on any history file. This enables the ability to
not perform expensive diagnostic calculations if they aren't being asked
for in the history output. (Contributed by Keith Lindsay.)
. Add column burden diagnostics for all constituents except water vapor.
These are not put on the history tapes by default. The names of the
output variables are of the form 'TM'//cnst_name. Not enabled for SCAM.
(Contributed by Keith Lindsay.)
. Add modcomm_gatscat to namelist (bugfix); fix defaults for some modcomm
options; cleanup of comments. (Contributed by Art Mirin.)
Bugs fixed (include bugzilla ID):
. Add modcomm_gatscat to namelist
Describe any changes made to build system: none
Describe any changes made to the namelist:
. Add modcomm_gatscat to namelist
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
models/atm/cam/tools/icesst/README.bfb
List all subroutines added and what they do:
. column burden diagnostics for all constituents except water vapor
models/atm/cam/src/physics/cam1/constituent_burden.F90
List all existing files that have been modified, and describe the changes:
. enable icesst tools to read new input datasets
models/atm/cam/tools/icesst/README
models/atm/cam/tools/icesst/bcgen/solver.f90
models/atm/cam/tools/icesst/regrid/regrid.f90
models/atm/cam/tools/icesst/regrid/wrap_nf.f90
. add column burden diagnostics
models/atm/cam/src/physics/cam1/diagnostics.F90
. add new query function to history
models/atm/cam/src/control/history.F90
. FV and modcomm bugfixes and cleanup
models/atm/cam/src/control/runtime_opts.F90
models/atm/cam/src/dynamics/fv/dyn_comp.F90
models/atm/cam/src/dynamics/fv/spmd_dyn.F90
models/utils/pilgrim/mod_comm.F90
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: All PASS
tempest: All PASS
bangkok/lf95: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: NONE
===============================================================
===============================================================
Tag name: cam3_3_20
Originator(s): erik
Date: Wed Aug 16 09:34:33 MDT 2006
On-line Summary: Fix bugs from cam3_3_15 for working on phoenix
Purpose of changes: Changes needed for CCSM to work on phoenix.
Bugs fixed (include bugzilla ID): 204 and 205
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: self, jwolfe
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/unit/control/run_time_test
Remove old style namelists for comparision, add netcdf directories
explicitly for bangkok.
M models/atm/cam/src/control/con_cam.F90
Move start of timers to earlier in program.
M models/SVN_EXTERNAL_DIRECTORIES
Update to share3_060814. Fixes problem in shr_file_mod for phoenix.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: all tests pass
tempest: all tests pass
bangkok/lf95: all tests pass
Summarize any changes to answers: bit-for-bit
===============================================================
===============================================================
Tag name: cam3_3_19
Originator(s): Jim Edwards (homme, history, phys_grid),
Rory Kelly (homme),
Pat Worley (phys_grid)
Date: 8-14-2006
One-line Summary: Moved dynamic core specific functions from history
to specific cores.
Purpose of changes: Continuing work on general dynamics/physics API
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: None
Describe any changes made to the namelist: None
List any changes to the defaults for the boundary datasets: None
Describe any substantial timing or memory changes: None
Code reviewed by: Originators, Brian Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
The dynamics fields that were previously added to the history output from
the history_defaults file have been moved to dyn_init so that each dycore
is responsible for declaring the fields for which outfld calls will be made
using the dynamics decomposition.
Moving addfld calls into dyn_init created a circular dependency, i.e.,
dyn_comp -> history -> io_dist -> dyn_comp
io_dist uses dyn_state (in dyn_comp) to access the decomposition info
needed for gathers. We broke the chain by putting dyn_state in a separate
module, i.e.,
dyn_comp -> history -> io_dist -> dyn_internal_state
Another side effect of moving addfld calls into dyn_init was to require
moving the setting of the dyngrid_set variable (and give it the more
appropriate name dyndecomp_set) to a place in front of these calls. It was
previously being set in initcom which is after the dyn_init call. It
should be set as soon as the parameters that determine the dynamics
decomposition are known. It can probably still be set earlier than it is
currently.
The gather/scatter methods in phys_grid have been modified to work when the
global field is on either a lat/lon grid or an unstructured grid.
(Contributed by Pat Worley.)
The wrapper routines in wrap_nf have been moved into a module. This
provides interface checking for these routines.
More work has been done to the dycore interface for HOMME in the
dynamics/homme directory. This work is still in progress.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: all tests pass
tempest: all tests pass
bangkok/lf95: all tests pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_3_18
Originator(s): bundy (Dani Bundy Coleman)
Date: Aug 11 2006
One-line Summary: Fix and improve prognostic aerosol packages
Purpose of changes: Some of the existing prognostic aerosol packages
did not function out-of-the-box. Furthermore, the dust package was
outdated, the sea salt was actually diagnostic and the carbon
offered only one simple scenerio.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system:
Added configure option for prognostic aerosol packages
-prog_aero sulfur,caer4,caer16,dust,seasalt
Invoke with comma-separated list of any combination of the above.
This adds the correct number of advected constituents to the build and
defines cpp tokens required for dust, seasalt and the new 16-constituent
carbon aerosol package.
Describe any changes made to the namelist:
The configure option for prognostic aerosols is used by build-namelist to
set namelist variables related to the aerosol packages.
Added namelist variables to control implementation & feedback of dust & sea salt.
aero_progsslt
Set to .TRUE. to turn on prognostic seal salt aerosols
should be set by build-namelist, as needs cppdefs too
aero_feedback_progsslt
Set to .TRUE. to enable feedback of prognostic sea salt
aerosols. NOT YET IMPLEMENTED
aero_dust
Set to .TRUE. to turn on dust prognostic aerosols
should be set by build-namelist, as needs cppdefs too
aero_feedback_dust
Set to .TRUE. to enable feedback of dust
prognostic aerosols.
To avoid confusion with the existing co_emis_file namelist variable,
renamed emissions file for prognostic carbon aerosol:
OLD co_emis
NEW caer_emis
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
When run without aerosols, there are no changes.
Code reviewed by: Brian Eaton
List all subroutines eliminated:
models/atm/cam/src/physics/cam1/dust.F90
module information moved to (existing) dust_intr.F90
List all subroutines added and what they do:
models/atm/cam/src/physics/cam1/progseasalts_intr.F90
new module includes methods for registering, initializing, setting
indices, and calculating sinks of prognostic sea salt aerosol
progseasalts_register_cnst
progseasalts_init_cnst
set_progseasalts_idx
progseasalts_initialize
progseasalts_time_interp
ProgseasaltsDryDep
models/atm/cam/src/control/camsrfexch_types.F90
hub2atm_setopts
called by dust.F90:dust_register_cnst to allocate ram1 & fv in srfflx_state
models/atm/cam/src/physics/cam1/aerosol_intr.F90
aerosol_getopts ! get runtime options
Also added
models/drv/seq_flds/dust
models/drv/seq_flds/dust/seq_flds_mod.F90
add tokens for new exchange components in x2a and l2x strings
List all existing files that have been modified, and describe the changes:
M models/atm/cam/bld/CAM_config.pm
add AEROSOLS attribute
M models/atm/cam/bld/DefaultCAM_INPARM_Namelist.xml
changed co_emis to caer_emis
add caer_emis_16
M models/atm/cam/bld/cam_inparm.pm
Add AEROSOLS attribute to namelist object
Use settings from config_cache to set aero_carbon and caer_emis
All carbon surface emissions in caer_emis
Use config_cache to set prognostic_sulfur and prescribed_sulfur
Use config_cache to set soil_erod
M models/atm/cam/bld/config_cam_eul_defaults.xml
M models/atm/cam/bld/config_cam_fv_defaults.xml
M models/atm/cam/bld/config_cam_homme_defaults.xml
M models/atm/cam/bld/config_cam_sld_defaults.xml
M models/atm/cam/bld/config_trop_chem_mozart_defaults.xml
M models/atm/cam/bld/config_waccm_ghg_defaults.xml
M models/atm/cam/bld/config_waccm_mozart_defaults.xml
add prog_aero to defaults xml files
M models/atm/cam/bld/configure
add "-prog_aero " option
add consistency check for specified packages
increment nadv with the number of constituents in specified packages
M models/atm/cam/src/control/atm_comp_mct.F90
transfer fv, ram1 & dust fluxes if dust or progseasalt is running
M models/atm/cam/src/control/camsrfexch_types.F90
add ram1,fv to srfflx_state type definition
add subroutine hub2atm_setopts( aero_dust_in )
called by dust.F90:dust_register_cnst
to allocate ram1 & fv pointers
M models/atm/cam/src/control/runtime_opts.F90
change namelist var co_emis to caer_emis
changed comments about sea salt to diagnostic
added namelist variables for prognostic sea salt & dust
M models/atm/cam/src/physics/cam1/aer_optics.F90
if prognostic dust is running, use different optics
M models/atm/cam/src/physics/cam1/aerosol_intr.F90
new subroutine aerosol_getopts to get run-time settings
add dust & progseasalt implementation & control flags
take out old ifdef DUSTs
initialize dry deposition module if any aerosol is running
generalize interfaces to carbon_intr
M models/atm/cam/src/physics/cam1/caer.F90
add new carbon aerosol scenario with 16 constituents
invoke with configure -prog_aero caer16
M models/atm/cam/src/physics/cam1/caerbnd.F90
add new caer16 carbon emission scenario
M models/atm/cam/src/physics/cam1/carbon_intr.F90
generalize interface for both caer4 & caer16 packages
D models/atm/cam/src/physics/cam1/dust.F90
moved information into dust_intr.F90
M models/atm/cam/src/physics/cam1/dust_intr.F90
move calcarm, d3ddflux subroutines to drydep_mod.F90
moved dust.F90 information into dust_intr.F90
M models/atm/cam/src/physics/cam1/hk_conv.F90
modify checks for small constituent values
causes roundoff diffs to any constituent with mixing ration < 1.e-300
M models/atm/cam/src/physics/cam1/physpkg.F90
remove ram1, fv from comsrf
A models/atm/cam/src/physics/cam1/progseasalts_intr.F90
new module to handle prognostic sea salt aersols
M models/atm/cam/src/physics/cam1/seasalt_intr.F90
added method to return ixsslt
M models/atm/cam/src/physics/cam1/sulchem.F90
add wet deposition diagnostics to dummy args of chemwdepdr
M models/atm/cam/src/physics/cam1/sulfur_intr.F90
add SO4 wet deposition diagnostics to history (not default)
remove call to inidrydep (now down in aerosol_intr if any aerosol is running)
move outfld calls to sulfur_wet_intr from chemwdepdr
M models/atm/cam/src/physics/cam1/tphysac.F90
remove fv and ram1 dummy args (in srfflx_state2d?)
pass srfflx_state2d to aerosol_drydep_intr instead of its components
M models/atm/cam/src/physics/cam1/wetdep.F90
add optional args to wetdepa
A models/drv/seq_flds/dust
A models/drv/seq_flds/dust/seq_flds_mod.F90
added directory for configure-time dust options
add tokens for new exchange components in x2a and l2x strings
M models/drv/seq_flds_indices/seq_flds_indices.F90
add indices for fv, ram1, flxdst[1-4] in x2a, l2x sections (no ifdefs here)
M models/drv/seq_mct/mrg_x2a_mct.F90
copy from surface attribute vectors to atm attvec
NOTE these land mods were checked into the clm trunk with clm3_expa_64
M src/biogeochem/DUSTMod.F90
OLD dmt_vma = 2.524e-6_r8 ! [m] Mass median diameter analytic She84 p.75 Table1
NEW dmt_vma = 3.500e-6_r8 ! [m] Mass median diameter analytic
M src/main/clm_atmlnd.F90
add land-to-atmosphere communication of fv,ram1 & dust fluxes
(only active if defined DUST or PROGSEASALT )
M src/main/lnd_comp_mct.F90
add land-to-atmosphere communication of fv,ram1 & dust fluxes
(only active if defined DUST or PROGSEASALT )
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky:
030 bl336 TBL.sh wm4h outfrq3s 9 ..................................FAIL!
Round-off changes to chemistry constituents Np and N2p are caused
by a modification to the Hack shallow convection scheme
models/atm/cam/src/physics/cam1/hk_conv.F90
tempest: all PASS
bangkok/lf95: all PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: waccm_mozart
- what platforms/compilers: bluesky tested, probably all
- nature of change (roundoff; larger than roundoff but same climate; new
climate): roundoff
If bitwise differences were observed, how did you show they were no worse
than roundoff? cprnc output shows 2 diffs in an array of 218592 elements;
code change only operates on numbers less than 1.e-300
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
===============================================================
Tag name: cam3_3_17
Originator(s): mvr
Date:
One-line Summary: update to clm3_expa_66; work-around for bluevista
compiler bug; update to new pgi compiler and libs
Purpose of changes: wanted to use new pgi compilers which meant an
update to clm tag with work-around for code the compilers didn't like;
new operating system on bluevista forced some mods in the build
system and test scripts
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
- removed noopt flag from compile and link commands in debug,smp mode
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
- added spun-up CAM and CLM initial files for the FV 0.9x1.25 resolution.
Describe any substantial timing or memory changes:
Code reviewed by: myself, eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/TSB.ccsm.sh
M models/atm/cam/test/system/TCT.ccsm.sh
- modified for ccsm test to work in new bluevista os
M models/atm/cam/test/system/TBL.sh
M models/atm/cam/test/system/TBR.sh
M models/atm/cam/test/system/TER.sh
M models/atm/cam/test/system/TRX.sh
M models/atm/cam/test/system/TEQ.sh
- test scripts modified to do ALL output file comparisons, even if one fails
M models/atm/cam/test/system/test_driver.sh
- modified for new bluevista os; updated bangkok libraries and 6.1-3 pgi compiler
M models/atm/cam/tools/scam/ui/configure
- updated to link with pgi 6.1-3 libraries
M models/atm/cam/bld/DefaultCLM_INPARM_Namelist.xml
M models/atm/cam/bld/DefaultCAM_INPARM_Namelist.xml
- added spun-up CAM and CLM initial files for the FV 0.9x1.25 resolution.
M models/atm/cam/bld/Makefile.in
- removed noopt flag from compile and link commands in debug,smp mode (work-
around for bluevista compiler bug)
M models/atm/cam/bld/run-ibm.csh
- updated to work on bluevista as well as bluesky
M models/atm/cam/bld/run-pc.csh
- updated bangkok libraries and 6.1-3 pgi compiler
M models/atm/cam/src/dynamics/fv/dynamics_vars.F90
- bug fix for out-of-bounds array reference caught by pgi compiler
M models/SVN_EXTERNAL_DIRECTORIES
- updated to new clm tag (clm3_expa_66)
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky:
005 bl131 TBL.sh e32c11dh pghgsul 9 ...............................FAIL! rc= 7
008 bl132 TBL.sh e32sdh ghgrmp 9 ..................................FAIL! rc= 7
010 bl133 TBL.sh e32pdh aqpgro 3 ..................................FAIL! rc= 7
013 bl134 TBL.sh e32dh adia 9 .....................................FAIL! rc= 7
019 bl331 TBL.sh f4c11gdh pghgsul 9 ...............................FAIL! rc= 7
022 bl332 TBL.sh f4sdh ghgrmp 9 ...................................FAIL! rc= 7
024 bl333 TBL.sh f4pdh aqpgro 3 ...................................FAIL! rc= 7
027 bl334 TBL.sh f4dh adia 9 ......................................FAIL! rc= 7
037 bl354 TBL.sh fm2dh outfrq3s 9 .................................FAIL! rc= 7
040 bl355 TBL.sh fmo2dh off2x2.5 9 ................................FAIL! rc= 7
044 bl531 TBL.sh s32c11dh pghgsul 9 ...............................FAIL! rc= 7
047 bl532 TBL.sh s32sdh ghgrmp 9 ..................................FAIL! rc= 7
049 bl533 TBL.sh s32pdh aqpgro 3 ..................................FAIL! rc= 7
052 bl534 TBL.sh s32dh adia 9 .....................................FAIL! rc= 7
055 cs998 TCS.ccsm.sh ERS f19_g13 K ...............................FAIL! rc= 2
-all baseline tests in debug mode running omp fail due to removal of noopt flag
-ccsm test fails, but passes when using ccsm sandbox with upcoming mods
tempest:
all PASS
bangkok/lf95:
012 bl153 TBL.sh e64m outfrq24h -2 ................................FAIL! rc= 7
028 bl353 TBL.sh f2m outfrq24h -2 .................................FAIL! rc= 7
040 bl553 TBL.sh s64m outfrq24h -2 ................................FAIL! rc= 5
-these were deemed acceptable roundoff errors
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations: non-debug
- what platforms/compilers: bangkok/lf95
- nature of change (roundoff; larger than roundoff but same climate; new
climate): roundoff
If bitwise differences were observed, how did you show they were no worse
than roundoff? analysis of nstep0 differences
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_3_16
Originator(s): Will Sawyer, Art Mirin
Date: Wed Jul 19 02:15:17 MDT 2006
One-line Summary:
Purpose of changes: Upgrade to newest pilgrim version;
streamlining of FV to remove unneeded allocations;
resynching FV dycore with that from GEOS5;
corrections in namelists for high resolution runs;
update of FV namelist options (removal of set_eta);
update of benergy to run with XY decomposition
Bugs fixed (include bugzilla ID): (following have no bugzilla ID)
D-resolution runs on Phoenix (reported by Worley)
Conservative-mode bug (reported by Suarez in GEOS5)
Compilation bug tp_core.F90 (reported by Trayanov in GEOS5)
Tracer transpose bug on certain architectures (e.g. Linux)
with mod_method=1 (reported by Mirin, et al.)
Describe any changes made to build system: none
Describe any changes made to the namelist:
Removed use_eta (comctl)
Added dyn_conservative (comctl)
Changes to the names of Topo, SST and initial files for
0.9x1.25 and 0.5x0.625 resolutions.
List any changes to the defaults for the boundary datasets:
RESOLUTION="0.9x1.25">atm/cam/topo/USGS-gtopo30_0.9x1.25_remap_c051027.nc
RESOLUTION="0.5x0.625" >atm/cam/sst/sst_HadOIBl_bc_0.5x0.625_1949_2001_c040402.nc
Describe any substantial timing or memory changes:
FV memory usage has decreased -- intermediate buffering of
tracers is being avoided.
Code reviewed by: ourselves
List all subroutines eliminated:
set_decomp, y_decomp (mod_comm) -- no longer used
create_vars, restore_vars, record_state, record_vars,
destroy_state, destroy_vars (dyn_comp) -- no longer used
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
M models/utils/pilgrim/parutilitiesmodule.F90
-- Changes to initialization routine to support revised mod_comm
-- Changed default communication method to mod_method=0
-- MPI types only initialized for mod_method=1
-- Updated documentation
M models/utils/pilgrim/mod_comm.F90
-- Extensive refactoring and updates to documentation
-- Dynamic allocation of buffers (minimalistic, local allocation)
-- Removal of mod_method=3 (MPI1)
M models/atm/cam/bld/DefaultCLM_INPARM_Namelist.xml
M models/atm/cam/bld/DefaultCAM_INPARM_Namelist.xml
-- Changed default files for 0.9x1.25 and 0.5x0.625 resolutions.
M models/atm/cam/src/control/runtime_opts.F90
M models/atm/cam/src/control/comctl.h
-- Added dyn_conservative, modcomm_gatscat
-- removed use_eta, tracertrans
M models/atm/cam/src/physics/cam1/phys_grid.F90
-- Changes to modmax_alltoall method (now method 13)
M models/atm/cam/src/dynamics/sld/spmd_dyn.F90
M models/atm/cam/src/dynamics/eul/spmd_dyn.F90
-- Added dyn_conservative, modcomm_gatscat
-- removed use_eta, tracertrans
M models/atm/cam/src/dynamics/fv/pmgrid.F90
-- Added mod_gatscat method, default = 0
M models/atm/cam/src/dynamics/fv/benergy.F90
-- Extensively rewritten to support XY decomposition (instead of YZ)
M models/atm/cam/src/dynamics/fv/tp_core.F90
-- Compilation bug fixed (reported by Trayanov in GEOS5)
M models/atm/cam/src/dynamics/fv/par_xsum.F90
-- Sequential execution bug fix, refactoring
M models/atm/cam/src/dynamics/fv/dp_coupling.F90
M models/atm/cam/src/dynamics/fv/dryairm.F90
-- Introduced T_TRACERS type for constituents
M models/atm/cam/src/dynamics/fv/inidat.F90
-- Introduced T_TRACERS type for constituents
-- T3 array has become i,j,k (was i,k,j)
M models/atm/cam/src/dynamics/fv/dyn_comp.F90
-- Introduced T_TRACERS type for constituents
-- Introduced dyn_conservative mode
-- Removed unneeded allocations, removed unneeded subroutines
-- Moved location of call to benergy into XY decomposition section
M models/atm/cam/src/dynamics/fv/p_d_adjust.F90
-- Introduced T_TRACERS type for constituents
M models/atm/cam/src/dynamics/fv/spmd_dyn.F90
-- Added dyn_conservative, modcomm_gatscat
-- removed use_eta, tracertrans
-- changed location and argument list of parinit()
M models/atm/cam/src/dynamics/fv/inital.F90
-- Revised call to dyn_init
M models/atm/cam/src/dynamics/fv/cd_core.F90
-- cosmetic changes to FVstart/stopclock
M models/atm/cam/src/dynamics/fv/sw_core.F90
-- loop index bug fixed (reported by Putman)
M models/atm/cam/src/dynamics/fv/diag_dynvar_ic.F90
-- T3 array has become i,j,k (was i,k,j)
-- Introduced T_TRACERS type for constituents
M models/atm/cam/src/dynamics/fv/dynamics_vars.F90
-- Bug fix for conservation mode (now: grid%klastp = km_in+1)
-- Added ACOSU
M models/atm/cam/src/dynamics/fv/geopk.F90
-- Changes for mod_geopk mode to reduce MPI_TYPE usage
M models/atm/cam/src/dynamics/fv/stepon.F90
-- Removed set_eta code (namelist variable: use_eta)
-- Introduced T_TRACERS type for constituents
M models/atm/cam/src/dynamics/fv/fv_prints.F90
-- Introduced T_TRACERS type for constituents
M models/atm/cam/src/dynamics/fv/restart_dynamics.F90
-- Introduced T_TRACERS type for constituents
-- Revised call to dyn_init
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: All tests pass except the CCSM comparison test
055 cs998 TCS.ccsm.sh ERS f19_g13 K ...............................FAIL! rc= 2
Justification: was failing in cam3_3_14 and 15. See explanations there.
tempest: all tests passed
bangkok/lf95: all tests passed
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
==> No changes to answers
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate: no change
URL for AMWG diagnostics output used to validate new climate: no change
===============================================================
===============================================================
Tag name: cam3_3_15
Originator(s): Erik Kluzek
Date: Thu Jul 13 21:17:44 MDT 2006
One-line Summary:
Purpose of changes: Put in top level initialization and clock objects at driver level
for moving forward with sequential CCSM.
This separates out the driver from the internals of the CAM time-manager. Making
the top level driver separable from CAM itself, and moves it toward a Sequential
CCSM where the sub-components can be swapped for other CCSM components (such
as the CCSM data models or active ice and active ocean models).
Bugs fixed (include bugzilla ID): 153
Fix bug in test_driver.sh where if BL_ROOT was NOT set -- it would erase the
entire contents of your /ptmp/$USER directory after it was finished.
Describe any changes made to build system: Add csm_share/eshr directory to Filepath
Describe any changes made to the namelist: Split into 4 namelists.
ccsm_inparm ----- Sequential CCSM driver level initialization information
(passed to all subcomponent models: atm, lnd, ice, ocn)
timemgr_inparm -- Sequential CCSM driver level clock information
(passed to all subcomponent models: atm, lnd, ice, ocn)
cam_inparm ------ CAM specific namelist
clm_inparm ------ CLM specific namelist
The ccsm_inparm namelist is managed by the shr_inputInfo_mod.F90 module and objects.
The timemgr_inparm namelist is managed by the eshr_timemgr_mod.F90 module and objects.
These are both sequential CCSM top level driver objects that are passed into
sub-components (atmosphere, land, sea-ice, and ocean) to manage information that is
shared between sub-components.
CHANGE SO THAT THE NAMELIST IS NO LONGER READ FROM STDIN -- THE NAMELIST NAME IS
ASSUMED AND THE FILE OPENED EXPLICITLY. When namelists read the file is opened, and
namelists are read until the correct one is found, then the file is closed. If the
needed namelist does not exist -- OR THERE IS AN ERROR IN IT -- the program will abort
with an error.
The interface to build-namelist still uses the input namelists of: camexp and clmexp
and will divide the relevant namelist items to the appropriate namelists.
Option added to build-namelist: -ignore_ic_date
by default build-namelist will match the start date/time (start_ymd,start_tod) for
initial condition datasets. If a suitable dataset with the correct starting date
(as well as resolution and other matching criteria) isn't found -- an initial condition
dataset will NOT be provided.
If you use the -ignore_ic_date option is used datasets will be matched without
examining the starting date/time of the dataset.
Namelist items removed:
restart_nsteps --> use restart_option and restart_n
restart_nhours --> use restart_option and restart_n
restart_ndays ---> use restart_option and restart_n
restart_monthly -> use restart_option and restart_n
restart_yearly --> use restart_option and restart_n
no_restart ------> use restart_option and restart_n
nelapse ---------> use stop_option and stop_n
nestep ----------> Functionality removed
nsrest ----------> Use start_type
Namelist items moved to different namelist:
Old namelist item New namelist used in
================= ====================
archive_dir ccsm_inparm (now refers to the top level archive directory)
aqua_planet ccsm_inparm
brnch_retain_casename ccsm_inparm
mss_irt ccsm_inparm
mss_wpass ccsm_inparm
calendar timemgr_inparm
stop_ymd timemgr_inparm
stop_tod timemgr_inparm
start_ymd timemgr_inparm
start_tod timemgr_inparm
ref_ymd timemgr_inparm
ref_tod timemgr_inparm
perpetual_run timemgr_inparm
perpetual_ymd timemgr_inparm
Namelist items that changed names:
Old namelist item New namelist item name New namelist used in
================= ====================== ====================
eccen orb_eccen timemgr_inparm
obliq orb_obliq timemgr_inparm
mvelp orb_mvelp timemgr_inparm
iyear_AD orb_iyear_AD timemgr_inparm
ctitle case_desc ccsm_inparm
caseid case_name ccsm_inparm
adiabatic atm_adiabatic ccsm_inparm
ideal_phys atm_ideal_phys ccsm_inparm
Namelist items added:
Namelist item Description New namelist used in
========================= ====================================== ====================
start_type ---------------> How simulation will startup ccsm_inparm
Valid options:
startup = Use initial files
continue = Use restart files to continue simulation
branch = Use restart files to branch simulation
restart_pfile ------------> Driver level restart pointer file name ccsm_inparm
restart_file -------------> Driver level restart file name ccsm_inparm
restart_file_override ----> Override list of items from restart file ccsm_inparm **
restart_file_TGRoverride -> Override list of items from restart file ccsm_inparm **
restart_option -----------> Driver level restart frequency type timemgr_inparm
Valid options:
nsteps ---- Write restarts every restart_n time-steps
ndays ----- Write restarts every restart_n days
nmonths --- Write restarts every restart_n months
nyears ---- Write restarts every restart_n years
monthly --- Write restarts at beginning of each month
yearly ---- Write restarts at beginning of the year
end ------- Write restarts only at end of simulation
none ------ Don't write any restarts
restart_n ----------------> Driver level restart frequency value timemgr_inparm
stop_option --------------> Driver level restart frequency type timemgr_inparm
Valid options:
nsteps ---- Stop after stop_n time-steps
ndays ----- Stop after stop_n days
nmonths --- Stop after stop_n months
nyears ---- Stop after stop_n years
date ------ Stop at stop_ymd/stop_tod date/time
stop_n -------------------> Driver level restart frequency value timemgr_inparm
stop_final_ymd -----------> The final date to run to (YYYYMMDD format) timemgr_inparm
atm_cpl_dt ---------------> Atmosphere coupling frequency (sec) timemgr_inparm
Example creation of the old and new namelist using build-namelist:
With cam3_3_14:
build-namelist -runtype initial -namelist "&camexp nelapse=-1, restart_monthly=.true., iyear_AD=1990/ "
with cam3_3_15:
build-namelist -runtype startup -namelist "&camexp stop_option='ndays', stop_n=1, restart_option='monthly' , orb_iyear_AD=1990 /"
So build-namelist figures out that all of the above options belong in the timemgr_inparm namelist
and deal with it appropriately. And results in the following namelist:
&ccsm_inparm
case_name = 'camrun'
start_type = "startup"
/
&timemgr_inparm
atm_cpl_dt = 1200
orb_iyear_ad = 1950
restart_option = 'monthly'
start_ymd = 901
stop_n = 1
stop_option = 'ndays'
/
&cam_inparm
absems_data = '/fs/cgd/csm/inputdata/atm/cam/rad/abs_ems_factors_fastvx.c030508.nc'
aeroptics = '/fs/cgd/csm/inputdata/atm/cam/rad/AerosolOptics_c050419.nc'
bnd_topo = '/fs/cgd/csm/inputdata/atm/cam/topo/USGS-gtopo30_64x128_c050520.nc'
bndtvaer = '/fs/cgd/csm/inputdata/atm/cam/rad/AerosolMass_V_64x128_clim_c031022.nc'
bndtvo = '/fs/cgd/csm/inputdata/atm/cam/ozone/pcmdio3.r8.64x1_L60_clim_c970515.nc'
bndtvs = '/fs/cgd/csm/inputdata/atm/cam/sst/sst_HadOIBl_bc_64x128_clim_c020411.nc'
dtime = 1200
ncdata = '/fs/cgd/csm/inputdata/atm/cam/inic/gaus/cami_0000-09-01_64x128_L26_c030918.nc'
/
&clm_inparm
fpftcon = '/fs/cgd/csm/inputdata/lnd/clm2/pftdata/pft-physiology-cn16.c040719'
fsurdat = '/fs/cgd/csm/inputdata/lnd/clm2/srfdata/cam/clms_3.1_64x128_c050523.nc'
/
How to override information on the restart file from the namelist: **
By default for a continue or branch type of simulation -- driver information is
read from the restart file and namelist information is ignored. In order to explicitly
use driver level restart information you have to do the following:
1.) Set given namelist item(s) you want to override.
2.) Set restart_file_override (or restart_file_TGRoverride for timemgr_inparm namelist)
to include the colon delimited list of items you want to override on the namelist.
3.) To override the case_name -- set the brnch_retain_casename namelist item to true.
List of ccsm_inparm items that can be overridden: mss_irt, mss_wpass, and case_desc
(and case_name is brnch_retain_casename is set to true)
List of timemgr_inparm items that can be overridden: restart_option, restart_n
Examples
build-namelist -runtype continue -namelist \
"&camexp mss_irt=45, mss_wpass='example', restart_file_override='mss_irt:mss_wpass' /"
The above example will override the values of mss_irt, mss_wpass from the restart file
with those given on the namelist. Note if restart_file_override were missing, the
values on the restart file would be used an the values on the namelist ignored.
build-namelist -runtype continue -namelist \
"&camexp restart_option='nyears', restart_n=2, restart_file_TGRoverride='restart_option:restart_n' /"
The above example will override the values of restart_option, and restart_n from
the restart file with those given on the namelist. Note if restart_file_override
were missing, the values on the restart file would be used an the values on the
namelist ignored.
List any changes to the defaults for the boundary datasets: make sure IC file names are
consistent with dates in file.
Also add ic_ymd and ic_tod to describe the date (YYYYMMDD format) and time-of-day (sec)
that the given initial condition file represents.
Describe any substantial timing or memory changes: none
Code reviewed by: Brian Eaton, Mariana Vertenstein, Mat Rothstein
Externals updated:
share3_060710 ------------ Add shr_inputinfo_mod/eshr_timemgr_mod codes.
clm3_expa_65 ------------- Use new share objects.
esmf_wrf_timemgr_060616 -- Same behavior as ESMF, changes needed for new share.
MCT2_2_1_060706 ---------- Changes get_zeits.c to work on new bangkok compiler.
List all subroutines eliminated:
In models/atm/cam/src/control/restart.F90 <-- determine if restart from SyncClock
restart_init
restart_is_write_step
update_next_write_time
In models/atm/cam/src/utils/time_manager.F90 <-- remove unused functions.
get_clock
get_curr_ESMF_Time
calc_nestep <---- not needed as SynClock determines stop-time.
List all subroutines added and what they do:
runtime_opts_setNLFile (runtime_opts.F90) -- Set the namelist filename.
ccsmini_sendgridgetorb (ccsm_msg.F90) ------ Send grid and get orbit info
at initialization.
get_calendar (time_manager.F90) ------------ Get calendar type using.
timemgr_check_restart (time_manager.F90) --- Check the restart info for consistency.
List all files eliminated:
atm/cam/bld/DefaultCLMEXPNamelist.xml (change name)
atm/cam/bld/clm2exp.pm (change name)
atm/cam/bld/CAM_namelist.p (change name)
atm/cam/bld/DefaultCAMEXPNamelist.xml (change name)
List all files added and what they do:
atm/cam/test/unit/control/filterhead.pl -- filters unit test log data to compare results
with a different version.
atm/cam/bld/nl_descrips.pm --------------------- Describes the driver level namelists.
atm/cam/bld/clm_inparm.pm ---------------------- Manages the clm_inparm namelist.
atm/cam/bld/DefaultCAM_INPARM_Namelist.xml ----- Default values for the cam_inparm namelist.
atm/cam/bld/DefaultCLM_INPARM_Namelist.xml ----- Defaults for the clm_inparm
namelist (renamed from old CAMEXP file)
atm/cam/bld/DefaultTIMEMGR_INPARM_Namelist.xml - Default values for the timemgr_inparm namelist.
atm/cam/bld/DefaultCCSM_INPARM_Namelist.xml ---- Default values for the ccsm_inparm namelist.
atm/cam/bld/NamelistsDescriptions.xml ---------- List of items on driver level namelists.
atm/cam/bld/timemgr_inparm.pm ------------------ Manages the timemgr_inparm namelist.
atm/cam/bld/cam_inparm.pm ---------------------- Manages the cam_inparm namelist.
atm/cam/bld/SeqCCSM_namelist.pm ---------------- Manages all namelists needed.
atm/cam/bld/ccsm_inparm.pm --------------------- Manages the ccsm_inparm namelist
List all existing files that have been modified, and describe the changes:
Pass CCSMInit and SyncClock objects down to sub-models.
ocn/dom/ocn_comp_mct.F90
ocn/som/ocn_comp_mct.F90
ice/csim4/ice_comp_mct.F90
Update unit tests for time_manager module.
atm/cam/test/unit/control/configure
atm/cam/test/unit/control/run_time_test
atm/cam/test/unit/control/test_time.F90
Change test_driver system to work with new build-namelist options and new
namelist behavior and changes to namelist item names.
atm/cam/test/system/test_driver.sh
atm/cam/test/system/TSB.ccsm.sh
atm/cam/test/system/TBR.sh
atm/cam/test/system/TER.sh
atm/cam/test/system/TRX.sh
atm/cam/test/system/TSM.sh
atm/cam/test/system/nl_files/idphys
atm/cam/test/system/nl_files/ghgrmp
atm/cam/test/system/nl_files/outfrq3s
atm/cam/test/system/nl_files/aqpgro
atm/cam/test/system/nl_files/no_ttrac
atm/cam/test/system/nl_files/off2x2.5
atm/cam/test/system/nl_files/pghgsul
atm/cam/test/system/nl_files/ttrac_lb1
atm/cam/test/system/nl_files/ttrac_lb2
atm/cam/test/system/nl_files/ttrac
atm/cam/test/system/nl_files/ttrac_lb3
atm/cam/test/system/nl_files/adia
atm/cam/test/system/nl_files/outfrq24h
Pass CCSMInit and SyncClock objects down as appropriate into scam. Upgrade
make to work with new PGI compiler on bangkok.
atm/cam/tools/scam/configure
atm/cam/tools/scam/testscript
atm/cam/tools/scam/userdata/crmtest26.out <--- Needed for compiler upgrade on bangkok.
atm/cam/tools/scam/ui/configure
atm/cam/tools/scam/scm_init/init_model.F90
atm/cam/tools/scam/scm_init/scam_srfdata_MCT.F90
atm/cam/tools/scam/scm_init/scam_run.F90
atm/cam/tools/scam/scm_init/inital.F90
atm/cam/tools/scam/scm_init/scamMod.F90
atm/cam/tools/scam/scm_init/setiopupdate.F90
Work with new PGI compiler on bangkok. Add eshr to Filepath.
Changes to get Darwin (Mac OS-X) to work. Remove HIDE_SHR_MSG as
unneeded now.
atm/cam/bld/configure
atm/cam/bld/Makefile.in
Change to work in the context of the 4 new namelists. Add -ignore_ic_date option
in. Have CAM_config.pm keep track of new items added to the XML configuration file.
atm/cam/bld/build-namelist
atm/cam/bld/clmexp.pm
atm/cam/bld/namelist.pm
atm/cam/bld/camexp.pm
atm/cam/bld/atmlndnl.pm
atm/cam/bld/CAM_config.pm
Change run scripts to work with new build-namelist and do NOT redirect stdin.
atm/cam/bld/run-ibm.csh
atm/cam/bld/run-pc.csh
atm/cam/bld/run-sgi.csh
atm/cam/bld/run-lightning.csh
atm/cam/bld/run-darwin.csh
Pass CCSMInit and SyncClock objects down as needed. Use them to
initialize CAM internal data.
atm/cam/src/control/readinitial.F90
atm/cam/src/control/runtime_opts.F90
atm/cam/src/control/restart.F90
atm/cam/src/control/history.F90
atm/cam/src/control/atm_comp_mct.F90
atm/cam/src/control/startup_initialconds.F90
atm/cam/src/control/units.F90
atm/cam/src/control/filenames.F90
atm/cam/src/control/ioFileMod.F90
atm/cam/src/control/cam_comp.F90
atm/cam/src/utils/time_manager.F90
atm/cam/src/dynamics/fv/metdata.F90
Create CCSMInit and SyncClock driver level data at top level concurrent CAM
driver pass down as needed. Set orbital information in SyncClock from data sent
from coupler.
atm/cam/src/control/con_cam.F90
atm/cam/src/control/ccsm_msg.F90
Remove dependence on CAM specific data and modules and use CCSMInit
and SyncClock driver level objects.
drv/seq_mct_drv/seq_ccsm_drv.F90
drv/seq_mct/mrg_x2a_mct.F90
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: All tests pass except the CCSM comparison test
055 cs998 TCS.ccsm.sh ERS f19_g13 K ...............................FAIL! rc= 2
CCSM failed because the new CCSM scripts for this tag don't work with the old CAM
Thus the comparison to the old CCSM tag and cam3_3_14 had to be done by hand.
The comparison is exact if esmf_wrf_timemgr is updated in cam3_3_14 to
esmf_wrf_timemgr_051212. If not the difference is bit-for-bit until the 25th time-step
when the difference is off by roundoff. This was documented in the cam3_3_4 commit
where we had a roundoff difference due to the new esmf_wrf_timemgr.
For running CCSM test use:
env CAM_CCSMROOT=/fs/cgd/csm/models/atm/newchg_ccsm3_1_beta34
tempest: none
bangkok/lf95: none
CAM tag used for the baseline comparison tests if different than previous
tag: cam3_3_14
Summarize any changes to answers: none (bit-for-bit)
CCSM (changes to roundoff)
CCSM answers change because esmf_wrf_timemgr updated from an
older version 050309 to the version used here 060616.
===============================================================
===============================================================
Tag name: cam3_3_14
Originator(s): mvr
Date: 060612
One-line Summary:
Update the ccsm tests within cam's test suite to reflect new changes to
ccsm's scripts; minor bug fixes to test scripts; cray x1 mods in newcprnc
Purpose of changes: new naming conventions were introduced to ccsm's
test scripts; newcprnc required mods for cray x1
Bugs fixed (include bugzilla ID): applied fix for bug #43 to cam's newcprnc
code just as it was applied to clm's
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: myself
List all subroutines eliminated:
D models/atm/cam/tools/newcprnc/cprnc.f90
- file renamed due to pre-processor directives added for cray x1
List all subroutines added and what they do:
A + models/atm/cam/tools/newcprnc/cprnc.F90
- new file with directives for cray x1
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/TSB.ccsm.sh
- mods to reflect name changes introduced in ccsm scripts; added cleanup after
successful ccsm test; minor bug fixes
M models/atm/cam/test/system/TCT.ccsm.sh
- mods to reflect name changes introduced in ccsm scripts; now dumps output
of create_test to output log
M models/atm/cam/test/system/test_driver.sh
- test suite now uses prod queue rather than debug on phoenix
M models/atm/cam/test/system/gen_test_table.sh
- utility updated to include all platforms supported by cam's test driver
M models/atm/cam/test/system/input_tests_master
- mods to reflect name changes introduced in ccsm scripts
M models/atm/cam/test/system/TCB.ccsm.sh
- mods to reflect name changes introduced in ccsm scripts
M models/atm/cam/tools/newcprnc/nfwrappers.f90
- bug fix for intent of variables in wrap_open (see bug #43)
M models/atm/cam/tools/newcprnc/Makefile
- added mods for cray x1
M posttag_cron_bangkok.sh
M posttag_cron_bluesky.sh
M posttag_cron_lightning.sh
M posttag_cron_bluevista.sh
- cron scripts modified to make use of "collections" area for cam tags
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky:
055 cs998 TCS.ccsm.sh ERS f19_g13 K ...............................FAIL! rc= 2
- ccsm test fails when comparing to baseline due to changes in naming
conventions...tests were verified by hand
tempest: all pass
bangkok/lf95: all pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
Tag name: cam3_3_13
Originator(s): Jim Edwards
Date: 060606
One-line Summary: addition of HOMME dynamic core, removal of plat/plon from physics
Purpose of changes: development of a new dynamic core prototype, generalization of
model to non-rectangular horizontal grids
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: Added support for homme dynamics in configure.
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: myself, Brian Eaton
List all subroutines eliminated:
List all subroutines added and what they do: a dynamics/homme subdirectory was added
with stubs to interface the (external) homme model
A models/atm/cam/src/dynamics/homme
A models/atm/cam/src/dynamics/homme/initcom.F90
A models/atm/cam/src/dynamics/homme/pmgrid.F90
A models/atm/cam/src/dynamics/homme/dp_coupling.F90
A models/atm/cam/src/dynamics/homme/dycore.F90
A models/atm/cam/src/dynamics/homme/dyn_comp.F90
A models/atm/cam/src/dynamics/homme/inidat.F90
A models/atm/cam/src/dynamics/homme/stepon.F90
A models/atm/cam/src/dynamics/homme/restart_dynamics.F90
A models/atm/cam/src/dynamics/homme/commap.F90
A models/atm/cam/src/dynamics/homme/spmd_dyn.F90
A models/atm/cam/src/dynamics/homme/dyn_grid.F90
A models/atm/cam/src/dynamics/homme/README
A models/atm/cam/src/dynamics/homme/inital.F90
A models/atm/cam/bld/config_cam_homme_defaults.xml
List all existing files that have been modified, and describe the changes:
Changed FV CPP macro STAGGERED to the more appropriate and general DYN_STATE_INTERFACE
U models/atm/cam/src/control/runtime_opts.F90
U models/atm/cam/src/control/ccsm_msg.F90
U models/atm/cam/src/control/startup_initialconds.F90
U models/atm/cam/src/control/cam_comp.F90
U models/atm/cam/src/utils/time_manager.F90
Removed references to plat and plon.
U models/atm/cam/src/physics/cam1/comsrf.F90
U models/atm/cam/src/physics/cam1/dmsbnd.F90
U models/atm/cam/src/physics/cam1/acbnd.F90
U models/atm/cam/src/physics/cam1/dust_intr.F90
U models/atm/cam/src/physics/cam1/tracers_suite.F90
U models/atm/cam/src/physics/cam1/boundarydata.F90
U models/atm/cam/src/physics/cam1/drydep_mod.F90
U models/atm/cam/src/physics/cam1/restart_physics.F90
U models/atm/cam/src/physics/cam1/stratiform.F90
U models/atm/cam/src/physics/cam1/water_tracers.F90
U models/atm/cam/src/physics/cam1/co2_data_flux.F90
U models/atm/cam/src/physics/cam1/volcanicmass.F90
U models/atm/cam/src/physics/cam1/physpkg.F90
U models/atm/cam/src/physics/cam1/wetdep.F90
U models/atm/cam/src/physics/cam1/prescribed_aerosols.F90
U models/atm/cam/src/physics/cam1/carbon_intr.F90
U models/atm/cam/src/physics/cam1/sulchem.F90
U models/atm/cam/src/physics/cam1/co2_cycle.F90
U models/atm/cam/src/physics/cam1/sulbnd.F90
U models/atm/cam/src/physics/cam1/soxbnd.F90
U models/atm/cam/src/physics/cam1/caerbnd.F90
U models/atm/cam/src/physics/cam1/aerosol_intr.F90
U models/atm/cam/src/physics/cam1/advnce.F90
U models/atm/cam/src/physics/cam1/sulfur_intr.F90
Added support for HOMME dycore
U models/atm/cam/bld/configure
U models/atm/cam/bld/resolution_parameters.xml
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: none
tested with CAM_CCSMROOT=/fs/cgd/csm/collections/ccsm3_1_beta29
tempest: none
bangkok/lf95: none
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
Tag name: cam3_3_12
Originator(s): mvr, jedwards
Date: 060602
One-line Summary: test suite now supports phoenix (cray x1);
added testing of coupled model to bluevista; bug fix
for fv coupled runs
Purpose of changes: wanted test coverage of cray x1; testing
of coupled model had been restricted to just bluesky;
new features in ccsm's scripts available for use
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets:
- added DMS_emmissions, oxid, and SOx_emissions files for fv 2x2.5
- updated to newer SOx_emissions file for T85
Describe any substantial timing or memory changes:
Code reviewed by: myself
List all subroutines eliminated:
none
List all subroutines added and what they do:
A models/atm/cam/test/system/config_files/fmo2m
A models/atm/cam/test/system/config_files/f2c11m
A models/atm/cam/test/system/config_files/e128m
A models/atm/cam/test/system/config_files/e128c11m
A models/atm/cam/test/system/config_files/fm1.9m
A models/atm/cam/test/system/config_files/f1.9pm
A models/atm/cam/test/system/config_files/f1.9m
A models/atm/cam/test/system/config_files/e128pm
- new configuration options for testing on phoenix
A models/atm/cam/test/system/nl_files/off2x2.5p
- new namelist options for testing on phoenix
A models/atm/cam/test/system/TCT.ccsm.sh
- new test script needed for testing of coupled model
A models/atm/cam/test/system/tests_posttag_phoenix
A models/atm/cam/test/system/tests_posttag_robin
- new sets of default posttag tests for robin and phoenix
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/TSB.ccsm.sh
M models/atm/cam/test/system/TCB.ccsm.sh
- scripts for testing of coupled model now supports bluevista, phoenix;
build of coupled model can now be done separately prior to run;
updated to take advantage of new features in ccsm scripts
M models/atm/cam/test/system/tests_posttag_bluesky
M models/atm/cam/test/system/tests_pretag_bluesky
- defaults test sets modified to have fv ccsm test be pretag, eul posttag
M models/atm/cam/test/system/tests_posttag_bluevista
- default posttag tests on bluevista now include ccsm tests
M models/atm/cam/test/system/test_driver.sh
- support added for robin/phoenix; MPI_TYPE_MAX now set for all platforms
M models/atm/cam/test/system/input_tests_master
M models/atm/cam/test/system/CAM_runcmnd.sh
- new tests added to master list for phoenix
M models/atm/cam/bld/camexp.pm
- mod to allow for casenames of length 80 characters
M models/atm/cam/bld/DefaultCAMEXPNamelist.xml
- added default input files for fv 2x2.5; updated to a newer T85 file
M models/atm/cam/src/control/ccsm_msg.F90
M models/atm/cam/src/control/history.F90
- bug fixes to remove call to getfil for files not yet in existence -
would fail on phoenix when it tried to find it on mass store
M models/atm/cam/src/chemistry/trop_mozart/mo_imp_sol.F90
- bug fix for line length exceeding max number of characters
M models/atm/cam/src/dynamics/fv/dyn_grid.F90
- bug fix for problem introduced in cam3_3_11 for fv coupled runs
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky:
001 sb998 TSB.ccsm.sh ER.01a 1.9x2.5_gx1v3 K ......................FAIL! rc= 6
055 cs998 TCS.ccsm.sh ER.01a 1.9x2.5_gx1v3 K ......................FAIL! rc= 2
- failed due to mods in test scripts; test passes when run manually
tempest:
all pass
bangkok/lf95:
all pass
CAM tag used for the baseline comparison tests if different than previous
tag: fv ccsm test used cam3_3_10
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff? b4b
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
Tag name: cam3_3_11
Originator(s): pworley, Jim Edwards
Date: 5/25/2006
One-line Summary: phys_grid changes for single index method to support non-rectangular grids
Purpose of changes: to support non-rectangular grids
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: myself, Brian Eaton, Mariana, Pat Worley
List all subroutines eliminated:
cam_domain,
get_cid_all_p,
get_chunk_coord_owner_p,
get_chunk_owner_p
get_cid_p
get_block_coord_d
get_block_coord_cnt_d
get_block_col_cnt_d
get_lon_d
get_lat_d
List all subroutines added and what they do:
get_gcol_all_p, get the global column id for all columns in the chunk
get_area_all_p, get the area of all columns in the chunk
get_block_gcol_d get column indices for given block
get_gcol_block_d get global block indices and local columns index for given global column index
get_horiz_grid_d: provides the total number of global columns visible to the physics
the lat, lon location and the surface area of each column
(functions)
get_area_p : get area of column
get_gcol_p : get gcol id of column
get_gcol_owner_p : get task of column gcol in physics
get_block_gcol_cnt_d : get number of columns in given block
get_gcol_block_cnt_d : get number of blocks containing data
from a given global column index
get_horiz_grid_cnt_d : get number of columns in dynamics grid (visible to physics)
List all existing files that have been modified, and describe the changes:
The cam_domain module was removed and these files modified to
get the mct grid initialization data directly from the physics
chunk on the task. This is both a simplification and a generalization
of the method introduced in cam3_3_9
M models/ocn/dom/ocn_comp_mct.F90
M models/ocn/som/ocn_comp_mct.F90
M models/ice/csim4/ice_comp_mct.F90
M models/atm/cam/src/control/ccsm_msg.F90
M models/atm/cam/src/control/atm_comp_mct.F90
M models/atm/cam/src/control/cam_comp.F90
M models/drv/seq_mct/seq_domain_mct.F90
Chunks (decomposed grid in the physics) were changed from
2-D arrays in the horizontal to 1D.
This allows the generalization of cam physics to
non-rectangular grids.
M models/atm/cam/src/physics/cam1/physics_types.F90
M models/atm/cam/src/physics/cam1/phys_grid.F90
M models/atm/cam/src/dynamics/sld/dyn_grid.F90
M models/atm/cam/src/dynamics/eul/dyn_grid.F90
M models/atm/cam/src/dynamics/fv/dyn_grid.F90
Support for the historical reduced grid method has been removed
if a reduced grid is reintroduced in the future it should use the
more general method introduced here.
M models/atm/cam/src/chemistry/trop_mozart/mo_lightning.F90
M models/atm/cam/src/chemistry/trop_mozart/mo_ub_vals.F90
M models/atm/cam/src/chemistry/trop_mozart/mo_tropopause.F90
M models/atm/cam/src/chemistry/trop_mozart/mo_sulf.F90
M models/atm/cam/src/chemistry/trop_mozart/mo_chemini.F90
M models/atm/cam/src/chemistry/trop_mozart/mo_airplane.F90
M models/atm/cam/src/chemistry/trop_mozart/chemistry.F90
M models/atm/cam/src/chemistry/trop_mozart/mo_lb_vals.F90
M models/atm/cam/src/chemistry/trop_mozart/mo_drydep.F90
M models/atm/cam/src/chemistry/trop_mozart/mo_photo.F90
M models/atm/cam/src/chemistry/trop_mozart/mo_ch4_lbc.F90
M models/atm/cam/src/chemistry/trop_mozart/mo_strato_sad.F90
M models/atm/cam/src/chemistry/trop_mozart/mo_srf_emissions.F90
M models/atm/cam/src/chemistry/waccm_mozart/mo_tgcm_ubc.F90
M models/atm/cam/src/chemistry/waccm_mozart/mo_sulf.F90
M models/atm/cam/src/chemistry/waccm_mozart/mo_airplane.F90
M models/atm/cam/src/chemistry/waccm_mozart/mo_drydep.F90
M models/atm/cam/src/chemistry/waccm_mozart/mo_srf_emissions.F90
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: All tests pass
tempest: All tests pass
bangkok/lf95: All tests pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_3_10
Originator(s): mvr
Date: 060524
One-line Summary: added gensom option to configure; added test of gensom
vars to test suite; updated externals to new timing files; cleanup
Purpose of changes: SOM input fields should not be generated and written
to history tapes by default; unused integer return codes of some
timing routines were causing problems on phoenix; write statements
and empty directories were left behind from previous commits
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: -gensom option added to configure
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: not including the SOM
input fields by default may show slight improvement
Code reviewed by: myself, eaton
List all subroutines eliminated:
D models/atm/cam/test/system/config_files/f4c11dh
- config options file replaced with one that includes -gensom option for testing
D models/atm/cam/src/ocnsice
D models/atm/cam/src/ocnsice/dom
D models/atm/cam/src/ocnsice/som
- removed empty directories left behind from previous commit
List all subroutines added and what they do:
A models/atm/cam/test/system/config_files/f4c11gdh
- replacement config options file with -gensom option for testing
List all existing files that have been modified, and describe the changes:
M models/utils/SVN_EXTERNAL_DIRECTORIES
- updating to new timing tag with mods for replacing int return codes with void
M models/atm/cam/test/system/input_tests_master
- adding test of the new configure option (gensom) to an existing test
M models/atm/cam/bld/configure
- modified to handle new option for generating SOM input fields
M models/atm/cam/bld/config_cam_eul_defaults.xml
M models/atm/cam/bld/config_trop_chem_mozart_defaults.xml
M models/atm/cam/bld/config_waccm_ghg_defaults.xml
M models/atm/cam/bld/config_waccm_mozart_defaults.xml
M models/atm/cam/bld/config_cam_fv_defaults.xml
M models/atm/cam/bld/config_cam_sld_defaults.xml
- added default values for gensom
M models/atm/cam/src/control/atm_comp_mct.F90
M models/drv/seq_mct/mrg_x2a_mct.F90
M models/drv/seq_mct/seq_domain_mct.F90
- removed write statements left behind from previous commits
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky:
019 bl331 TBL.sh f4c11gdh pghgsul 9 ...............................FAIL! rc= 5
- failed because this is a new test and did not exist in previous tag
tempest:
021 bl331 TBL.sh f4c11gdh pghgsul 9 ...............................FAIL! rc= 5
- failed because this is a new test and did not exist in previous tag
bangkok/lf95:
none
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_3_9
Originator(s): Mariana Vertenstein
Date: Wed May 17 13:05:13 MDT 2006
One-line Summary: created new generate som mode for mct coupling
Purpose of changes: Remove output of ice/ocn specific fields from
mrg_x2a_mct.F90 and move then to atm_comp_mct.F90. All routines
in mct_drv/ should be devoid of cam specific information.
Bugs fixed (include bugzilla ID): None
Describe any changes made to build system: default seq_flds_mod.F90
to be used is currently in drv/seq_flds/gensom. This corresponds
to extra fields being sent to the atm which would only be used
for SOM input fields generation. Changes need to be made to
the testing scripts and configure to support a gensom option.
Describe any changes made to the namelist: None
List any changes to the defaults for the boundary datasets: None
Describe any substantial timing or memory changes: The extra
fields being sent to atm_comp_mct.F90 might have a very small
performance penalty. However, a fix for this is expected in the
very near future.
Code reviewed by: Myself (discussed these changes with Brian Eaton)
List all subroutines eliminated:
List all subroutines added and what they do:
A drv/seq_flds/gensom/seq_flds_mod.F90
sends fields to atm that are needed for SOM input data generation
List all existing files that have been modified, and describe the changes:
M atm/cam/src/control/history_defaults.F90
M atm/cam/src/physics/cam1/diagnostics.F90
M ocn/som/ocn_comp.F90
moved history file initialization from diagnostics.F90 and history_defaults.F90 for:
TSOCN&IC, QFLUX, QFLUX_FT, QFLUX_TH, QFLUX_A2, FOCN, OIE, OIERATE, ONF, MLDANN
M ice/csim4/ice_comp.F90
moved history file initialization from diagnostics.F90 and history_defaults.F90 for:
MELTB ,MELTT ,MELTL ,GROWB ,FRAZIL ,FLOOD ,FRZMLT ,NRGERROR,DELTAICE
NRGICE ,IIERATE ,F_ICE ,F_OCN ,FRZMLTMX,IMBAL ,EICEIN ,EICEOUT , TSICERAD&IC
M atm/cam/src/control/ccsm_msg.F90
moved history file initialization from diagnostics.F90 and history_defaults.F90 for:
CPLRAINC,CPLRAINL,CPLSNOWC,CPLSNOWL,CPLPRCER
M atm/cam/tools/scam/scm_init/scam_run.F90
M drv/seq_mct_drv/seq_ccsm_drv.F90
removed call to mrg_x2a_run1_mct
renamed call to mrg_x2a_run2_mct as call to mrg_x2a_run_mct
M drv/seq_mct/mrg_x2a_mct.F90
removed subroutine mrg_x2a_run1_mct
renamed subroutine mrg_x2a_run2_mct as call to mrg_x2a_run_mct
M atm/cam/bld/configure
set default path for seq_flds to $srcdir/drv/seq_flds/gensom
M atm/cam/src/control/atm_comp_mct.F90
added private subroutines atm_gensom1_mct and atm_gensom2_mct
that provide fuctionality previously done in
mrg_x2a_run1_mct and mrg_x2a_run2_mct - they are only invoked
if indices in seq_flds_indices for SOM generation fields are
non-zero (this will only be the case if
$srcdir/drv/seq_flds/gensom/seq_flds_mod.F90 is used.
M drv/seq_flds_indices/seq_flds_indices.F90 added indices necessary for SOM data generation
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: all pass
tempest: all pass
bangkok/lf95: all pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_3_8
Originator(s): Mariana Vertenstein
Date: Sat May 13 17:48:49 MDT 2006
One-line Summary: to implement MCT domains in each component
Purpose of changes: To implement MCT domains and implement
a top level domain checker for cam and the surface components
using these domains (this replaces use statements within clm)
Note that cam3_3_7 was identical to this tag - but was tagged
with the wrong land externals - this tag fixes this problem.
Bugs fixed (include bugzilla ID): None
Describe any changes made to build system: None
Describe any changes made to the namelist: None
List any changes to the defaults for the boundary datasets: None
Describe any substantial timing or memory changes: None
all changes are in the initialization phase - however,
the impact of timing due to initialization has not been tested
Code reviewed by: Tony Craig and Rob Jacob
List all subroutines eliminated: None
List all subroutines added and what they do:
A models/atm/cam/src/control/cam_domain.F90
A new routine that can be used by both ccsm_msg.F90 and the
atm_comp_mct.F90 routine to calculate the cam grid and areas
(this code was previously contained in ccsm_msg.F90).
List all existing files that have been modified, and describe the changes:
M models/atm/cam/src/control/atm_comp_mct.F90
M models/ocn/dom/ocn_comp_mct.F90
M models/ocn/som/ocn_comp_mct.F90
M models/ice/csim4/ice_comp_mct.F90
M models/lnd/clm2/src/main/lnd_comp_mct.F90
M models/atm/cam/tools/scam/scm_init/init_model.F90
M models/atm/cam/tools/scam/scm_init/scam_srfdata_MCT.F90
M models/drv/seq_mct_drv/seq_ccsm_drv.F90
M models/drv/seq_mct/seq_mct_mod.F90
D models/drv/seq_mct/seq_mct_init.F90 => A models/drv/seq_mct/seq_init_mct.F90
In the above routines:
* domains were implemented using the MCT general grid data structures
* the current general grid components that I am defining are
lon,lat,area,mask and maxfrac (maxfrac is the maximum fraction
that the model can have on any gridcell - for the landthis
corresponds to the landfrac, for the ocn and ice this should
correspond to 1.-landfrac when everyone is on the same grid)
* domain checking is only done on the master processor at
initialization (via an MCT gather on the general grid)
* essentially, domains enable grid and fraction checking to be done
at the top level by the master processor and eliminates the use
statements contained in the land model. The consistency check for
fractional land is somewhat different than in the current flux
coupler. Basically, when all the models are on the same grid, you
want the "maximum fraction" from each component to be consistent.
What this boils down to is that the max_frac from the ocn/ice must
equal (1.-landfrac)to within some eps (I have specified 1.e-13).
In the ccsm cpl6 system, the ocn determines the landfrac,and on the
ocean grid maxfrac is either 1 or 0. This is clearly different
than in the current system.
M models/atm/cam/src/physics/cam1/comsrf.F90
M models/atm/cam/src/physics/cam1/restart_physics.F90
M models/atm/cam/src/dynamics/sld/inidat.F90
M models/atm/cam/src/dynamics/eul/inidat.F90
M models/atm/cam/src/dynamics/fv/inidat.F90
-removed landfrac_glob.F90 from above routines
M models/atm/cam/src/control/ccsm_msg.F90
- made changes to use cam_domain.F90 for determining grid information
M models/ocn/dom/ocn_comp.F90
M models/ocn/som/ocn_comp.F90
M models/ice/csim4/ice_comp.F90
M atm/cam/src/physics/cam1/diagnostics.F90
- moved calls for the following initial file fields:
SNOWHICE&IC, ICEFRAC&IC, SICTHK&IC, TSICERAD&IC
to models/ice/csim4/ice_comp.F90
- moved calls for the following initial file fields:
TSOCN&IC
to models/ocn/som/ocn_comp.F90 and models/ocn/dom/ocn_comp.F90
- removed references to initial file file fields TS&IC since this
intiial field it was not being used by any routine
M atm/cam/src/physics/cam1/restart_physics.F90
- removed references to landfrac_glob and all COUP_CSM if-defs
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: all pass
tempest: all pass
bangkok/lf95: all pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_3_7
Originator(s): Mariana Vertenstein
Date: Sat May 13 17:48:49 MDT 2006
One-line Summary: to implement MCT domains in each component
Note that tag had a problem with the land externals - another tag
was made using the correct land externals
===============================================================
===============================================================
Tag name: cam3_3_6
Originator(s): eaton, tcraig
Date: Mon May 8 11:26:41 MDT 2006
One-line Summary: Fix CAM log problem in CCSM runs
Purpose of changes:
. Fix problem with CAM log file not being written during CCSM run. The
fix was provide by tcraig in bugzilla bug 102.
. Set MPI_TYPE_MAX environment variable for testing on tempest
(sgi-o3800). This problem started in cam3_3_3 with FV code updates.
It is only a problem on machines that have a default value which is too
small. Currently setting MPI_TYPE_MAX=100000 is sufficient. The
problem has also been observed on phoenix (cray-x1e).
Bugs fixed (include bugzilla ID): 102
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: myself, tcraig
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/src/control/con_cam.F90
. rearrange initialization as follows:
call cpl_interface_init(cpl_fields_atmname, mpicom) ! call mpi_init
call spmdinit() ! set masterproc
call shr_msg_chdir('atm') ! all PE's chdir
call shr_msg_chStdin('atm') ! all PE's redirect unit 5
if (masterproc) call shr_msg_chStdout('atm') ! redir unit 6
call ESMF_Initialize() ! init ESMF time manager
call ccsm_seq_timer_init() ! init timing library
models/atm/cam/test/system/test_driver.sh
. add "export MPI_TYPE_MAX=100000" for testing on tempest
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: Only ran CCSM test -- PASS.
tempest: All PASS.
bangkok/lf95: no tests done -- none of the code mods are exercised on bangkok
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_3_5
Originator(s): sawyer, eaton
Date: Sun May 7 14:58:53 MDT 2006
One-line Summary: Fix omega calc for FV offline mode
Purpose of changes: Fix omega calc for FV offline mode
Bugs fixed (include bugzilla ID):
The fix for the FV omega calc added in cam3_3_3 was inadvertently left
out of the offline version of FV.
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: myself, sawyer
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
models/atm/cam/src/dynamics/fv/dyn_comp.F90
. call compute_vdot_gradp for both prognostic and offline FV
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: All PASS except:
040 bl355 TBL.sh fmo2dh off2x2.5 9 ................................FAIL! rc= 7
This baseline test is expected to fail. Only the OMEGA field is different.
tempest: All PASS.
bangkok/lf95: All PASS.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_3_4
Originator(s): Erik Kluzek
Date: May/01/2006
One-line Summary: Update clm, csm_share, and esmf_wrf_timemgr, use shr_file_mod for archiving
Purpose of changes: First step in moving toward using ESMF clock interface at top level
use shr_file_mod syntax for archive_dir.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: None
Describe any changes made to the namelist: None
List any changes to the defaults for the boundary datasets: None
Describe any substantial timing or memory changes: None
Code reviewed by: Tony Craig, Mariana Vertenstein, Brian Kauffman (csm_share and clm)
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
M models/drv/seq_mct_drv/seq_ccsm_drv.F90
M test/unit/control/configure
M test/unit/control/run_time_test
M tools/scam/scm_init/init_model.F90
M src/control/ccsm_msg.F90
M src/control/con_cam.F90
M src/control/history.F90
M src/control/filenames.F90
M src/control/ioFileMod.F90
M src/physics/waccm/chemistry.F90
Also use new version of CLM (shrgetput08_cammct05_clm3_expa_58), csm_share (share3_060428) and
esmf_wrf_timemgr (esmf_wrf_timemgr_060501)
Changes needed to use new versions of above libraries. Make sure a getfil is done before
a open-file. All programs need to do a ESMF_Initialize. ioFileMod uses shr_file_mod.F90 rather
than it's own functionality. filenames sets up archive_dir for shr_file_mod.F90 syntax (with
mss: prescript). Time unit tests use Lahey instead of PGI compiler on Linux.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: None
tempest: None
bangkok/lf95:
012 bl153 TBL.sh e64m outfrq24h -2 ................................FAIL! rc= 7
028 bl353 TBL.sh f2m outfrq24h -2 .................................FAIL! rc= 7
040 bl553 TBL.sh s64m outfrq24h -2 ................................FAIL! rc= 7
esmf_wrf_timemgr real-r8 julian day calculations are off by roundoff on Linux platforms.
It's off by roundoff on other platforms after 4-years. According to the unit tests it
remains within roundoff even for a very long simulation time.
CAM tag used for the baseline comparison tests if different than previous
tag: cam3_3_3
Summarize any changes to answers, i.e.,
- what code configurations: outfrq24
- what platforms/compilers: Linux lf95/pgi
If bitwise differences were observed, how did you show they were no worse
than roundoff? cprnc shows RMS difference in the 1e-17-1.e-20 range
Also unit-tests for time-manager shows that answers for julian day remain
bounded within roundoff even for a very long simulation time.
===============================================================
===============================================================
Tag name: cam3_3_3
Originator(s): sawyer, eaton
Date: Tue Apr 25 16:17:05 MDT 2006
One-line Summary: FV dycore improvements, fix omega diagnostic
Purpose of changes:
o The XY decomposition is now the only one which is
visible to the outside user (YZ is hidden). This
simplifies the CAM-specific part of FV (code inside
the dynamics/fv directory but outside of the portable
FV core) enormously.
o A dynamical core interface has
been written which is ESMF-like (init, run, final)
using an import and export container, and implementing
a dynamics state.
o Correct calculation of OMEGA, and related diagnostics.
o Resynchronized versions of FV dycores and PILGRIM
between GEOS5 and CAM (except GEOS5 vertical remapping).
o Numerous other software engineering improvements
(all bit-for-bit).
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes:
The timings for the FV dycore are roughly the same on bluevista. There
is about a 10% slowdown on phoenix.
Code reviewed by: sawyer, eaton
List all subroutines eliminated:
models/atm/cam/src/dynamics/fv/FVCAM_GridCompMod.F90
. renamed dyn_comp.F90
models/atm/cam/src/dynamics/fv/gmean.F90
. move this subroutine into the new module mean_module.F90
models/atm/cam/src/dynamics/fv/prognostics.F90
. prognostic state info now in dyn_import and dyn_export
List all subroutines added and what they do:
models/atm/cam/src/dynamics/fv/diag_module.F90
. contains the new routine compute_vdot_gradp which supplies a fix for
the OMEGA calculation
models/atm/cam/src/dynamics/fv/dyn_comp.F90
. was FVCAM_GridCompMod.F90
models/atm/cam/src/dynamics/fv/mean_module.F90
. new module contains the old gmean.F90 and new XY version of gmean
List all existing files that have been modified, and describe the changes:
models/atm/cam/src/control/history.F90
. add ability to work with FV fields using XY decomp -- it appears that the
ability to use YZ decomp has been maintained.
. generalize the beg/end index names
. allow history restart write/read to work with XY decomp
models/atm/cam/src/control/cam_comp.F90
models/atm/cam/src/control/restart.F90
models/atm/cam/src/control/startup_initialconds.F90
. add STAGGERED ifdef to support passing dynamics import/export types through top
level interfaces
models/atm/cam/src/dynamics/fv/diag_dynvar_ic.F90
. use XY decomp only
models/atm/cam/src/dynamics/fv/dp_coupling.F90
. use XY decomp only
. use dyn import/export types
models/atm/cam/src/dynamics/fv/dryairm.F90
. modify for XY decomp
models/atm/cam/src/dynamics/fv/dynamics_vars.F90
. move ESMF code inside GEOS_MODE ifdefs
. remove some tracertrans related code
models/atm/cam/src/dynamics/fv/epvd.F90
. update for XY decomp
models/atm/cam/src/dynamics/fv/fv_prints.F90
. converted to XY decomp
. ifdef out unused code to find max/min constituent values
models/atm/cam/src/dynamics/fv/inidat.F90
. use dyn_import/export_t containers
. remove dependency on prognostics
. input fields now scattered to XY decomp
models/atm/cam/src/dynamics/fv/inital.F90
. use dyn_import/export_t containers
. remove dependency on prognostics
models/atm/cam/src/dynamics/fv/io_dist.F90
. add methods needed for read/write of XY decomp from/to restart files
models/atm/cam/src/dynamics/fv/mapz_module.F90
. add option for cubic interpolation
models/atm/cam/src/dynamics/fv/metdata.F90
. remove unneeded ghosting of met_us, met_vs (it's done in cd_core).
models/atm/cam/src/dynamics/fv/p_d_adjust.F90
. modify to use XY decomp only
models/atm/cam/src/dynamics/fv/par_xsum.F90
. remove CPP conditional surrounding par_xsum_r4
models/atm/cam/src/dynamics/fv/pfixer.F90
. remove SPMD ifdefs that surrounded !$omp directives.
models/atm/cam/src/dynamics/fv/restart_dynamics.F90
. remove prognostics module
. add dyn_out as intent(in) arg to write_restart_dynamics
. modify read/write methods to work with XY decomp
. add dyn_in/out as intent(out) args to read_restart_dynamcis
models/atm/cam/src/dynamics/fv/spmd_dyn.F90
. remove tracertrans-related variables
models/atm/cam/src/dynamics/fv/stepon.F90
. remove prognostics by adding dyn_state to dyn_comp
. use dyn_import_t and dyn_export_t from dyn_comp
. get rid of lots of local xy arrays that are now part of the import/export
states
. add dyn_in, dyn_out dummy args to stepon_init
. the components of dyn_in are being initialized rather than local variables
. add dyn_in, dyn_out as dummy args to stepon_run1
. replace actual args to diag_dynvar_ic with components of dyn_out
(previously used prognostics module data).
. replace actual args to dyn_run with dyn_state, dyn_in, dyn_out.
Previously passed prognostics module vars and local xy vars.
. use dyn_out as actual arg to d_p_coupling instead of all the local xy
arrays.
. add dyn_in, dyn_out as intent(out) args to stepon_run2
. use dyn_in as actual arg to p_d_coupling instead of mix of prognostics
and local xy arrays
. add dyn_in, dyn_out as intent(out) args to stepon_run3
. pass components of dyn_out as actual args to fv_out
models/atm/cam/src/dynamics/fv/sw_core.F90
. change an index range in one calc
models/atm/cam/src/dynamics/fv/te_map.F90
. added some SPMD ifdefs
models/atm/cam/src/dynamics/fv/trac2d.F90
. add local variable frac to private declaration in !$omp and !CSD$ directives
models/atm/cam/src/dynamics/fv/uv3s_update.F90
. update to use XY decomp. Note that u3s and v3s are no longer passed with
ghost regions because the ghost region is taken care of in local variables.
models/atm/cam/src/dynamics/eul/dyn_grid.F90
models/atm/cam/src/dynamics/fv/dyn_grid.F90
models/atm/cam/src/dynamics/sld/dyn_grid.F90
. move ptimelevels from prognostics to here.
models/atm/cam/src/physics/cam1/phys_buffer.F90
models/atm/cam/src/physics/cam1/restart_physics.F90
. get ptimelevels from dyn_grid instead of prognostics
models/utils/pilgrim/mod_comm.F90
models/utils/pilgrim/parutilitiesmodule.F90
models/utils/pilgrim/puminterfaces.F90
models/utils/pilgrim/redistributemodule.F90
. merge with GEOS5 version of pilgrim
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: All PASS except:
019 bl331 TBL.sh f4c11dh pghgsul 9 ................................FAIL! rc= 7
022 bl332 TBL.sh f4sdh ghgrmp 9 ...................................FAIL! rc= 7
027 bl334 TBL.sh f4dh adia 9 ......................................FAIL! rc= 7
030 bl336 TBL.sh wm4h outfrq3s 9 ..................................FAIL! rc= 7
034 bl351 TBL.sh f2h outfrq24h -2 .................................FAIL! rc= 7
037 bl354 TBL.sh fm2dh outfrq3s 9 .................................FAIL! rc= 7
tempest: All PASS except:
017 bl314 TBL.sh wg10dh outfrq3s 9 ................................FAIL! rc= 7
021 bl331 TBL.sh f4c11dh pghgsul 9 ................................FAIL! rc= 7
024 bl332 TBL.sh f4sdh ghgrmp 9 ...................................FAIL! rc= 7
029 bl335 TBL.sh f4dh idphys 9 ....................................FAIL! rc= 7
031 bl352 TBL.sh f2o outfrq24h -2 .................................FAIL! rc= 7
bangkok/lf95: All PASS except:
016 bl311 TBL.sh f10c8mdm ttrac 9 .................................FAIL! rc= 7
024 bl312 TBL.sh f10sdm ghgrmp 9 ..................................FAIL! rc= 7
028 bl353 TBL.sh f2m outfrq24h -2 .................................FAIL! rc= 7
All the baseline tests that fail on bluesky, tempest, and bangkok do so
due to changes in the diagnostic OMEGA and OMEGAT fields.
CAM tag used for the baseline comparison tests if different than previous
tag: cam3_3_1
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_3_2
Originator(s): Jim Edwards
Date: 4/24/2006
One-line Summary: Eliminate common block comlun
Purpose of changes: code clean up
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: self
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
In control directory:
file comlun.h removed
readinitial.F90 - converted to netcdf f90 interface, modified for ncol format
initial file
runtime_opts.F90, history.F90 - removed reference to ncid_ini and ncid_topo
startup_initial_cond.F90 - made ncid_ini and ncid_topo module variables
added function interfaces initial_file_get_id and topo_file_get_id
to allow read only access to these.
ioFileMod.F90 - made info messages print from masterproc only
error_messages.F90 - added an option integer variable in handle_ncerr to
print the error line number
physics/waccm/tgcm_forcing.F90: removed unused reference to comlun.h
physics/waccm/chemistry.F90: removed unused reference to ncid_trc and comlun.h
physics/cam1/co2_data_flux.F90: removed unused reference to comlun.h
physics/cam1/boundarydata.F90: changed calls to handle_ncerr to add line number
dynamics/*/inidat.F90: changed read_inidat to pass in ncid_ini, ncid_topo
{eul,sld}/spegrd.F90
{eul,sld}/linemsdyn.F90
sld/tfilt_massfix.F90 - removed unused refs to comlun.h
tools/scam/scm_init/inital.F90
tools/scam/scm_init/init_model.F90 - removed refs to comlun.h
ice/csim4/ice_comp.F90
ice/csim4/ice_data.F90
ocn/som/somini.F90
ocn/som/ocn_comp.F90
ocn/som/somint.F90
ocn/dom/ocn_comp.F90
ocn/dom/sst_data.F90 - replaced refs to comlun.h with calls to functions or
local variables as appropriate.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: All pass
tempest: All pass
bangkok/lf95: All pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_3_1
Originator(s): Mariana Vertnestein
Date: 4/18/2006
One-line Summary: bug fixes necessary for CCSM phoenix
Purpose of changes: Implemented several minor bugs fixes necessary
to run CAM in CCSM on phoenix (Cray X1)
Bugs fixed (include bugzilla ID): None
Describe any changes made to build system: None
Describe any changes made to the namelist: None
List any changes to the defaults for the boundary datasets: None
Describe any substantial timing or memory changes: None
Code reviewed by: myself, Jon Wolfe, Brian Eaton
List all subroutines eliminated: None
List all subroutines added and what they do: None
List all existing files that have been modified, and describe the changes:
M src/control/con_cam.F90
Modifications put in so that only the lead/master MPI process
redirects its stdout to the appropriate component model log file when
running fully coupled, instead of having every process "open" the same
log file. This change solves the problem of logs getting trashed on
phoenix (which has been recently experienced). The only down side is
that warning and error messages generated by other processes
(non-master processes) get stuck in the job's stdout file, so users
may want to archive job output files as well as component model logs.
In addition, when debugging problems with the model system, certain
error messages which may have previously appeared in component model
logs may now appear in the job output.
M src/dynamics/fv/trac2d.F90
M src/dynamics/fv/cd_core.F90
Removed streaming directives for COUP_CSM mode where a routine is called
from within a streaming region. This problem does not exist in stand-alone
CAM mode, but causes the system to hang or crash in coupled mode. This is
bug will be reported to Cray, but a fix has currently been implemented so
that production runs can occur on phoenix.
the following changes was made for cpp directives:
#if defined(USE_OPM => #if (!definedUSE_OMP) && (!defined COUP_CSM)
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: no pre-tag testing was done
tempest: no pre-tag testing was done
bangkok/lf95: no pre-tag testing was done
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_3_0
Originator(s): Mariana Vertenstein
Date: 4/18/2006
One-line Summary: Directory restructuring for sequential ccsm
Purpose of changes:
Create new directory structure and mct file renaming for
the next step in the creation of a sequential ccsm starting
from stand-alone cam and for the next step in the ESMF stage
1 evaluation effort.
Bugs fixed (include bugzilla ID): None
Describe any changes made to build system:
bld/configure was changed to take into account the new directory
structure and to remove the obsolete mode for the sea ice package "ccmice"
Describe any changes made to the namelist: None
List any changes to the defaults for the boundary datasets: None
Describe any substantial timing or memory changes: None
Code reviewed by: Myself, Brian Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
The following lists the directory restucturing and file renaming that was done
D atm/cam/src/ocnsice/dom/sst_data.F90 => A models/ocn/dom/sst_data.F90
D atm/cam/src/ocnsice/dom/ocn_types.F90 => A models/ocn/dom/ocn_types.F90
D atm/cam/src/ocnsice/dom/albocean.F90 => A models/ocn/dom/albocean.F90
D atm/cam/src/ocnsice/dom/print_coverage.F90 => A models/ocn/dom/print_coverage.F90
D atm/cam/src/ocnsice/dom/parpbl.h => A models/ocn/dom/parpbl.h
D atm/cam/src/ocnsice/dom/ocn_comp.F90 => A models/ocn/dom/ocn_comp.F90
D atm/cam/src/ocnsice/dom/albedo.h => A models/ocn/dom/albedo.h
D atm/cam/src/ocnsice/dom/wtrc_flxoce.F90 => A models/ocn/dom/wtrc_flxoce.F90
D atm/cam/src/ocnsice/dom/MCT_ocn_comp.F90 => A models/ocn/dom/flxoce.F90
D atm/cam/src/ocnsice/dom/flxoce.F90 => A models/ocn/dom/ocn_comp_mct.F90
D atm/cam/src/ocnsice/dom/srfoce.F90 => A models/ocn/dom/srfoce.F90
D atm/cam/src/ocnsice/som/ocn_types.F90 => A models/ocn/som/ocn_types.F90
D atm/cam/src/ocnsice/som/albocean.F90 => A models/ocn/som/albocean.F90
D atm/cam/src/ocnsice/som/mixed_layer_globalcalcs.F90 => A models/ocn/som/mixed_layer_globalcalcs.F90
D atm/cam/src/ocnsice/som/ocn_srf.F90 => A models/ocn/som/ocn_srf.F90
D atm/cam/src/ocnsice/som/parpbl.h => A models/ocn/som/parpbl.h
D atm/cam/src/ocnsice/som/somint.F90 => A models/ocn/som/somint.F90
D atm/cam/src/ocnsice/som/ocn_comp.F90 => A models/ocn/som/ocn_comp.F90
D atm/cam/src/ocnsice/som/ocean_data.F90 => A models/ocn/som/ocean_data.F90
D atm/cam/src/ocnsice/som/somini.F90 => A models/ocn/som/somini.F90
D atm/cam/src/ocnsice/som/MCT_ocn_comp.F90 => A models/ocn/som/flxoce.F90
D atm/cam/src/ocnsice/som/flxoce.F90 => A models/ocn/som/ocn_comp_mct.F90
D ice/csim4/MCT_ice_comp.F90 => A models/ice/csim4/ice_comp_mct.F90
A models/drv/seq_mct/seq_mct_mod.F90
D atm/cam/src/control/MCT_atmocn_cpl.F90 => A models/drv/seq_mct/map_atmocn_mct.F90
D atm/cam/src/control/MCT_atmlnd_cpl.F90 => A models/drv/seq_mct/map_atmlnd_mct.F90
D atm/cam/src/control/MCT_iceocn_cpl.F90 => A models/drv/seq_mct/map_iceocn_mct.F90
D atm/cam/src/control/MCT_atmice_cpl.F90 => A models/drv/seq_mct/map_atmice_mct.F90
D atm/cam/src/control/MCT_lndhub_comp.F90 => A models/drv/seq_mct/mrg_x2l_mct.F90
D atm/cam/src/control/MCT_atmhub_comp.F90 => A models/drv/seq_mct/mrg_x2a_mct.F90
D atm/cam/src/control/MCT_icehub_comp.F90 => A models/drv/seq_mct/mrg_x2i_mct.F90
D atm/cam/src/control/MCT_ocnhub_comp.F90 => A models/drv/seq_mct/mrg_x2o_mct.F90
D atm/cam/src/control/MCT_seq.F90 => A models/drv/seq_mct/seq_mct_init.F90
D atm/cam/src/control/cam.F90 => A models/drv/seq_mct_drv/seq_ccsm_drv.F90
D atm/cam/src/control/seq_fields_indices.F90 => A models/drv/seq_flds_indices/seq_flds_indices.F90
D atm/cam/src/control/seq_fields_mod.F90 => A models/drv/seq_flds/default/seq_flds_mod.F90
D atm/cam/src/control/MCT_atm_comp.F90 => A models/atm/cam/src/control/atm_comp_mct.F90
List all existing files that have been modified, and describe the changes:
M atm/cam/tools/scam/configure
M atm/cam/tools/scam/scm_init/init_model.F90
M atm/cam/tools/scam/scm_init/scam_srfdata_MCT.F90
M atm/cam/tools/scam/scm_init/scam_run.F90
M atm/cam/bld/configure
Modified these files to work with above directory restructuring and file renaming
M models/SVN_EXTERNAL_DIRECTORIES
Modified to use clm branch tag cammct05_clm3_expa_58
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. => All failed tests must be justified.
bluesky: None
tempest: None
bangkok/lf95: None
CAM tag used for the baseline comparison tests if different than previous
tag: cam3_2_60 (this is the previous tag)
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_2_60
Originator(s): Mariana Vertenstein
Date: 4/11/2006
One-line Summary: Introduction of MCT CAM-surface coupling
Purpose of changes:
Incoporate MCT cam-surface component coupling for all
CAM surface components. Introduce the appropriate thin-layer
design for MCT coupling that can also be extended to ESMF.
Bugs fixed (include bugzilla ID): None
Describe any changes made to build system: None
Describe any changes made to the namelist: None
List any changes to the defaults for the boundary datasets: None
Describe any substantial timing or memory changes:
***These have to still be determined***
Code reviewed by:
Erik Kluzek
List all subroutines eliminated:
D atm/cam/src/physics/cam1/camhub_comp.F90
List all subroutines added and what they do:
A atm/cam/src/control/MCT_atmhub_comp.F90
A atm/cam/src/control/MCT_lndhub_comp.F90
A atm/cam/src/control/MCT_ocnhub_comp.F90
A atm/cam/src/control/MCT_icehub_comp.F90
-introduced these routines to merge necessary
input for atm, lnd, ocn and ice states
(e.g. MCT_atm_hub_comp.F90 merges lnd,ocn and ice
states and fluxes to create the necessary input
atm attribute vector)
These hub component routines act only on MCT attribute vectors
A atm/cam/src/control/MCT_atmlnd_cpl.F90
A atm/cam/src/control/MCT_iceocn_cpl.F90
A atm/cam/src/control/MCT_atmocn_cpl.F90
A atm/cam/src/control/MCT_atmice_cpl.F90
A atm/cam/tools/scam/scm_init/scam_srfdata_MCT.F90
-introduced new MCT couplers
(e.g. MCT_atmlnd_cpl.F90 maps between atm and lnd domains)
These couplers act only on MCT attribute vectors
A atm/cam/src/ocnsice/dom/ocn_types.F90
-introduced new dom specific data structures (ocn_in, ocn_out)
so that MCT could map directory to the dom input/output states
A atm/cam/src/ocnsice/som/ocn_types.F90
-introduced new som specific data structures (ocn_in, ocn_out)
so that MCT could map directory to the dom input/output states
A ice/csim4/ice_types.F90
-introduced new csim specific data structures (ice_in, ice_out)
so that MCT could map directory to the dom input/output states
A ice/csim4/MCT_ice_comp.F90
-introduced MCT_ocn_comp.F90 that map MCT attribute vectors
to ice_in and ice_out to MCT attribute vectors
A atm/cam/src/ocnsice/dom/MCT_ocn_comp.F90
-introduced MCT_ocn_comp.F90 that map MCT attribute vectors
to ocn_in and ocn_out to MCT attribute vectors
A atm/cam/src/ocnsice/som/MCT_ocn_comp.F90
-introduced MCT_ocn_comp.F90 that map MCT attribute vectors
to ocn_in and ocn_out to MCT attribute vectors
A atm/cam/src/control/seq_fields_indices.F90
-contains integer indices for all possible character strings that
represent cam and cam surface model couling
A atm/cam/src/control/con_cam.F90
-replaced part of cam.F90 that dealt with running cam in COUP_CSM mode
List all existing files that have been modified, and describe the changes:
M ice/csim4/ice_data.F90
-removed srfflx_state2d from input argument and replaced it with frac
(frac a frac_t defined type now specific to ice/csim4)
M ice/csim4/ice_dh.F
-renamed surface_state2d -> ice_in (where ice_in is now of type ice_in_t not surface_state)
A ice/csim4/print_coverage.F90 -> D atm/cam/src/ocnsice/som/print_coverage.F90
M ice/csim4/ice_srf.F90
-renamed surface_state2d -> ice_in (where ice_in is now of type ice_in_t not surface_state)
M ice/csim4/ice_diagnostics.F
-renamed surface_state2d -> ice_in (where ice_in is now of type ice_in_t not surface_state)
D ice/csim4/camice.F90-> A ice/csim4/ice_comp.F90
-renamed camice.F90 -> ice_comp.F90 and introduced ice_in and ice_out
states to replace srfflx_parm2d_ocn, srfflx_state2d.
-added functionality to read landfrac from cam initial file and store it
as a module variable, rather than use it from camsrf.
M atm/cam/tools/scam/configure
M atm/cam/tools/scam/testscript
M atm/cam/tools/scam/scm_init/init_model.F90
M atm/cam/tools/scam/scm_init/scam_srfdata.F90
M atm/cam/tools/scam/scm_init/scam_run.F90
M atm/cam/tools/scam/scm_init/inital.F90
M atm/cam/tools/scam/scm_init/scamMod.F90
M atm/cam/tools/scam/ui/configure
-incorporated changes into above scam routines so that
they could work with new MCT coupling to surface models
M atm/cam/tools/cprnc/Makefile
M atm/cam/src/control/MCT_atm_comp.F90
-introduced MCT_ocn_comp.F90 that map MCT attribute vectors
to cam_in and cam_out to MCT attribute vectors
M atm/cam/src/control/ccsm_msg.F90
-extended restart functionality to use logic previously in restart_physics
M atm/cam/src/control/cam.F90
-this is now a top level MCT coupling driver for stand-alone CAM
M atm/cam/src/control/seq_fields_mod.F90
-generalized routine to account for cam-ice and cam-ocn coupling
M atm/cam/src/control/restart.F90
-made changes to removed srfflx_parm and srfflx_state from camsrfexch_typees
M atm/cam/src/control/startup_initialconds.F90
-removed call to close ncid_topo
-removed srfflx_state2d argument to read_inidat and initial_conds
M atm/cam/src/control/cam_comp.F90
-removed srfflx_parm from camsrfexch_types use
-removed cam_in from call to cam_initial and read_restart
-added call to hub2atm_alloc
M atm/cam/src/physics/cam1/comsrf.F90
-removed snowhland, tsocn and renamed landfrac_field->landfrac_glob
M atm/cam/src/physics/cam1/diagnostics.F90
-removed outfld calls for TBOT,SNOWHLND,TSOCN
-removed comsrf uses for tsocn
-removed snowhland as input argument
M atm/cam/src/physics/cam1/tphysbc.F90
-removed snowh ans input argument
M atm/cam/src/physics/cam1/restart_physics.F90
-removed calls to output srfflx_state2d fields - moved these to
ccsm_msg.F90 and MCT_atmhub_comp.F90.
M atm/cam/src/physics/cam1/physpkg.F90
D atm/cam/src/physics/cam1/camsrfexch_types.F90 -> A atm/cam/src/control/camsrfexch_types.F90
-removed module routines bounding, verify_fractions, srfflx_parm_reset,
srfcomp2hub_alloc
-renamed atm2srf_alloc->atm2hub->alloc
-removed srfflx_parm derived type and added snowhland as part of srfflx_state
definition
D atm/cam/src/physics/cam1/srfxfer.F90 -> A atm/cam/src/control/srfxfer.F90
-cosmetic changes only
M atm/cam/src/physics/cam1/tphysac.F90
-removed snowh as input argument
M atm/cam/src/ocnsice/dom/sst_data.F90
- removed landfrac from argument list
M atm/cam/src/ocnsice/dom/albocean.F90
- removed landfrac from argument list
D atm/cam/src/ocnsice/dom/camoce.F90 -> A atm/cam/src/ocnsice/dom/ocn_comp.F90
-renamed camoce.F90 -> ocn_comp.F90 and introduced ocn_in and ocn_out
states to replace srfflx_parm2d_ocn, srfflx_state2d.
-added functionality to read landfrac from cam initial file and store it
as a module variable, rather than use it from camsrf.
-introuced ocn_IC_OUTPUT routine to output TSOCN to initial file.
M atm/cam/src/ocnsice/som/albocean.F90
- removed landfrac from argument list
D atm/cam/src/ocnsice/som/print_coverage.F90
-removed routine, no longer needed
M atm/cam/src/ocnsice/som/mixed_layer_globalcalcs.F90
-added landfrac_glob to argument list, replace landfrac_field->landfrac_glob
M atm/cam/src/ocnsice/som/ocn_srf.F90
-removed Tffresh, rhow, cp_ocn from ice_constants use statement and used
SHR_CONST_ROWSW, SHR_CONST_CPSW and SHR_CONST_TKFREZ from shr_const_mod instead
D atm/cam/src/ocnsice/som/somoce.F90 -> A atm/cam/src/ocnsice/som/ocn_comp.F90
-renamed somoce.F90 -> ocn_comp.F90 and introduced ocn_in and ocn_out
states to replace srfflx_parm2d_ocn, srfflx_state2d.
-added functionality to read landfrac from cam initial file and store it
as a module variable, rather than use it from camsrf.
-introuced ocn_IC_OUTPUT routine to output TSOCN to initial file.
-added landfrac_glob as argument to gmean and check_conservation calls
-repalce Focn->ocn_out%frzmlt
-removed use of camice (Focn, frzmlt, aice, sicthk, snowhice now obtained
via ocn_in and ocn_out)
-removed ice_constants use for rhow, cp_ocn and used
SHR_CONST_ROWSW, SHR_CONST_CPSW from shr_const_mod instead
M atm/cam/src/chemistry/trop_mozart/mo_gas_phase_chemdr.F90
M atm/cam/src/chemistry/trop_mozart/chemistry.F90
-removed snowhland from camsrf use and added to argument list (part of cam_in)
M atm/cam/src/dynamics/sld/stepon.F90
M atm/cam/src/dynamics/eul/stepon.F90
-removed srfflx_state and srfflx_parm use statements
M atm/cam/src/dynamics/sld/inidat.F90
M atm/cam/src/dynamics/eul/inidat.F90
M atm/cam/src/dynamics/fv/inidat.F90
-changed varialbe names (surface_state->cam_out)
-introduced landfrac_glob and removed use of landfrac from comsrf
-removed srfflx_state use from camsrfexch_types
M atm/cam/src/dynamics/sld/inital.F90
M atm/cam/src/dynamics/eul/inital.F90
M atm/cam/src/dynamics/fv/inital.F90
-changed varialbe names (surface_state->cam_out)
-removed srfflx_state2d as input argument
-removed call to hub2atm_alloc and renamed atm2srf_alloc->atm2hub_alloc
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky:
005 bl131 TBL.sh e32c11dh pghgsul 9 ...............................FAIL! rc= 7
008 bl132 TBL.sh e32sdh ghgrmp 9 ..................................FAIL! rc= 7
015 bl151 TBL.sh e64h outfrq24h -2 ................................FAIL! rc= 7
019 bl331 TBL.sh f4c11dh pghgsul 9 ................................FAIL! rc= 7
022 bl332 TBL.sh f4sdh ghgrmp 9 ...................................FAIL! rc= 7
030 bl336 TBL.sh wm4h outfrq3s 9 ..................................FAIL! rc= 7
034 bl351 TBL.sh f2h outfrq24h -2 .................................FAIL! rc= 7
037 bl354 TBL.sh fm2dh outfrq3s 9 .................................FAIL! rc= 7
040 bl355 TBL.sh fmo2dh off2x2.5 9 ................................FAIL! rc= 7
044 bl531 TBL.sh s32c11dh pghgsul 9 ...............................FAIL! rc= 7
047 bl532 TBL.sh s32sdh ghgrmp 9 ..................................FAIL! rc= 7
054 bl551 TBL.sh s64h outfrq24h -2 ................................FAIL! rc= 7
All these tests only contain differences in the following fields
LHFLXOI, SHFLXOI, FSNSOI
with RMS values on the order of 1.e-15
In addition, TBL.f2h.outfrq24h contained a difference in QFLX on the order of 1.-e23
So even though the TBL tests fail, this tag does not change answers.
tempest:
014 bl152 TBL.sh e64o outfrq24h -2 ................................FAIL! rc= 7
diffs set in at NSTEP=47
021 bl331 TBL.sh f4c11dh pghgsul 9 ................................FAIL! rc= 7
only QRS is different in cprnc.out with an RMS on the order of 3.e-24
024 bl332 TBL.sh f4sdh ghgrmp 9 ...................................FAIL! rc= 7
only SNOWHICE is different in cprnc.out with an RMS on the order of 1.e-23
031 bl352 TBL.sh f2o outfrq24h -2 .................................FAIL! rc= 7
only VD01 is different in cprnc.out with an RMS on the order of 4.e-25
bangkok/lf95:
012 bl153 TBL.sh e64m outfrq24h -2 ................................FAIL! rc= 7
diffs set in at NSTEP=29
016 bl311 TBL.sh f10c8mdm ttrac 9 .................................FAIL! rc= 7
differences are only in SHOWHICE with RMS value of 5.e-21
028 bl353 TBL.sh f2m outfrq24h -2 .................................FAIL! rc= 7
diffs set in at nstep=9
040 bl553 TBL.sh s64m outfrq24h -2 ................................FAIL! rc= 7
diffs set in at nstep=3
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_2_59
Originator(s): mvr, eaton, edwards
Date: 4/6/06
One-line Summary: collection of bug fixes
Purpose of changes:
bug fixes for ccsm testing, cray directive on phoenix, compiler workaround
on bluevista, coupled runs on tempest; added cron scripts to repository
Bugs fixed (include bugzilla ID):
bug#68 is resolved
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: myself, eaton, edwards
List all subroutines eliminated:
D models/atm/cam/test/system/td_nightly_aix.sh
D models/atm/cam/test/system/td_nightly_linux.sh
- post-tag cron jobs were renamed
List all subroutines added and what they do:
A + models/atm/cam/test/system/posttag_cron_bangkok.sh
A + models/atm/cam/test/system/posttag_cron_bluesky.sh
A models/atm/cam/test/system/posttag_cron_lightning.sh
A models/atm/cam/test/system/posttag_cron_bluevista.sh
- post-tag cron job scripts for machine-specific testing of cam
List all existing files that have been modified, and describe the changes:
M models/atm/cam/test/system/TSB.ccsm.sh
M models/atm/cam/test/system/TCB.ccsm.sh
- mods to make cam's ccsm test scripts compatible with changes in ccsm scripts
M models/atm/cam/test/system/test_driver.sh
M models/atm/cam/test/system/CAM_runcmnd.sh
- test scripts now use unix utility 'hostname' to determine local machine
M models/atm/cam/src/physics/cam1/dmsbnd.F90
- removed debug print statement left behind - trips up coupled runs on tempest
M models/atm/cam/src/physics/cam1/physpkg.F90
- added workaround for code introduced in cam3_2_58 which trips xlf bug on bluevista
M models/atm/cam/src/dynamics/fv/fv_prints.F90
- fix of bug in cpp definition CPP_PRT_PREFIX on phoenix
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: all PASS
tempest: all PASS
bangkok/lf95: all PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_2_58
Originator(s): Jim Edwards
Date: 4/4/2006
One-line Summary: changes for ncol dataset support
Purpose of changes: Allows data input on non-rectangular grids using an alternative input file format
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets: none, no new boundary datasets
are provided at this time
Describe any substantial timing or memory changes:
Code reviewed by: myself, Eaton
List all subroutines eliminated: renamed file zonalbndydata.F90 to boundarydata.F90
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
renamed file zonalbndydata.F90 to boundarydata.F90
dmsbnd.F90 prescribed_aerosols.F90 : added support for ncol based input dataset
chemistry.F90 ozone_data.F90: changed module names from zonalndydata to boundarydata
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: one fails
055 cs999 TCS.ccsm.sh ER.01a T31_gx3v5 K ..........................FAIL! rc= 2
That's due to a change in the CCSM scripts in ccsm3_1_beta24. The test
passes when run in ccsm3_1_beta23. Mat is modifying the CAM test to work
with the new CCSM scripts.
tempest: all pass
bangkok/lf95: all pass
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_2_57
Originator(s): pworley
Date: Sun Apr 2 12:22:18 MDT 2006
One-line Summary: Performance optimizations, primarily for Cray-X1E
Purpose of changes:
To restore X1E performance lost in recent check-ins; to fix X1E and XT3
bugs.
Bugs fixed (include bugzilla ID):
PGI compile problem; aqua planet error on X1E.
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes:
Performance on X1E doubled compared to 3_2_56 when running T85 on 128
processors.
Code reviewed by: myself, eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
a) Vectorized routines chunk_to_buff and buff_to_chunk used by MCT in
land/atmosphere communication and in coupler/atmosphere communication:
M atm/cam/src/physics/cam1/phys_grid
- buff_to_chunk, chunk_to_buff: interface clean up and vector-friendly
reimplementation
M atm/cam/src/control/ccsm_msg.F90
M atm/cam/src/control/MCT_atm_comp
- changed parameters in calls to buff_to_chunk and chunk_to_buff
b) Fixed bug on X1E caused by diagnostic message output inside streamed
code segment
M atm/cam/src/dynamics/fv/fill_module.F90
- surrounded (non fatal) error test with ifdef DEBUG cpp logic
c) Introduced workaround for PGI compiler bug
M lnd/clm2/src/biogeophys/Hydrology2Mod.F90
- removed unnecessary "use" of spmdMod
d) Optimized performance of MCT Rearranger routine
M utils/mct/mct/m_Rearranger
- added new optional parameter ALLTOALL and modified code in rearrange_
to implement an MPI_Alltoallv option for interprocessor communication.
- reordered the send/receive pattern for the non-ALLTOALL implementation
to decrease link and processor contention.
- Reduced number of allocate/deallocate requests.
M utils/mct/mpi-serial/collective.c
M utils/mct/mpi-serial/mpi.h
- add mpi_alltoallv interfaces (from Ray Loy)
M lnd/clm2/src/main/MCT_atmlnd_cpl
- added optional parameters VECTOR and ALLTOALL to calls to MCT_Rearrange,
setting VECTOR to .true. if CPP_VECTOR is defined, and setting
ALLTOALL to .true. if SYSUNICOS is defined. Otherwise they are
set to .false. .
Note that ALLTOALL=.true. may be a better choice for all systems,
but I have only verified this on the X1E.
e) Improved vector performance in CLM/MCT interface.
M lnd/clm2/src/main/MCT_lnd_comp
- made minor change to index logic to improve vectorization on the X1E
f) Improved performance (both vector and nonvector) in routines that
map to/from CLM grid
M areaMod
- added support for mapping multiple fields in one call to
gridmap_mapparray
- reimplemented gridmap_mapparray to improve vector performance,
incuding changing calling interface
M clm_atmlnd
- modified logic in clm_mapa2l and clm_mall2a to call gridmap_array once
for all fields
- made other minor modifications to improve vectorization
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: All PASS except:
055 cs999 TCS.ccsm.sh ER.01a T31_gx3v5 K ..........................FAIL! rc= 2
That's due to a change in the CCSM scripts in ccsm3_1_beta24. The test
passes when run in ccsm3_1_beta23. Mat is modifying the CAM test to work
with the new CCSM scripts.
tempest: All PASS
bangkok/lf95: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_2_56
Originator(s): mvr
Date: 3/27/06
One-line Summary:
test_driver.sh made to run interactive; test scripts modified for use
on lightning, bluevista; new tests added; some bugs fixes
Purpose of changes:
wanted to test on lightning, bluevista
Bugs fixed (include bugzilla ID):
ccsm tests work again
Describe any changes made to build system:
LAPACK library replaced with ESSL library as default for WACCM_MOZART
configurations on AIX
Describe any changes made to the namelist:
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: myself, eaton
List all subroutines eliminated:
D atm/cam/test/system/input_tests_aix
D atm/cam/test/system/input_tests_linux
D atm/cam/test/system/input_tests_irix
D atm/cam/test/system/input_tests_aix_nightly
D atm/cam/test/system/input_tests_linux_nightly
- these files were renamed
List all subroutines added and what they do:
A + atm/cam/test/system/tests_pretag_bluesky
A + atm/cam/test/system/tests_pretag_bangkok
A + atm/cam/test/system/tests_pretag_tempest
A + atm/cam/test/system/tests_posttag_bluesky
A + atm/cam/test/system/tests_posttag_bangkok
- newly renamed files
A atm/cam/test/system/tests_posttag_bluevista
A atm/cam/test/system/tests_posttag_lightning
- new files containing tests for respective platforms
A atm/cam/test/system/config_files/f1.9h
- new config options file for testing fv 1.9x2.5 resolution
A atm/cam/test/system/TPF.sh
- new test script for performance testing
List all existing files that have been modified, and describe the changes:
M atm/cam/test/system/test_driver.sh
- made to run interactive (ie will create/spawn command file to machine's
batch queue); added support for lightning, bluevista; added env var to
specify ccsm tag to be used for testing
M atm/cam/test/system/CAM_runcmnd.sh
- added logic for setting run command on lightning, bluevista
M atm/cam/test/system/TCB.sh
- uses new env var for setting options to configure
M atm/cam/test/system/TSB.ccsm.sh
- bug with ccsm test now fixed - files/dirs now compatible with ccsm scripts
M atm/cam/test/system/TCB.ccsm.sh
- now specifies resources for ccsm test; general cleanup
M atm/cam/test/system/TBL.sh
- minor cleanup
M atm/cam/test/system/td_nightly_aix.sh
- posttag cron script modified with new call to test_driver.sh;
now spawns testing on lightning, bluevista
M atm/cam/test/system/td_nightly_linux.sh
- posttag cron script modified with new call to test_driver.sh
M atm/cam/test/system/input_tests_master
- added performance test, ccsm test w/bgc, ccsm test w/trop_mozart chemistry
M atm/cam/test/system/tag_email.sh
- posttag cron script now used 'mail' to notify developers of new tag
M atm/cam/bld/configure
- now uses ESSL rather than LAPACK library for waccm_mozart on aix
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky:
055 cs999 TCS.ccsm.sh ER.01a T31_gx3v5 K ..........................FAIL! rc= 2
- compatibility bug w/ccsm scripts fixed, this test should pass going forward
tempest:
- none
bangkok/lf95:
004 bl111 TBL.sh e8c8mdm ttrac 9 ..................................FAIL! rc= 5
008 bl112 TBL.sh e8sdm ghgrmp 9 ...................................FAIL! rc= 5
010 bl113 TBL.sh e8pdm aqpgro 3 ...................................FAIL! rc= 5
012 bl153 TBL.sh e64m outfrq24h -2 ................................FAIL! rc= 5
016 bl311 TBL.sh f10c8mdm ttrac 9 .................................FAIL! rc= 5
024 bl312 TBL.sh f10sdm ghgrmp 9 ..................................FAIL! rc= 5
026 bl313 TBL.sh f10pdm aqpgro 3 ..................................FAIL! rc= 5
028 bl353 TBL.sh f2m outfrq24h -2 .................................FAIL! rc= 5
032 bl511 TBL.sh s8c8mdm ttrac 9 ..................................FAIL! rc= 5
036 bl512 TBL.sh s8sdm ghgrmp 9 ...................................FAIL! rc= 5
038 bl513 TBL.sh s8pdm aqpgro 3 ...................................FAIL! rc= 5
040 bl553 TBL.sh s64m outfrq24h -2 ................................FAIL! rc= 5
- baseline tests failed due to change in env vars used in testing; these tests
should pass going forward
CAM tag used for the baseline comparison tests if different than previous
tag: ccsm testing used cam3_2_51
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_2_55
Originator(s): Francis Vitt
Date: Fri Mar 24 16:51:04 MST 2006
One-line Summary: Fixed bug in mo_drydep.F90
Purpose of changes: To compile and run on lighting
Bugs fixed (include bugzilla ID): 60
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes:
Code reviewed by: Jeff Lee ( Yen-Huei Lee)
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
U models/atm/cam/src/chemistry/trop_mozart/mo_drydep.F90
changed parameter n_drydep_species from 52 to 53
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky:
037 bl354 TBL.sh fm2dh outfrq3s 9 .................................FAIL! rc= 7
040 bl355 TBL.sh fmo2dh off2x2.5 9 ................................FAIL! rc= 7
Dry deposition species have changed. Bit for bit comparison to
previous version is expected to fail.
055 cs999 TCS.ccsm.sh ER.01a T31_gx3v5 K ..........................FAIL! rc= 2
CCSM scripts have changed causing this test to fail
tempest: All PASS
bangkok/lf95: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_2_54
Originator(s): Jim Edwards
Date: 03/21/06 12:36PM
One-line Summary: addresses bug reported in http://bugs.cgd.ucar.edu/show_bug.cgi?id=56
Purpose of changes:
Bugs fixed (include bugzilla ID): 56 ccsm cannot read co2 data file
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by:
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
co2_data_flux.F90 changed variable for date and sec to explicitly allocated
arrays.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
This file is used only with ccsm, no stand-alone tests are available. We have confirmed the
change using ccsm3_1_beta23.
===============================================================
===============================================================
Tag name: cam3_2_53
Originator(s): Francis Vitt
Date: Thu Mar 16 09:00:35 MST 2006
One-line Summary:
Changes to trop_mozart chemistry for CCSM coupled mode.
Purpose of changes:
To run CAM with trop_mozart in CCSM and pass acceptance tests
Bugs fixed (include bugzilla ID):
Describe any changes made to build system:
Describe any changes made to the namelist:
New initial coniditions files are used from trop_mozart chemistry.
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Brian Eaton
List all subroutines eliminated:
List all subroutines added and what they do:
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/DefaultCAMEXPNamelist.xml
changed IC data files for trop_mozart
models/atm/cam/src/chemistry/trop_mozart/chemistry.F90
Changed loop indexing where constituent fields are output
to history file to be compatable with BGC package
models/atm/cam/src/chemistry/trop_mozart/mo_drydep.F90
Added H2 to the dry deposition species list
models/atm/cam/src/chemistry/trop_mozart/mo_srf_emissions.F90
Initialize surface fluxes to zero before setting to data
models/atm/cam/src/chemistry/trop_mozart/mo_strato_sad.F90
Increaesed the length of filename to 265 characters
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky:
037 bl354 TBL.sh fm2dh outfrq3s 9 .................................FAIL! rc= 7
040 bl355 TBL.sh fmo2dh off2x2.5 9 ................................FAIL! rc= 7
These bit-for-bit comparisons to the previous version of trop_mozart
chemistry are expected fail since H2 was added to dry deposition
and different initial conditions are used
055 cs999 TCS.ccsm.sh ER.01a T31_gx3v5 K ..........................FAIL! rc= 2
CCSM scripts have changed causing this test to fail
tempest: All PASS
bangkok/lf95: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name:cam3_2_52
Originator(s): Jim Edwards
Date: Thu Mar 9 17:11:49 MST 2006
One-line Summary: netcdf f90 interface, interpaerosols update
Purpose of changes: non lat/lon grid integration
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: Brian Eaton
List all subroutines eliminated: copies of bilin and kinds were eleminated from
the interpaerosols directory; subroutine lininterp was overloaded to
avoid several names for the same basic functionality
List all subroutines added and what they do: lininterp2d1d - interpolate from a 2d lat/lon field to a 1d ncols field - this subroutine is private in interpolate_data.F90 and accesible through the lininterp interface.
List all existing files that have been modified, and describe the changes:
A host of files were changed in the physics/cam1 directory to move from netcdf f77 interface to the f90 interface.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky:
tempest:
bangkok/lf95:
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_2_51
Originator(s): eaton
Date: Fri Feb 17 18:27:20 MST 2006
One-line Summary: configure and Makefile mods
Purpose of changes:
. Modify Makefile to enable threading with lf95 and pathf90.
*** N.B. *** lf95 (v6.2) still has severe per thread stacksize limits.
I was only able to get threading to work with the --threadheap option
which puts the thread memory on the heap. That option however doesn't
work with debug flags -g, --chk, or --chkglobal. Set up the Makefile
to allow threading only in non-debug mode.
. Modify configure so that the checks for include files and libraries are
not done when the -ccsm flag is set since the CAM Makefile is not used
in that case.
. Modify configure to require that the configure script and all
associated configuration files live in a specified subdirectory of the
source tree for the source being built. This is for robustness. Using
configure with source or configuration files from outside a source tree
that represents a fixed source version is fraught with peril.
. Add -linker option to the configure script to enable specifying a
linker which is different from the Fortran compiler. This is
particulary useful on Linux platforms where specifying mpif90 as the
linker automatically provides the correct linkage for the MPI
libraries. Note that this depends on the user's PATH being set so that
the correct mpif90 is used, i.e., the one that's consistent with the
Fortran compiler being used.
. Modify configure to run its tests in a subdirectory of the CAM build
directory. This avoids having the Makefile build the dependency list
for the full CAM just to run the tests, and consequently they run much
faster.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: described above
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated:
bld/Makefile -- renamed Makefile.in
List all subroutines added and what they do:
bld/Makefile.in -- was Makefile
List all existing files that have been modified, and describe the changes:
bld/configure
. Remove -cam_cfg option and CAM_CFGDIR environment variable: require all
configuration files to be in the same directory as the configure script.
. remove -cam_root option and CAM_ROOT environment variable: require
configure to be located in the CAM src tree.
. Modifications for CCSM build: delete setting of locations for all
external include/mod/lib directories. These are only needed for the CAM
Makefile which is not produced when doing a CCSM build.
. Remove -esmf_* options. This was used with the ESMF prototype library
which is no longer supported. Will re-implement ESMF options when we
start linking the new ESMF library.
. add -linker option
. use it to set the USER_LINKER macro in the Makefile
. Run tests in a subdirectory of the CAM build directory.
bld/Makefile.in
. add "--openmp --threadheap 4096" to FFLAGS and LDFLAGS when SMP is
defined for lf95 compiler
. add -mp to FFLAGS and LDFLAGS when SMP is defined for pathf90 compiler
. Remove default settings of macros that are set by configure.
. Remove ESMF macros except for ESMF_MOD and ESMF_LIB. Currently these are
not set by configure, but they will be when we start linking to the ESMF
library.
. add LINKER macro to be used to link cam executable.
. set LINKER to $(USER_LINKER) if it's defined, otherwise use platform
default which is $(FC) on most platforms, but is "$(CC) -f90lib" on the
SX6 and ES.
bld/mkDepends
. add fix from Tom Henderson to include Fortran syntax includes in the
dependencies (previously only recognized cpp syntax).
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: All PASS.
tempest: All PASS.
bangkok/lf95: All PASS.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
NONE
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_2_50
Originator(s): eaton
Date: Sat Feb 4 08:28:03 MST 2006
One-line Summary: Fix FV memory leak & non-adv tracers. SCAM bugfixes. Fix Cray build problem.
Purpose of changes:
1. Fix for FV memory leak found by Siddhartha Ghosh
2. Mods FV dycore from Jean-Francois Lamarque to advect only the advected
constituents
3. Mod from Rob Jacob to fix the MCT problem on the Cray (replace
"system" call with "ishell")
4. Bug fixes for scam from John Truesdale
. Corrects a C++ usage error when using the newest gcc compiler
. Allows the model to use scam generated initial conditions.
. Allows the model to use user generated data for input.
Bugs fixed (include bugzilla ID):
Described in Purpose of changes section. The memory leak in FV is
bugzilla ID 26.
Describe any changes made to build system:
Define the SYSUNICOS cpp macro when running on Cray.
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
1. FV memory leak
dynamics/fv/FVCAM_GridCompMod.F90
. move allocation of q_internal to top of run method, and deallocation to
bottom of method -- like all the rest of the allocation/deallocation
pairs.
2. Mods FV dycore
dynamics/fv/FVCAM_GridCompMod.F90
. call create_vars with ntotq as actual arg rather than nq
dynamics/fv/inital.F90
. use pcnst as actual arg corresponding to dummy arg nq in Initialize arg
list
dynamics/fv/restart_dynamics.F90
. use pcnst as actual arg corresponding to dummy arg nq in Initialize arg
list
3. Mod to fix the MCT problem on the Cray
bld/Makefile
. add -DSYSUNICOS to FPPFLAGS for UNICOS/mp build
4. Bug fixes for scam
scm_init/runtype.h
scm_init/readsaveinit.F90
ui/LoadData.ui
ui/plot.h
ui/dataset.cpp
ui/LoadDataImpl.cpp
ui/ncfile.h
Changes to external code:
The following fix was made in $SVNREPOS/mct/trunk/mpeu
m_FileResolv.F90
. use ishell rather than system command when SYSUNICOS is #define'd
A new tag for MCT was created: $SVNREPOS/mct/mct/trunk_tags/MCT2_2_0_060203
and CAM's external property for MCT was changed to that tag.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: All PASS
tempest: All PASS
bangkok/lf95: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag: cam3_2_49
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_2_49
Originator(s): Tony Craig
Date: Thu Feb 2 11:10:17 MST 2006
One-line Summary:
Update CLM source code to clm3_expa_56
Purpose of changes:
Bring in a latest version of clm3 which includes changes in
infrastructure to support finemesh integrations. This version
of clm is 100% backward compatable with namelist and datasets
and is bfb with recent version of clm3.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: new optional namelist
added to clm, fatmgrid.
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: tcraig, mvr
List all subroutines eliminated: No changes in CAM
List all subroutines added and what they do: No changes in CAM
List all existing files that have been modified, and describe the changes:
No changes in CAM
models/lnd/clm2/src/main:
initGridIndexMod.F90,lnd2atmMod.F90,CNiniTimeVar.F90,
abortutils.F90,clm_comp.F90,driver.F90,clm_camMod.F90,atmdrvMod.F90,
subgridRestMod.F90,accFldsMod.F90,clmtypeInitMod.F90,initializeMod.F90,
pftdynMod.F90,iniTimeConst.F90,histFileMod.F90,program_csm.F90,
clm_atmlnd.F90,clm_varsur.F90,clm_csmMod.F90,restFileMod.F90,
surfFileMod.F90,controlMod.F90,initSurfAlbMod.F90,initSubgridMod.F90,
clm_varctl.F90,ndepFileMod.F90,initGridCellsMod.F90,MCT_lnd_comp.F90,
program_off.F90,domainMod.F90,decompMod.F90,areaMod.F90,clmtype.F90,
histFldsMod.F90
models/lnd/clm2/src/riverroute:
RtmMod.F90
models/lnd/clm2/src/biogeochem:
CASAMod.F90,DUSTMod.F90,CNPhenologyMod.F90,STATICEcosysDynMod.F90,
DGVMMod.F90,CNrestMod.F90,VOCEmissionMod.F90,CNNDynamicsMod.F90,
CNVegStructUpdateMod.F90
models/lnd/clm2/src/biogeophys:
BalanceCheckMod.F90,SurfaceRadiationMod.F90,SoilTemperatureMod.F90,
Biogeophysics1Mod.F90,Biogeophysics2Mod.F90,FrictionVelocityMod.F90,
Hydrology1Mod.F90,Hydrology2Mod.F90,BiogeophysicsLakeMod.F90,
HydrologyLakeMod.F90,BareGroundFluxesMod.F90,CanopyFluxesMod.F90
SurfaceAlbedoMod.F90
Code changes: *** Only in clm code ***
Merge atm2lnd_state_type, atm2lnd_flux_type. Same for lnd2atm state/flux.
Related changes in clm3 and elsewhere in code.
Add domainMod.F90 and domain_type. Migrate grid data into domain type.
Instantiate adomain(atm/coarse), ldomain(lnd/finemesh), rdomain(rtm),
ddomain(atmdrv external data) in model.
Add lats, latn, lonw, lone 2d arrays and associated code changes.
Cleanup areaMod.F90; merging subroutines, removing redundant code, eliminate
*_point routines.
Remove numlon
Add decomp_type for gcelldc and gcellsn. Remove redundant data in other
arrays related to addressing physical space and logical space.
Clean up interface in set_landunit subroutines. Remove redundant code.
Clean up procs and clumps datatypes, removing redundant data.
Migrate clm3 topology data to pointers from copies
Add gridmap_type for interpolation and associated code and routines to
support the type.
Add clm_atmlnd.F90 file for upscale/downscale code. Add clm_mapa2l
and clm_mapl2a to carry out mapping associated with upscale/downscale.
Add gridmap_setmapsFM for generation of weights for downscale/upscale
routines.
forc_ndep should not be in atm2lnd_type.
Reuse new code as much as possible throughout.
Rename latixy and longxy to latc and lonc.
Remove fullgrid attribute.
Add pftm to domain datatype and history file
Modify coupling to handle coarse <-> finemesh for standlaone, cam, and
ccsm.
Add normalized area to history files
Add lat_a, lon_a, latdeg_a, londeg_a to clm3 datatype for atm lats/lons.
required in SurfaceAlbedo computation where the the zenith angle has
to be based on the atm (coarse) grid, not the fine clm grid.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: All PASS
tempest: All PASS
bangkok/lf95: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag: cam3_2_48
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
Answers are bit-for-bit.
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_2_48
Originator(s): Francis Vitt
Date:
One-line Summary:
Update to the tropospheric chemistry code and fixes to the offline driver.
Purpose of changes:
The updates to the tropospheric chemistry code where done to include
into CAM the latest MOZART chemistry and to provide the option to
to use interactive dry deposition and interactive photolysis rates
rather than the prescribed table look up methods.
Changes to the offline dynamics driver were done to fix a total energy
conservation issue that was evident with simulations longer than two
months. A fix to a bug which occurred in high resolution runs is also
included.
Bugs fixed (include bugzilla ID):
Describe any changes made to build system: none
Describe any changes made to the namelist:
added:
character(len=256) :: depvel_file
character(len=256) :: depvel_lnd_file
character(len=256) :: clim_soilw_file
character(len=256) :: season_wes_file
character(len=256) :: trop_pause_file
character(len=256) :: lght_landmask_file
real(r8) :: lght_no_prd_factor
logical :: xactive_prates
logical :: xactive_drydep
character(len=256) :: photopath
character(len=8) :: emis_type
integer :: emis_date
integer :: emis_yr_offset
character(len=256) :: met_filenames_list
removed:
chem_config
List any changes to the defaults for the boundary datasets:
Describe any substantial timing or memory changes:
Code reviewed by: Brian Eaton
List all subroutines eliminated:
models/atm/cam/src/chemistry/trop_mozart/mo_waccm_photo.F90
models/atm/cam/src/chemistry/trop_mozart/mo_read_chm_sim.F90
models/atm/cam/src/chemistry/trop_mozart/phtadj.F90
models/atm/cam/src/chemistry/trop_mozart/rxtmod.F90
List all subroutines added and what they do:
models/atm/cam/src/chemistry/trop_mozart/mo_wavelen.F90
models/atm/cam/src/chemistry/trop_mozart/mo_cnum.F90
models/atm/cam/src/chemistry/trop_mozart/mak_grp_vmr.F90
models/atm/cam/src/chemistry/trop_mozart/mo_phtadj.F90
models/atm/cam/src/chemistry/trop_mozart/mo_airmas.F90
models/atm/cam/src/chemistry/trop_mozart/mo_photoin.F90
models/atm/cam/src/chemistry/trop_mozart/mo_setaer.F90
models/atm/cam/src/chemistry/trop_mozart/mo_trislv.F90
models/atm/cam/src/chemistry/trop_mozart/mo_setair.F90
models/atm/cam/src/chemistry/trop_mozart/mo_waveall.F90
models/atm/cam/src/chemistry/trop_mozart/mo_seto2.F90
models/atm/cam/src/chemistry/trop_mozart/mo_tropopause.F90
models/atm/cam/src/chemistry/trop_mozart/mo_schu.F90
models/atm/cam/src/chemistry/trop_mozart/mo_rtlink.F90
models/atm/cam/src/chemistry/trop_mozart/mo_addpnt.F90
models/atm/cam/src/chemistry/trop_mozart/mo_constants.F90
models/atm/cam/src/chemistry/trop_mozart/mo_sim_dat.F90
models/atm/cam/src/chemistry/trop_mozart/mo_lymana.F90
models/atm/cam/src/chemistry/trop_mozart/time_utils.F90
models/atm/cam/src/chemistry/trop_mozart/mo_setz.F90
models/atm/cam/src/chemistry/trop_mozart/mo_calcoe.F90
models/atm/cam/src/chemistry/trop_mozart/mo_zadj.F90
models/atm/cam/src/chemistry/trop_mozart/mo_pchem.F90
models/atm/cam/src/chemistry/trop_mozart/mo_ps2str.F90
models/atm/cam/src/chemistry/trop_mozart/mo_sphers.F90
models/atm/cam/src/chemistry/trop_mozart/mo_setozo.F90
models/atm/cam/src/chemistry/trop_mozart/m_types.F90
models/atm/cam/src/chemistry/trop_mozart/mo_photo.F90
models/atm/cam/src/chemistry/trop_mozart/mo_tuv_inti.F90
models/atm/cam/src/chemistry/trop_mozart/mo_strato_sad.F90
models/atm/cam/src/chemistry/trop_mozart/mo_waveo3.F90
models/atm/cam/src/chemistry/trop_mozart/mo_rxt_mod.F90
models/atm/cam/src/chemistry/trop_mozart/mo_setcld.F90
models/atm/cam/src/chemistry/trop_mozart/mo_xsections.F90
models/atm/cam/src/chemistry/trop_mozart/mo_drydep_tables.F90
models/atm/cam/src/chemistry/trop_mozart/mo_inter.F90
models/atm/cam/src/chemistry/trop_mozart/mo_wavelab.F90
models/atm/cam/src/chemistry/trop_mozart/mo_params.F90
List all existing files that have been modified, and describe the changes:
models/atm/cam/bld/config_trop_chem_mozart_defaults.xml
the number of advected constituents has changed to 99 for the tropospheric chemistry
models/atm/cam/bld/camexp.pm
models/atm/cam/bld/DefaultCAMEXPNamelist.xml
made changes to be consistent with the changes to the namelist variables
models/atm/cam/src/control/readinitial.F90
moved from stepon the adjustment to the top interface pressure when zero
models/atm/cam/src/control/runtime_opts.F90
see namelist changes above
models/atm/cam/src/control/wrap_nf.F90
added subroutines wrap_inq_varndims and wrap_get_vara_text
models/atm/cam/src/control/history.F90
made module variable fieldname_len public and increase its value to 16
models/atm/cam/src/utils/string_utils.F90
made GLC function public
models/atm/cam/src/physics/cam1/comsrf.F90
added fsds field
models/atm/cam/src/physics/cam1/diagnostics.F90
increased the length of string variable dcconnam
models/atm/cam/src/physics/cam1/tphysbc.F90
changed the intent on fsds argument variable to inout
models/atm/cam/src/physics/cam1/check_energy.F90
don't adjust the total global energy in offline mode
models/atm/cam/src/physics/cam1/restart_physics.F90
models/atm/cam/src/physics/cam1/constituents.F90
increased the length of string variable names
models/atm/cam/src/physics/cam1/physpkg.F90
removed module variable fsds
models/atm/cam/src/physics/cam1/chemistry.F90
made the chemistry interface consistent with trop_mozart
models/atm/cam/src/physics/cam1/tphysac.F90
the arguments to subroutine chem_timestep_tend has changed
models/atm/cam/src/physics/cam1/vertical_diffusion.F90
increased the length of vdiffnam string variable
models/atm/cam/src/physics/waccm/chemistry.F90
made the chemistry interface consistent with trop_mozart
models/atm/cam/src/chemistry/trop_mozart/mo_lu_solve.F90
models/atm/cam/src/chemistry/trop_mozart/mo_mean_mass.F90
models/atm/cam/src/chemistry/trop_mozart/mo_jlong.F90
models/atm/cam/src/chemistry/trop_mozart/mo_nln_matrix.F90
models/atm/cam/src/chemistry/trop_mozart/mo_gas_phase_chemdr.F90
models/atm/cam/src/chemistry/trop_mozart/mo_lu_factor.F90
models/atm/cam/src/chemistry/trop_mozart/mo_lightning.F90
models/atm/cam/src/chemistry/trop_mozart/mo_ub_vals.F90
models/atm/cam/src/chemistry/trop_mozart/mo_chemini.F90
models/atm/cam/src/chemistry/trop_mozart/mo_indprd.F90
models/atm/cam/src/chemistry/trop_mozart/mo_aerosols.F90
models/atm/cam/src/chemistry/trop_mozart/mo_adjrxt.F90
models/atm/cam/src/chemistry/trop_mozart/mo_imp_sol.F90
models/atm/cam/src/chemistry/trop_mozart/mo_lin_matrix.F90
models/atm/cam/src/chemistry/trop_mozart/m_rxt_id.F90
models/atm/cam/src/chemistry/trop_mozart/chemistry.F90
models/atm/cam/src/chemistry/trop_mozart/mo_drydep.F90
models/atm/cam/src/chemistry/trop_mozart/mo_wetdep.F90
models/atm/cam/src/chemistry/trop_mozart/mo_setrxt.F90
models/atm/cam/src/chemistry/trop_mozart/mo_seasalt.F90
models/atm/cam/src/chemistry/trop_mozart/chem_mods.F90
models/atm/cam/src/chemistry/trop_mozart/mo_prod_loss.F90
models/atm/cam/src/chemistry/trop_mozart/mo_usrrxt.F90
models/atm/cam/src/chemistry/trop_mozart/m_het_id.F90
models/atm/cam/src/chemistry/trop_mozart/m_spc_id.F90
models/atm/cam/src/chemistry/trop_mozart/mo_srf_emissions.F90
update the tropospheric chemistry code to be consistent with MOZART 4
models/atm/cam/src/chemistry/waccm_mozart/chemistry.F90
made the chemistry interface consistent with trop_mozart
models/atm/cam/src/dynamics/fv/pfixer.F90
changed variable names and cleaned up code
models/atm/cam/src/dynamics/fv/spmd_dyn.F90
fixed a bug in compute_gsfactors which occurs when 2D decomposition is used
models/atm/cam/src/dynamics/fv/stepon.F90
removed the adjustment to the top interface pressure if zero
models/atm/cam/src/dynamics/fv/metdata.F90
Made changes to get the observed surface pressure at sub-timestep
interval times. This is needed for the high resolution bug fix.
models/atm/cam/src/dynamics/fv/FVCAM_GridCompMod.F90
Made changes needed to fix the bug in the offline driver used in
high resolution simulations.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky:
037 bl354 TBL.sh fm2dh outfrq3s 9 .................................FAIL! rc= 7
040 bl355 TBL.sh fmo2dh off2x2.5 9 ................................FAIL! rc= 7
Changes to the trop_mozart chemistry have caused the base line comparison
to fail as expected.
tempest: All PASS
bangkok/lf95: All PASS
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_2_47
Originator(s): eaton, mvr
Date: Thu Jan 19 12:01:57 MST 2006
One-line Summary:
Workaround for xlf90 (v9.1.0.3) bug (on bluevista) & test script updates.
Purpose of changes:
The workaround for the compiler bug allows running on bluevista. Note
that there is still an unresolved bug on the platform affecting the
trop_mozart chemistry mode.
The TBR.sh test script has been modified to work around a GPFS
problem (sometimes files that have just been copied don't appear in the
output from an "ls" command executed in the directory that the files were
just copied to) which causes the branch test to fail intermittently on bluesky.
Misc changes in the overnight test scripts.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: eaton, mvr
List all subroutines eliminated:
models/atm/cam/test/system/update_collections.sh
List all subroutines added and what they do:
models/atm/cam/test/system/tag_email.sh
List all existing files that have been modified, and describe the changes:
models/atm/cam/src/physics/cam1/physpkg.F90
. in phys_init pass phys_state(begchunk:endchunk) to chem_init and
ozone_data_init
models/atm/cam/test/system/TBR.sh
models/atm/cam/test/system/td_nightly_aix.sh
models/atm/cam/test/system/td_nightly_linux.sh
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: All PASS.
tempest: All PASS.
bangkok/lf95: All PASS.
CAM tag used for the baseline comparison tests if different than previous
tag: cam3_2_46
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
Answers are bit-for-bit.
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_2_46
Originator(s): eaton
Date: Mon Jan 9 16:58:42 MST 2006
One-line Summary: refactor restart logic and fix restart bug
Purpose of changes:
Provide control of when restart files are written that is decoupled from
the behavior of CAM's history file writes.
Also did some cleanup work:
. get rid of anncyc, nlhst, itsst
. move nlres and lbrnch into restart module -- get it out of comctl.h
. call set_restart_filepath from restart_setopts -- remove from runtime_opts
. move the restart filepaths into the restart module -- get them out of filepaths.F90
Bugs fixed (include bugzilla ID):
At cam3_2_24 the restart writes for fields needed by CAM's ocean and sea
ice modules were moved into the run methods for those modules. This broke
the writing of restart files in all situations except at the end of the
model run when the restart triggers off the nlend flag.
The problem is that the calls to write the restart files for the ocean and
ice models are made before atm_run4 which makes the calls to write CAM's
history file. Since the old logic to determine when it's time to write a
restart file depended on the state in CAM's history module, and that state
was being updated after the calls to write the ocean and ice restart files,
the ocn and ice restarts were out of sync with CAM's restarts except when
nlend=.true. The fix was to implement a separate logic for writing restart
files that doesn't depend on the state of CAM's history module.
Describe any changes made to build system: none
Describe any changes made to the namelist:
The namelist variables nrefrq and itsst have been eliminated.
New namelist variables added:
logical :: no_restart ! true => turn off all restart writes (default: false)
integer :: restart_nsteps ! restart interval in steps
integer :: restart_nhours ! restart interval in hours
integer :: restart_ndays ! restart interval in days
logical :: restart_monthly ! true => write restart at begining of each month (default: true)
logical :: restart_yearly ! true => write restart at begining of each year
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: mvertens reviewed CLM code changes
List all subroutines eliminated:
control/rstwr.F90
. remove this file -- subroutine added to restart module
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
control/cam.F90
. change rstwr from external to local variable. add init before timestep
loop.
. add rstwr as actual arg to ocn_run, ice_run, atm_run4
. add rstwr actual arg to clm_camRun (lnd_run)
control/cam_comp.F90
. add rstwr dummy arg to cam_run4 and remove the external declaration
. change write_restart call logic to only use rstwr
. set rstwr using restart_is_write_step method
control/ccsm_msg.F90
. add rstwr intent(out) arg to ccsm_run and remove the external declaration
. move the logic to set rstwr from the old rstwr function into ccsm_run
. change write_restart_ccsm call logic to only use rstwr
control/comctl.h
. remove nrefrq
. remove anncyc -- no longer used
. remove nlhst -- it is equivalent to lbrnch
. remove itsst -- obsolete and broken
. move nlres and lbrnch to restart.F90
control/filenames.F90
. remove comctl.h
. remove nrefrq dependence on initializing rest_pfile
. move nrevsn and rest_pfile to restart module
control/history.F90
. add rstwr dummy arg to wrapup and remove the external declaration
control/restart.F90
. extend functionality of this module to include the control of setting the
restart write time.
. add restart_is_write_step method. This started as the rstwr
function in rstwr.F90.
. add restart_init to initialize the restart write time
. add restart_{default,set}opts methods for namelist
. restart_setopts provides checking of the namelist settings and echos a
summary to the log file.
. add next_write_{ymd,tod} to restart file
. replace use of nlhst with lbrnch
. add nlres and lbrnch control variables (moved here from comctl.h)
. add nsrest as intent(in) arg of the restart_setopts method - it's needed
to set the values of nlres and lbrnch
. remove comctl.h -- no longer needed
control/runtime_opts.F90
. remove nrefrq
. add restart write interval namelist variables
. add restart_{default,set}opts calls
. add mpibcast calls
. replace use of nlhst with lbrnch
. add nlres and lbrnch control variables (moved here from comctl.h)
. add nsrest as intent(in) arg of the restart_setopts method - it's needed
to set the values of nlres and lbrnch
. remove comctl.h -- no longer needed
control/startup_initialconds.F90
. add restart_init call to setup_initial method
control/time_manager.F90
. cleanup some comments
. remove target attribute from tm_cal
. remove public attribute from tm_clock
. add new methods:
timemgr_time_ge ! check if time2 is later than or equal to time1
timemgr_time_inc ! increment time instant by a given time interval
dynamics/eul/initcom.F90
. use nsrest instead of nlres in conditional
ice/csim4/camice.F90
. remove comctl.h from camice_write_restart -- not used
. add rstwr dummy arg to camice_run and remove the external declaration
. change camice_write_restart call logic to only use rstwr
. remove unused ref to nrevsn
ocnsice/dom/camoce.F90
. add rstwr dummy arg to camoce_run and remove the external declaration
. change camoce_write_restart call logic to only use rstwr
. remove anncyc -- no longer used
. remove itsst -- obsolete and broken
. remove unused ref to nrevsn
ocnsice/som/somoce.F90
. add rstwr dummy arg to somoce_run and remove the external declaration
. change somoce_write_restart call logic to only use rstwr
. remove unused ref to nrevsn
test/system/nl_files/pghgsul
test/system/nl_files/ttrac
. add restart_nsteps setting to match history write frequency (these are
the only namelists used in the TBR tests)
tools/scam/scm_init/scam_run.F90
. add rstwr actual arg to ocn_run and ice_run calls. Hardwire value to
.false. (scam doesn't ask cam to write a restart file)
Mods were made to the following CLM routines. These mods are accessed
via the external code svn:/clm2/branch_tags/clm3_expa_53_brnchT_cam01
clm/src/main/clm_camMod.F90
. add rstwr intent(in) arg to clm_camRun and use it as actual arg in call to
clm_run2
. remove setting of cam_nhtfrq, cam_mfilt to nhtfrq, mfilt.
clm/src/main/clm_comp.F90
. add rstwr intent(in) arg to clm_run2 and use it as actual arg in call to
driver2.
clm/src/main/clm_varctl.F90
. remove cam_nhtfrq and cam_mfilt -- no longer used
clm/src/main/controlMod.F90
. remove overwritting of hist_nhtfrq, hist_mfilt with cam_nhtfrq, cam_mfilt
clm/src/main/driver.F90
. add rstwr optional intent(in) arg to driver2 and use it in conditional for call to
restFile_write.
. check for presence of rstwr arg and use it to set local logical variable,
or use the do_restwrite method if it's not present. This allows driver2
to continue to be used by the program_csm and program_off drivers.
*** NOTE *** The following files have been moved from cam/src/control/
to cam/src/utils/. (The reason is that by moving utility functions into
a separate directory it's easier to build unit testers using CAM's build
procedure which requires that all source files in each specified
directory be built.)
abortutils.F90
cfort.h
spmd_utils.F90
string_utils.F90
time_manager.F90
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky: All PASS.
tempest: All PASS.
bangkok/lf95: All PASS.
CAM tag used for the baseline comparison tests if different than previous
tag: cam3_2_45
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
Answers are bit-for-bit.
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_2_45
Originator(s): eaton
Date: Sun Dec 18 14:55:02 MST 2005
One-line Summary: Bug fixes for memory leak and misc.
Purpose of changes: bug fixes
Bugs fixed (include bugzilla ID):
1. Fix memory leak introduced in cam3_0_20. Briefly the problem is that the
overloaded .not. operator defined in diffusion_solver.F90 returns an object
of type vdiff_selector for which memory is allocated, but this operator is
used as an actual argument in the call to compute_vdiff in
vertical_diffusion.F90, and so the memory is never assigned to a variable
which can be deallocated.
2. The T field in initial files for FV when 2D decomp is used are bad.
This was caused by a problem with an array index in
fv/diag_dynvar_ic.F90
3. The history restart files were being removed at the end of the run. The
normal CAM behavior is to not remove any files written at the end of the
run.
4. The filepath for tgcm data was being truncated at 80 characters.
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes:
Memory leak fixed. The severity of this problem depends on what the
compiler did with an object returned by a function that was being used as
an actual argument. The problem was most severe with the lf95 compiler.
Code reviewed by: myself
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
Bug 1.
physics/cam1/vertical_diffusion.F90
. replace fieldlist by fieldlist_moist
. add fieldlist_dry to module data and replace .not.fieldlist w/ fieldlist_dry
. add fieldlist_dry as actual arg to init_vdiff
. use fieldlist_dry in 2nd call to compute_vdiff
physics/cam1/diffusion_solver.F90
. add fieldlist_dry as dummy arg to init_vdiff and initialize it
Bug 2.
dynamics/fv/diag_dynvar_ic.F90
. The outfld call for T&IC needs the array referenced as t3(1,kfirst,j),
not t3(1,1,j).
Bug 3.
control/history.F90
. Change the argument in the putfil call that determines whether the file
is removed after it is archived from .true. to .not.nlend.
. Add nlend as a dummy arg to write_restart_history
control/restart.F90
. add nlend to the actual args of write_restart_history
Bug 4.
physics/waccm/tgcm_forcing.F90
. Change "character*80 locfn" to "character*256 locfn" at line 158
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky:
All PASS except:
004 br131 TBR.sh e32c11dh pghgsul 6+3 .............................FAIL! rc= 10
018 br331 TBR.sh f4c11dh pghgsul 6+3 ..............................FAIL! rc= 10
043 br531 TBR.sh s32c11dh pghgsul 6+3 .............................FAIL! rc= 9
tempest:
All PASS except:
003 br131 TBR.sh e32c11dh pghgsul 6+3 .............................FAIL! rc= 9
020 br331 TBR.sh f4c11dh pghgsul 6+3 ..............................FAIL! rc= 9
034 br531 TBR.sh s32c11dh pghgsul 6+3 .............................FAIL! rc= 9
bangkok/lf95:
All PASS except:
003 br111 TBR.sh e8c8mdm ttrac 6+3 ................................FAIL! rc= 9
015 br311 TBR.sh f10c8mdm ttrac 6+3 ...............................FAIL! rc= 9
031 br511 TBR.sh s8c8mdm ttrac 6+3 ................................FAIL! rc= 9
These tests are all expected to fail due to a bug in the current logic for
writing restart files.
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate):
If bitwise differences were observed, how did you show they were no worse
than roundoff? Answers are BFB.
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_2_44
Originator(s): Mariana Vertenstein, Rob Jacob
Date: 2005-12-16
One-line Summary: Incorprated MCT for cam-clm coupling
Purpose of changes: Replaced current cam-clm lp-coupling scheme with
new MCT coupling and interfaces.
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: MCT is now built as part of the
CAM executable (since it is all F90 code) - configure was changed so
that the MCT filepath could be incorporated into the CAM build
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none (from a preliminary
analysis of timing numbers). A more detailed timing analysis is planned.
Code reviewed by: Mariana Vertenstein, Rob Jacob
List all subroutines eliminated: none
List all subroutines added and what they do:
control/seq_fields_mod.F90
provides list of fields to be coupled using same mechanism as in
concurrent ccsm mode
control/MCT_seq.F90
control/MCT_atm_comp.F90
provide new MCT coupling interfaces
List all existing files that have been modified, and describe the changes:
physics/cam1/camhub_comp.F90
cleaned up interfaces
physics/cam1/prescribed_aerosols.F90
bug fix (provided by Andrew Conley) - does not change answers
in current tests
atm/cam/tools/scam/configure
atm/cam/tools/scam/scm_init/init_model.F90
atm/cam/tools/scam/scm_init/scam_run.F90
changes to SCAM in order to run with MCT cam-clm coupling
and also use new camhub_comp changes
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky:
004 br131 TBR.sh e32c11dh pghgsul 6+3 .............................FAIL! rc= 9
018 br331 TBR.sh f4c11dh pghgsul 6+3 ..............................FAIL! rc= 9
043 br531 TBR.sh s32c11dh pghgsul 6+3 .............................FAIL! rc= 10
tempest:
003 br131 TBR.sh e32c11dh pghgsul 6+3 .............................FAIL! rc= 9
020 br331 TBR.sh f4c11dh pghgsul 6+3 ..............................FAIL! rc= 9
034 br531 TBR.sh s32c11dh pghgsul 6+3 .............................FAIL! rc= 9
bangkok/lf95:
003 br111 TBR.sh e8c8mdm ttrac 6+3 ................................FAIL! rc= 9
015 br311 TBR.sh f10c8mdm ttrac 6+3 ...............................FAIL! rc= 9
031 br511 TBR.sh s8c8mdm ttrac 6+3 ................................FAIL! rc= 9
CAM tag used for the baseline comparison tests if different than previous
tag: previous tag
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate): results were bit-for-bit
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
Tag name: cam3_2_43
Originator(s): Mariana Vertenstein, Mat Rothstein
Date: 2005-12-15
One-line Summary: Updated clm code to clm3_expa_52
Purpose of changes:
Update clm code base to clm3_expa_52
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: Mariana Vertenstein, Keith Oleson
List all subroutines eliminated: none
List all subroutines added and what they do: none
List all existing files that have been modified, and describe the changes:
src/biogeochem/CASAMod.F90
src/biogeochem/DGVMEstablishmentMod.F90
src/biogeophys/CanopyFluxesMod.F90
src/main/clm_varcon.F90
src/main/time_manager.F90
- Put in changes to use esmf_wrf_timemgr code base
- Changes to clm_varcon.F90 and CanopyFluxesMod.F90 are
Changes made to the clm hydrology in tag clm3_expa_46 that were
increased the conductance of heat and moisture between the soil and the
canopy air space. This leads to the following changes in land climatology: global
increase in soil evaporation and a decrease in transpiration
and in the soil surface temperature.
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky:
004 br131 TBR.sh e32c11dh pghgsul 6+3 .............................FAIL! rc= 9
005 bl131 TBL.sh e32c11dh pghgsul 9 ...............................FAIL! rc= 7
008 bl132 TBL.sh e32sdh ghgrmp 9 ..................................FAIL! rc= 7
015 bl151 TBL.sh e64h outfrq24h -2 ................................FAIL! rc= 7
018 br331 TBR.sh f4c11dh pghgsul 6+3 ..............................FAIL! rc= 10
019 bl331 TBL.sh f4c11dh pghgsul 9 ................................FAIL! rc= 7
022 bl332 TBL.sh f4sdh ghgrmp 9 ...................................FAIL! rc= 7
030 bl336 TBL.sh wm4h outfrq3s 9 ..................................FAIL! rc= 7
034 bl351 TBL.sh f2h outfrq24h -2 .................................FAIL! rc= 7
037 bl354 TBL.sh fm2dh outfrq3s 9 .................................FAIL! rc= 7
040 bl355 TBL.sh fmo2dh off2x2.5 9 ................................FAIL! rc= 7
044 bl531 TBL.sh s32c11dh pghgsul 9 ...............................FAIL! rc= 7
047 bl532 TBL.sh s32sdh ghgrmp 9 ..................................FAIL! rc= 7
054 bl551 TBL.sh s64h outfrq24h -2 ................................FAIL! rc= 7
tempest:
003 br131 TBR.sh e32c11dh pghgsul 6+3 .............................FAIL! rc= 9
004 bl131 TBL.sh e32c11dh pghgsul 9 ...............................FAIL! rc= 7
007 bl132 TBL.sh e32sdh ghgrmp 9 ..................................FAIL! rc= 7
014 bl152 TBL.sh e64o outfrq24h -2 ................................FAIL! rc= 7
017 bl314 TBL.sh wg10dh outfrq3s 9 ................................FAIL! rc= 7
020 br331 TBR.sh f4c11dh pghgsul 6+3 ..............................FAIL! rc= 9
021 bl331 TBL.sh f4c11dh pghgsul 9 ................................FAIL! rc= 7
024 bl332 TBL.sh f4sdh ghgrmp 9 ...................................FAIL! rc= 7
031 bl352 TBL.sh f2o outfrq24h -2 .................................FAIL! rc= 7
034 br531 TBR.sh s32c11dh pghgsul 6+3 .............................FAIL! rc= 9
035 bl531 TBL.sh s32c11dh pghgsul 9 ...............................FAIL! rc= 7
038 bl532 TBL.sh s32sdh ghgrmp 9 ..................................FAIL! rc= 7
045 bl552 TBL.sh s64o outfrq24h -2 ................................FAIL! rc= 7
bangkok/lf95:
003 br111 TBR.sh e8c8mdm ttrac 6+3 ................................FAIL! rc= 9
004 bl111 TBL.sh e8c8mdm ttrac 9 ..................................FAIL! rc= 7
008 bl112 TBL.sh e8sdm ghgrmp 9 ...................................FAIL! rc= 7
012 bl153 TBL.sh e64m outfrq24h -2 ................................FAIL! rc= 7
015 br311 TBR.sh f10c8mdm ttrac 6+3 ...............................FAIL! rc= 9
016 bl311 TBL.sh f10c8mdm ttrac 9 .................................FAIL! rc= 7
024 bl312 TBL.sh f10sdm ghgrmp 9 ..................................FAIL! rc= 7
028 bl353 TBL.sh f2m outfrq24h -2 .................................FAIL! rc= 7
031 br511 TBR.sh s8c8mdm ttrac 6+3 ................................FAIL! rc= 9
032 bl511 TBL.sh s8c8mdm ttrac 9 ..................................FAIL! rc= 7
036 bl512 TBL.sh s8sdm ghgrmp 9 ...................................FAIL! rc= 7
040 bl553 TBL.sh s64m outfrq24h -2 ................................FAIL! rc= 7
CAM tag used for the baseline comparison tests if different than previous
tag:
Summarize any changes to answers: The above changes to clm lead to a new climate
If bitwise differences were observed, how did you show they were no worse
than roundoff? NA
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced:
- source tag (all code used must be in the repository):
cam3_2_40 (with clm source code corresponding to clm3_expa_48)
- platform/compilers:
bluesky
- configure commandline:
configure -ocn dom -dyn fv -spmd -smp -test -res 4x5
- build-namelist command (or complete namelist):
&camexp
absems_data = '/fis/cgd/cseg/csm/inputdata/atm/cam/rad/abs_ems_factors_fastvx.c030508.nc'
aeroptics = '/fis/cgd/cseg/csm/inputdata/atm/cam/rad/AerosolOptics_c050419.nc'
bnd_topo = '/fis/cgd/cseg/csm/inputdata/atm/cam/topo/USGS-gtopo30_4x5_remap_c050520.nc'
bndtvaer = '/fis/cgd/cseg/csm/inputdata/atm/cam/rad/AerosolMass_V_4x5_clim_c031022.nc'
bndtvo = '/fis/cgd/cseg/csm/inputdata/atm/cam/ozone/pcmdio3.r8.64x1_L60_clim_c970515.nc'
bndtvs = '/fis/cgd/cseg/csm/inputdata/atm/cam/sst/sst_HadOIBl_bc_4x5_clim_c030228.nc'
caseid = 'cam3_2_40_fv4x5dom'
dtime = 1800
iyear_ad = 1950
ncdata = '/fis/cgd/cseg/csm/inputdata/atm/cam/inic/fv/cami_0000-09-01_4x5_L26_c031217.nc'
nelapse = -1
nsrest = 0
/
&clmexp
fpftcon = '/fis/cgd/cseg/csm/inputdata/lnd/clm2/pftdata/pft-physiology-cn16.c040719'
fsurdat = '/fis/cgd/cseg/csm/inputdata/lnd/clm2/srfdata/cam/clms_3.1_4x5_c050523.nc'
/
- MSS location of output:
/OLESON/csm/cam3_2_40_fv4x5dom
MSS location of control simulations used to validate new climate
/OLESON/csm/cam3_2_29_fv4x5dom
URL for AMWG diagnostics output used to validate new climate:
http://www.cgd.ucar.edu/tss/clm/diagnostics/cam_clm2/cam3_2_40_fv4x5dom-cam3229_fv4x5dom/sets.htm
===============================================================
===============================================================
Tag name: cam3_2_42
Originator(s): mvr
Date: 051213
One-line Summary: csm_share, esmf_wrf_timemgr, timing now from
external sources; updated to share3_051205; new ChangeLog format
Purpose of changes: taking advantage of some subversion features
Bugs fixed (include bugzilla ID): none
Describe any changes made to build system: none
Describe any changes made to the namelist: none
List any changes to the defaults for the boundary datasets: none
Describe any substantial timing or memory changes: none
Code reviewed by: mvr, eaton
List all subroutines eliminated: all csm_share, esmf_wrf_timemgr, and
timing files;
D models/utils/timing
D models/utils/timing/private.h
D models/utils/timing/gptl_papi.c
D models/utils/timing/rtcrate.F90
D models/utils/timing/gptlutil.c
D models/utils/timing/threadutil.c
D models/utils/timing/gptl.c
D models/utils/timing/f_wrappers.c
D models/utils/timing/gptl.h
D models/utils/timing/README
D models/utils/timing/gptl.inc
D models/utils/esmf_wrf_timemgr
D models/utils/esmf_wrf_timemgr/ESMF_AlarmClockMod.F90
D models/utils/esmf_wrf_timemgr/ESMF_CalendarMod.F90
D models/utils/esmf_wrf_timemgr/ESMF_BaseMod.F90
D models/utils/esmf_wrf_timemgr/ESMF_TimeMgr.inc
D models/utils/esmf_wrf_timemgr/ESMF_Mod.F90
D models/utils/esmf_wrf_timemgr/c_esmc_subroutines.F90
D models/utils/esmf_wrf_timemgr/ESMF_BaseTimeMod.F90
D models/utils/esmf_wrf_timemgr/ESMF_ClockMod.F90
D models/utils/esmf_wrf_timemgr/ESMF_AlarmMod.F90
D models/utils/esmf_wrf_timemgr/ESMF_TimeMod.F90
D models/utils/esmf_wrf_timemgr/ESMF_Macros.inc
D models/utils/esmf_wrf_timemgr/ESMF_Stubs.F90
D models/utils/esmf_wrf_timemgr/ESMF_TimeIntervalMod.F90
D models/utils/esmf_wrf_timemgr/wrf_error_fatal.F90
D models/utils/esmf_wrf_timemgr/ESMF_FractionMod.F90
D models/utils/esmf_wrf_timemgr/Makefile
D models/csm_share
D models/csm_share/ChangeLog
D models/csm_share/unit_testers
D models/csm_share/unit_testers/test_shr_sys.F90
D models/csm_share/unit_testers/test_shr_orb.F90
D models/csm_share/unit_testers/Makefile
D models/csm_share/ChangeSum
D models/csm_share/shr
D models/csm_share/shr/shr_date_mod.F90
D models/csm_share/shr/shr_map_mod.F90
D models/csm_share/shr/shr_sys_mod.F90
D models/csm_share/shr/shr_file_mod.F90
D models/csm_share/shr/shr_vmath_mod.F90
D models/csm_share/shr/shr_timer_mod.F90
D models/csm_share/shr/shr_orb_mod.F90
D models/csm_share/shr/shr_kind_mod.F90
D models/csm_share/shr/shr_tInterp_mod.F90
D models/csm_share/shr/shr_mpi_mod.F90
D models/csm_share/shr/shr_const_mod.F90
D models/csm_share/shr/shr_msg_mod.F90
D models/csm_share/shr/shr_stream_mod.F90
D models/csm_share/shr/shr_ncread_mod.F90
D models/csm_share/shr/shr_alarm_mod.F90
D models/csm_share/shr/shr_cal_mod.F90
D models/csm_share/shr/shr_vmath_fwrap.c
D models/csm_share/shr/shr_string_mod.F90
D models/csm_share/README
D models/csm_share/cpl
D models/csm_share/cpl/cpl_kind_mod.F90
D models/csm_share/cpl/cpl_const_mod.F90
D models/csm_share/cpl/cpl_infobuf_mod.F90
D models/csm_share/cpl/cpl_fields_mod.F90.CO2A
D models/csm_share/cpl/cpl_fields_mod.F90.CO2B
D models/csm_share/cpl/cpl_comm_mod.F90
D models/csm_share/cpl/cpl_fields_mod.F90.CO2_DMSA
D models/csm_share/cpl/cpl_iobin_mod.F90
D models/csm_share/cpl/cpl_interface_mod.F90
D models/csm_share/cpl/cpl_fields_mod.F90
D models/csm_share/cpl/cpl_domain_mod.F90
D models/csm_share/cpl/cpl_bundle_mod.F90
D models/csm_share/cpl/cpl_contract_mod.F90
D models/csm_share/cpl/cpl_map_mod.F90
D models/csm_share/cpl/cpl_control_mod.F90
D models/csm_share/cpl/cpl_mct_mod.F90
D models/csm_share/cpl/cpl_iocdf_mod.F90
D models/csm_share/README.scm
- all these replaced with subversion external directories
D models/atm/cam/doc/BaseModelSum
D models/atm/cam/doc/BaseModelLog
- some documentation cleanup
List all subroutines added and what they do:
A models/utils/SVN_EXTERNAL_DIRECTORIES
A models/SVN_EXTERNAL_DIRECTORIES
- contain definitions for the external directories
A + models/atm/cam/doc/Copyright
- copy of the copyright information that'll be picked up by ccsm external pull
A models/atm/cam/doc/ChangeLog_template
- template for new ChangeLog format
List all existing files that have been modified, and describe the changes:
M models
M models/utils
- directories modified with properties for the external definitions
M models/atm/cam/doc/ChangeLog
- added documentation for this tag
If there were any failures reported from running test_driver.sh on any test
platform, and checkin with these failures has been OK'd by the gatekeeper,
then copy the lines from the td.*.status files for the failed tests to the
appropriate machine below. All failed tests must be justified.
bluesky:
004 br131 TBR.sh e32c11dh pghgsul 6+3 .............................FAIL! rc= 10
018 br331 TBR.sh f4c11dh pghgsul 6+3 ..............................FAIL! rc= 9
043 br531 TBR.sh s32c11dh pghgsul 6+3 .............................FAIL! rc= 9
tempest:
003 br131 TBR.sh e32c11dh pghgsul 6+3 .............................FAIL! rc= 9
020 br331 TBR.sh f4c11dh pghgsul 6+3 ..............................FAIL! rc= 9
034 br531 TBR.sh s32c11dh pghgsul 6+3 .............................FAIL! rc= 9
bangkok/lf95:
003 br111 TBR.sh e8c8mdm ttrac 6+3 ................................FAIL! rc= 9
015 br311 TBR.sh f10c8mdm ttrac 6+3 ...............................FAIL! rc= 9
031 br511 TBR.sh s8c8mdm ttrac 6+3 ................................FAIL! rc= 9
CAM tag used for the baseline comparison tests if different than previous
tag: prev
Summarize any changes to answers, i.e.,
- what code configurations:
- what platforms/compilers:
- nature of change (roundoff; larger than roundoff but same climate; new
climate): b4b
If bitwise differences were observed, how did you show they were no worse
than roundoff?
If this tag changes climate describe the run(s) done to evaluate the new
climate in enough detail that it(they) could be reproduced, i.e.,
- source tag (all code used must be in the repository):
- platform/compilers:
- configure commandline:
- build-namelist command (or complete namelist):
- MSS location of output:
MSS location of control simulations used to validate new climate:
URL for AMWG diagnostics output used to validate new climate:
===============================================================
===============================================================
cam3_2_41
Originator: jet ( John Truesdale)
Date: Fri Dec 9 11:56:01 MST 2005
Model: CAM
Version: CAM3.2.41
One-line Summary: Fixed SCAM bugs, added SCAM tests to testscript, made SCAM datasets consistent with CAM
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Ran test_driver.shl script: yes
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail: none (Get clearance for failed tests, enter lines from td.*.status files that fail)
bangkok:
003 br111 TBR.sh e8c8mdm ttrac 6+3 ................................FAIL! rc= 9
015 br311 TBR.sh f10c8mdm ttrac 6+3 ...............................FAIL! rc= 9
031 br511 TBR.sh s8c8mdm ttrac 6+3 ................................FAIL! rc= 9
tempest:
003 br131 TBR.sh e32c11dh pghgsul 6+3 .............................FAIL! rc= 9
020 br331 TBR.sh f4c11dh pghgsul 6+3 ..............................FAIL! rc= 9
034 br531 TBR.sh s32c11dh pghgsul 6+3 .............................FAIL! rc= 9
bluesky:
004 br131 TBR.sh e32c11dh pghgsul 6+3 .............................FAIL! rc= 10
018 br331 TBR.sh f4c11dh pghgsul 6+3 ..............................FAIL! rc= 10
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: One - fixed bug with scam specific code.
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
This is basically a bug fix for SCAM as well as adding some
small changes to get rid of autopromotion dependency in SCAM.
Changes are bit for bit for CAM/CCSM. I did update the datasets
used by SCM to by consistent with the latest datasets for CAM
so the CRM test output also needed to be updated to reflect this.
A list of files changed follows:
history_defaults.F90 - added cpp def BFB_CAM_SCAM_IOP around cam/scam
bfb specific code
runtime_opts.F90 - scam initializes with the name of the topo and initial
conditions datasets
startup_initialconds.F90 - made changes to include this file in scam
intial calls
inidat.F90 - changed scam specific code to use topo info off of topo dataset
.scam_defaults - updated to use topo file and newer boundary dataset files
scam/configure - Brian's changes to get rid of autopromotion
scam/testscript - added tests to check consistency of .scam_defaults
datasets with CAM namelist.
added tests for scam to run in default mode.
getnetcdfdata.F90 - Brian's changes to get rid of autopromotion
init_model.F90 - add topo dataset to scam
inital.F90 - call cam routine setup_initial from scam specific code.
ipc.h - added topo dataset to scam
runtype.h - added topo dataset to scam
scamMod.F90 - let scam initialize topo and initial conditions dataset names
scam_fifo.c - pass topo dataset to scam initialization
scam_run.F90 - fix bug with scam - don't run lsm if no land present
setlatlon.F90 - fix bug with scam - don't error out in iop run if iop
dataset has different column latitude and longitude than
initial data.
IOPSelectDateDlgImpl.cpp - cleanup
OptionsDlg.ui - add topo dataset to scam dataset tab in gui
OptionsDlgImpl.cpp - add topo dataset to scam dataset tab in gui
crm.cpp - add topo dataset to scam
dataset.cpp - add topo dataset to scam
fifomodel.cpp - add topo dataset to scam
manager.cpp - add topo dataset to scam
rpcmodel.cpp - add topo dataset to scam
crmtest26.out - crm radiation values changed slightly when we changed
the default datasets to match the current cam defaults
clm_camMod.F90 - fix bug with scam - don't run lsm if no land present also
get rid of spurious SCAM returns in this file
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_2_40
Originator: jedwards ( James Edwards)
Date: Fri Dec 2 17:24:53 MST 2005
Model: CAM
Version: CAM3.2.40
One-line Summary: minor cleanup and extended testing of cam3_2_39 commit
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Ran test_driver.shl script: yes
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail: testing was repeated against 3_2_38
bangkok:
003 br111 TBR.sh e8c8mdm ttrac 6+3 ................................FAIL! rc= 9
012 bl153 TBL.sh e64m outfrq24h -2 ................................FAIL! rc= 7
015 br311 TBR.sh f10c8mdm ttrac 6+3 ...............................FAIL! rc= 9
016 bl311 TBL.sh f10c8mdm ttrac 9 .................................FAIL! rc= 7
024 bl312 TBL.sh f10sdm ghgrmp 9 ..................................FAIL! rc= 7
026 bl313 TBL.sh f10pdm aqpgro 3 ..................................FAIL! rc= 7
028 bl353 TBL.sh f2m outfrq24h -2 .................................FAIL! rc= 7
031 br511 TBR.sh s8c8mdm ttrac 6+3 ................................FAIL! rc= 9
040 bl553 TBL.sh s64m outfrq24h -2 ................................FAIL! rc= 7
tempest:
003 br131 TBR.sh e32c11dh pghgsul 6+3 .............................FAIL! rc= 9
017 bl314 TBL.sh wg10dh outfrq3s 9 ................................FAIL! rc= 7
020 br331 TBR.sh f4c11dh pghgsul 6+3 ..............................FAIL! rc= 9
021 bl331 TBL.sh f4c11dh pghgsul 9 ................................FAIL! rc= 7
024 bl332 TBL.sh f4sdh ghgrmp 9 ...................................FAIL! rc= 7
026 bl333 TBL.sh f4pdh aqpgro 3 ...................................FAIL! rc= 7
031 bl352 TBL.sh f2o outfrq24h -2 .................................FAIL! rc= 7
034 br531 TBR.sh s32c11dh pghgsul 6+3 .............................FAIL! rc= 9
bluesky:
004 br131 TBR.sh e32c11dh pghgsul 6+3 .............................FAIL! rc= 9
015 bl151 TBL.sh e64h outfrq24h -2 ................................FAIL! rc= 7
018 br331 TBR.sh f4c11dh pghgsul 6+3 ..............................FAIL! rc= 9
019 bl331 TBL.sh f4c11dh pghgsul 9 ................................FAIL! rc= 7
022 bl332 TBL.sh f4sdh ghgrmp 9 ...................................FAIL! rc= 7
024 bl333 TBL.sh f4pdh aqpgro 3 ...................................FAIL! rc= 7
034 bl351 TBL.sh f2h outfrq24h -2 .................................FAIL! rc= 7
043 br531 TBR.sh s32c11dh pghgsul 6+3 .............................FAIL! rc= 10
054 bl551 TBL.sh s64h outfrq24h -2 ................................FAIL! rc= 7
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Brian Eaton
Restart files change: no
Changes answers: Yes (same-to-roundoff against 3_2_38)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
All reference to macros __FILE__ and __LINE__ have been removed.
File ozone_data.F90 was cleaned up to remove variables which are no longer used.
A further pergrow test was conducted with chemistry enabled since some of the changes in the previous commit involved chemistry but the pergrow test did not. This is why the test were rerun against 3_2_38 instead of 3_2_39.
Besides we wanted to close the CVS repository with a nice even number. :-)
===============================================================
===============================================================
cam3_2_39
Originator: jedwards ( James Edwards)
Date: Fri Dec 2 10:48:22 MST 2005
Model: CAM
Version: CAM3.2.39
One-line Summary: Consolidated zonal data input methods used by cam 1 chemistry and ozone files.
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Ran test_driver.shl script: yes
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail:
bluesky:
015 bl151 TBL.sh e64h outfrq24h -2 ................................FAIL! rc= 7
018 br331 TBR.sh f4c11dh pghgsul 6+3 ..............................FAIL! rc= 10
019 bl331 TBL.sh f4c11dh pghgsul 9 ................................FAIL! rc= 7
022 bl332 TBL.sh f4sdh ghgrmp 9 ...................................FAIL! rc= 7
024 bl333 TBL.sh f4pdh aqpgro 3 ...................................FAIL! rc= 7
034 bl351 TBL.sh f2h outfrq24h -2 .................................FAIL! rc= 7
043 br531 TBR.sh s32c11dh pghgsul 6+3 .............................FAIL! rc= 10
054 bl551 TBL.sh s64h outfrq24h -2 ................................FAIL! rc= 7
tempest:
003 br131 TBR.sh e32c11dh pghgsul 6+3 .............................FAIL! rc= 9
017 bl314 TBL.sh wg10dh outfrq3s 9 ................................FAIL! rc= 7
020 br331 TBR.sh f4c11dh pghgsul 6+3 ..............................FAIL! rc= 9
021 bl331 TBL.sh f4c11dh pghgsul 9 ................................FAIL! rc= 7
024 bl332 TBL.sh f4sdh ghgrmp 9 ...................................FAIL! rc= 7
026 bl333 TBL.sh f4pdh aqpgro 3 ...................................FAIL! rc= 7
031 bl352 TBL.sh f2o outfrq24h -2 .................................FAIL! rc= 7
034 br531 TBR.sh s32c11dh pghgsul 6+3 .............................FAIL! rc= 9
bangkok:
003 br111 TBR.sh e8c8mdm ttrac 6+3 ................................FAIL! rc= 9
012 bl153 TBL.sh e64m outfrq24h -2 ................................FAIL! rc= 7
015 br311 TBR.sh f10c8mdm ttrac 6+3 ...............................FAIL! rc= 9
016 bl311 TBL.sh f10c8mdm ttrac 9 .................................FAIL! rc= 7
024 bl312 TBL.sh f10sdm ghgrmp 9 ..................................FAIL! rc= 7
026 bl313 TBL.sh f10pdm aqpgro 3 ..................................FAIL! rc= 7
028 bl353 TBL.sh f2m outfrq24h -2 .................................FAIL! rc= 7
031 br511 TBR.sh s8c8mdm ttrac 6+3 ................................FAIL! rc= 9
040 bl553 TBL.sh s64m outfrq24h -2 ................................FAIL! rc= 7
041 sc999 scamtest.sh .............................................FAIL! rc= 4
6+3 tests are expected to fail due to problems in the testing procedure introduced in cam3_2_37.
TBL tests failed due to order of operation changes. A perturbation growth test was conducted and passed to show these are errors are order round-off.
The scamtest does not work for me in the test script, the scamtest was sucessfully run by hand.
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: Yes (same-to-roundoff - details noted above.)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Added file physics/cam1/zonalbndrydata.F90
moved netcdf interface in chemistry.F90 and ozone_data.F90 to this new file.
In the past the chemistry file read a full years data at the initial time and held it in memory while the ozone file read two months data and updated during the run when it needed new data. The new file supports both methods but requires that you choose one or the other at compile time. The second method is now the default.
Several minor changes were made in other files, mostly removing unused references to plat and plon. The interface was changed for one chemistry routine which required changes in all flavors of the chemistry files and dynamics inidat files. Below is a complete list of changed files:
models/atm/cam/src/chemistry/trop_mozart/chemistry.F90
models/atm/cam/src/chemistry/waccm_mozart/chemistry.F90
models/atm/cam/src/control/interpolate_data.F90
models/atm/cam/src/control/runtime_opts.F90
models/atm/cam/src/dynamics/eul/inidat.F90
models/atm/cam/src/dynamics/fv/inidat.F90
models/atm/cam/src/dynamics/sld/inidat.F90
models/atm/cam/src/physics/cam1/advnce.F90
models/atm/cam/src/physics/cam1/aerosol_intr.F90
models/atm/cam/src/physics/cam1/chemistry.F90
models/atm/cam/src/physics/cam1/drydep_mod.F90
models/atm/cam/src/physics/cam1/ozone_data.F90
models/atm/cam/src/physics/cam1/physpkg.F90
models/atm/cam/src/physics/cam1/ramp_scon.F90
models/atm/cam/src/physics/cam1/seasalt_intr.F90
models/atm/cam/src/physics/cam1/stratiform.F90
models/atm/cam/src/physics/cam1/tphysidl.F90
models/atm/cam/src/physics/cam1/tracers.F90
models/atm/cam/src/physics/cam1/volcanicmass.F90
models/atm/cam/src/physics/waccm/chemistry.F90
===============================================================
===============================================================
cam3_2_38
Originator: mirin ( Arthur Andrew Mirin)
Date: Tue Nov 29 16:50:01 MST 2005
Model: CAM
Version: CAM3.2.38
One-line Summary: (1) Restore FV options; (2) Cray-X1E communications, threading, streaming
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
No changes to file/directory structure
Substantial timing or memory changes: no
Requires change in run script: no
Ran test_driver.shl script: yes
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail: TBR.sh (Get clearance for failed tests, enter lines from td.*.status files that fail)
004 br131 TBR.sh e32c11dh pghgsul 6+3 .............................FAIL! rc= 9 (bluesky)
018 br331 TBR.sh f4c11dh pghgsul 6+3 ..............................FAIL! rc= 10 (bluesky)
043 br531 TBR.sh s32c11dh pghgsul 6+3 .............................FAIL! rc= 9 (bluesky)
003 br131 TBR.sh e32c11dh pghgsul 6+3 .............................FAIL! rc= 9 (tempest)
020 br331 TBR.sh f4c11dh pghgsul 6+3 ..............................FAIL! rc= 9 (tempest)
034 br531 TBR.sh s32c11dh pghgsul 6+3 .............................FAIL! rc= 9 (tempest)
003 br111 TBR.sh e8c8mdm ttrac 6+3 ................................FAIL! rc= 9 (bangkok)
015 br311 TBR.sh f10c8mdm ttrac 6+3 ...............................FAIL! rc= 9 (bangkok)
031 br511 TBR.sh s8c8mdm ttrac 6+3 ................................FAIL! rc= 9 (bangkok)
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Sawyer, Worley
Restart files change: no
Changes answers: no (bit-for-bit)
(Note: test_driver.sh only does checks this -- if you run it with "env BL_ROOT=")
Changes to CLM land-model: Cray-X1E OpenMP/CSD compatibility
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
(If not abort this tag and follow the procedures in)
http://www.cgd.ucar.edu/~cam/cam_checkins/checkin_process.shtml
Changes made:
(1) Restored nonworking FV options (geopktrans, tracertrans, modcomm-transpose,
modcomm_geopk)
(2) Added Shmem capability to Cray-X1E (USE_SHMEM if-def - affects mod_comm.F90)
(3) Allowed coexistence of OpenMP and CSD's on Cray-X1E provided they do not operate
on same loop index
(4) Removed nested OpenMP capability
Files changed:
models/atm/cam/bld: CAM_config.pm, config_trop_chem_mozart_defaults.xml, Makefile,
config_waccm_ghg_defaults.xml, config_cam_eul_defaults.xml, config_waccm_mozart_defaults.xml,
config_cam_fv_defaults.xml, configure, config_cam_sld_defaults.xml
models/atm/cam/src/control: runtime_opts.F90
models/atm/cam/src/dynamics/eul: dyndrv.F90, realloc4.F90, scan2.F90, scanslt.F90
models/atm/cam/src/dynamics/fv: FVCAM_GridCompMod.F90, geopk.F90, sw_core.F90,
cd_core.F90, pmgrid.F90, tp_core.F90, dynamics_vars.F90, spmd_dyn.F90, trac2d.F90
models/atm/cam/src/dynamics/sld: realloc4.F90, scan2.F90
models/atm/cam/src/physics/cam1: phys_grid.F90, radlw.F90
models/lnd/clm2/src/main: areaMod.F90, driver.F90, histFileMod.F90, atmdrvMod.F90,
filterMod.F90, initializeMod.F90
models/lnd/clm2/src/riverroute: RtmMod.F90
lnd/clm2/tools/mksurfdata: mkdynpftMod.F90, mklaiMod.F90, pmkpftMod.F90, mksoitex.F90,
mkglacier.F90, mklanwat.F90, mksoicol.F90, mkurban.F90
models/utils/pilgrim: mod_comm.F90
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_2_37
Originator: mvr ( Mathew Rothstein)
Date: Fri Nov 18 16:43:50 MST 2005
Model: CAM
Version: CAM3.2.37
One-line Summary: Bug fix for cam in ccsm; removed cam-specific code from csm_share; mod to branch test and other test suite cleanup
cam-bugs Requests resolved: none
Requires change in build system: yes
returned spmd mode to builds for ccsm (mistakenly removed with cam3_2_35)
Substantial timing or memory changes: yes, using spmd again in coupled mode
Requires change in run script: no
Ran test_driver.shl script: yes (does all of the following tests)
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail:
bluesky
004 br131 TBR.sh e32c11dh pghgsul 6+3 .............................FAIL! rc= 9
005 bl131 TBL.sh e32c11dh pghgsul 9 ...............................FAIL! rc= 6
008 bl132 TBL.sh e32sdh ghgrmp 9 ..................................FAIL! rc= 6
010 bl133 TBL.sh e32pdh aqpgro 3 ..................................FAIL! rc= 6
013 bl134 TBL.sh e32dh adia 9 .....................................FAIL! rc= 6
015 bl151 TBL.sh e64h outfrq24h -2 ................................FAIL! rc= 6
018 br331 TBR.sh f4c11dh pghgsul 6+3 ..............................FAIL! rc= 9
019 bl331 TBL.sh f4c11dh pghgsul 9 ................................FAIL! rc= 6
022 bl332 TBL.sh f4sdh ghgrmp 9 ...................................FAIL! rc= 6
024 bl333 TBL.sh f4pdh aqpgro 3 ...................................FAIL! rc= 6
027 bl334 TBL.sh f4dh adia 9 ......................................FAIL! rc= 6
030 bl336 TBL.sh wm4h outfrq3s 9 ..................................FAIL! rc= 6
034 bl351 TBL.sh f2h outfrq24h -2 .................................FAIL! rc= 6
037 bl354 TBL.sh fm2dh outfrq3s 9 .................................FAIL! rc= 6
040 bl355 TBL.sh fmo2dh off2x2.5 9 ................................FAIL! rc= 6
043 br531 TBR.sh s32c11dh pghgsul 6+3 .............................FAIL! rc= 9
044 bl531 TBL.sh s32c11dh pghgsul 9 ...............................FAIL! rc= 6
047 bl532 TBL.sh s32sdh ghgrmp 9 ..................................FAIL! rc= 6
049 bl533 TBL.sh s32pdh aqpgro 3 ..................................FAIL! rc= 6
052 bl534 TBL.sh s32dh adia 9 .....................................FAIL! rc= 6
054 bl551 TBL.sh s64h outfrq24h -2 ................................FAIL! rc= 6
055 cs999 TCS.ccsm.sh ER.01a T31_gx3v5 K ..........................FAIL! rc= 2
tempest
003 br131 TBR.sh e32c11dh pghgsul 6+3 .............................FAIL! rc= 9
004 bl131 TBL.sh e32c11dh pghgsul 9 ...............................FAIL! rc= 6
007 bl132 TBL.sh e32sdh ghgrmp 9 ..................................FAIL! rc= 6
009 bl133 TBL.sh e32pdh aqpgro 3 ..................................FAIL! rc= 6
012 bl135 TBL.sh e32dh idphys 9 ...................................FAIL! rc= 6
014 bl152 TBL.sh e64o outfrq24h -2 ................................FAIL! rc= 6
017 bl314 TBL.sh wg10dh outfrq3s 9 ................................FAIL! rc= 6
020 br331 TBR.sh f4c11dh pghgsul 6+3 ..............................FAIL! rc= 9
021 bl331 TBL.sh f4c11dh pghgsul 9 ................................FAIL! rc= 6
024 bl332 TBL.sh f4sdh ghgrmp 9 ...................................FAIL! rc= 6
026 bl333 TBL.sh f4pdh aqpgro 3 ...................................FAIL! rc= 6
029 bl335 TBL.sh f4dh idphys 9 ....................................FAIL! rc= 6
031 bl352 TBL.sh f2o outfrq24h -2 .................................FAIL! rc= 6
034 br531 TBR.sh s32c11dh pghgsul 6+3 .............................FAIL! rc= 9
035 bl531 TBL.sh s32c11dh pghgsul 9 ...............................FAIL! rc= 6
038 bl532 TBL.sh s32sdh ghgrmp 9 ..................................FAIL! rc= 6
040 bl533 TBL.sh s32pdh aqpgro 3 ..................................FAIL! rc= 6
043 bl535 TBL.sh s32dh idphys 9 ...................................FAIL! rc= 6
045 bl552 TBL.sh s64o outfrq24h -2 ................................FAIL! rc= 6
bangkok
003 br111 TBR.sh e8c8mdm ttrac 6+3 ................................FAIL! rc= 9
004 bl111 TBL.sh e8c8mdm ttrac 9 ..................................FAIL! rc= 6
008 bl112 TBL.sh e8sdm ghgrmp 9 ...................................FAIL! rc= 6
010 bl113 TBL.sh e8pdm aqpgro 3 ...................................FAIL! rc= 6
012 bl153 TBL.sh e64m outfrq24h -2 ................................FAIL! rc= 6
015 br311 TBR.sh f10c8mdm ttrac 6+3 ...............................FAIL! rc= 9
016 bl311 TBL.sh f10c8mdm ttrac 9 .................................FAIL! rc= 6
024 bl312 TBL.sh f10sdm ghgrmp 9 ..................................FAIL! rc= 6
026 bl313 TBL.sh f10pdm aqpgro 3 ..................................FAIL! rc= 6
028 bl353 TBL.sh f2m outfrq24h -2 .................................FAIL! rc= 6
031 br511 TBR.sh s8c8mdm ttrac 6+3 ................................FAIL! rc= 9
032 bl511 TBL.sh s8c8mdm ttrac 9 ..................................FAIL! rc= 6
036 bl512 TBL.sh s8sdm ghgrmp 9 ...................................FAIL! rc= 6
038 bl513 TBL.sh s8pdm aqpgro 3 ...................................FAIL! rc= 6
040 bl553 TBL.sh s64m outfrq24h -2 ................................FAIL! rc= 6
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, eaton
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
(If not abort this tag and follow the procedures in)
http://www.cgd.ucar.edu/~cam/cam_checkins/checkin_process.shtml
Changes made:
M models/atm/cam/bld/configure
- bug fix for running cam in coupled mode with spmd
M models/atm/cam/src/advection/slt/phcs.F90
M models/atm/cam/src/chemistry/waccm_mozart/apex_subs.F90
M models/atm/cam/src/control/gauaw_mod.F90
M models/atm/cam/src/dynamics/fv/cd_core.F90
M models/atm/cam/src/dynamics/fv/fill_module.F90
M models/atm/cam/src/dynamics/fv/geopk.F90
M models/atm/cam/src/dynamics/fv/pft_module.F90
M models/atm/cam/src/dynamics/fv/sw_core.F90
M models/csm_share/shr/shr_kind_mod.F90
- moved cam-specific code out of csm_share
M models/atm/cam/test/system/TBR.sh
M models/atm/cam/test/system/input_tests_master
- modified branch test to branch from an earlier restart file
M models/atm/cam/test/system/TCB.ccsm.sh
M models/atm/cam/test/system/TSB.ccsm.sh
M models/atm/cam/test/system/test_driver.sh
- cleaned up use of environment vars CAMROOT and CAM_ROOT
M models/atm/cam/test/system/td_nightly_aix.sh
M models/atm/cam/test/system/td_nightly_linux.sh
- just updating repository with mods to nightly scripts
M models/atm/cam/test/system/TCB.ccsm.sh
- ccsm test will now use tags from ccsm3_1_beta* sequence
M models/atm/cam/test/system/TBL.sh
M models/atm/cam/test/system/nl_files/adia
M models/atm/cam/test/system/nl_files/aqpgro
M models/atm/cam/test/system/nl_files/ghgrmp
M models/atm/cam/test/system/nl_files/idphys
M models/atm/cam/test/system/nl_files/no_ttrac
M models/atm/cam/test/system/nl_files/off2x2.5
M models/atm/cam/test/system/nl_files/outfrq24h
M models/atm/cam/test/system/nl_files/outfrq3s
M models/atm/cam/test/system/nl_files/pghgsul
M models/atm/cam/test/system/nl_files/ttrac
M models/atm/cam/test/system/nl_files/ttrac_lb1
M models/atm/cam/test/system/nl_files/ttrac_lb2
M models/atm/cam/test/system/nl_files/ttrac_lb3
- test suite now outputting history/restart files for each time sample
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_2_36
Originator: jedwards ( James Edwards)
Date: Fri Nov 11 09:58:15 MST 2005
Model: CAM
Version: CAM3.2.36
One-line Summary: rework of new interpolation method, bug fix in phys_gmean, clean up in waccm_mozart
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Ran test_driver.shl script: yes
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail: none
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Brian Eaton
Restart files change: no
Changes answers: no
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Reworked recently introduced interpolation changes to contain the method within a physics chunk and eliminate dependence on the phys_grid module.
Fixed a bug in phys_gmean in which the number of fields passed in was hardcoded, and the incoming number of fields argument was ignored.
Removed unused 'use ' statements in waccm_mozart.
===============================================================
===============================================================
cam3_2_35
Originator: erik ( Erik Kluzek)
Date: Tue Nov 8 12:41:27 MST 2005
Model: CAM
Version: CAM3.2.35
One-line Summary: Remove prototype ESMF and use WRF implimentation of ESMF_2_1_0 API for time-manager, allow build to use external ESMF
cam-bugs Requests resolved: none
Requires change in build system: yes
(Configure changed so that Filepath also points to models/utils/esmf_wrf_timemgr)
Substantial timing or memory changes: no
Requires change in run script: no
Ran test_driver.shl script: yes (does all of the following tests)
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail: none
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: yes
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Remove: models/utils/esmf:
Remove all prototype ESMF files.
Add: models/utils/esmf_wrf_timemgr:
Put in WRF implementation of ESMF_2_1_0 time-manager API.
Changes to build:
Change build so default is to build with WRF ESMF time-manager. Can also use
externally compiled ESMF_2_1_0.
Linking with external ESMF: non-default, have to build ESMF externally
and send appropriate command-line options to configure to use this option.
Get rid of ESMF_ROOT as unneeded now. Now have three
tokens for dealing with ESMF: ESMF_BLD, ESMF_LIB, and ESMF_MOD.
Setting any of those threes means to link with external ESMF.
If ESMF_BLD is set, the Makefile will figure out the path
based on uname and debug option to the default version of ESMF.
If ESMF_LIB and ESMF_MOD are set the full paths will be used
for these two. The external version of ESMF that you link to is
assumed to be a current download of ESMF, which has it's own build
system. The current ESMF does not have a build for several important
platforms: Cray, Pathscale, ES, NEC etc, therefore only the esmf_wrf_timemgr
version can be used there. Also there were bugs with linking that will
need to be worked out on our standard platforms, before this will work
reliably
Change configure so SPMD information isn't duplicated in configure output
files.
Changes to time_manager:
Update time_manager.F90 so uses the ESMF_2_1_0 API instead of the prototype
ESMF API. Create unit-tests in models/atm/cam/tests/unit/control to do
extensive testing of the time-manager interface. Runs different time-intervals and different ways of setting the stop-date, with one simulation period from
year 0 to year 9999. Answers were compared to the previous time-manager and
found to be bit-for-bit on IBM, SGI and Linux-PGI compilers.
Calendar can not be set to GREGORIAN right now, only NOLEAP is allowed.
esmf_wrf_timemgr could be extended to allow both calendars, or we can start
using ESMF externally to handle both calendars.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_2_34
Originator: jedwards ( James Edwards)
Date: Thu Nov 3 17:00:55 MST 2005
Model: CAM
Version: CAM3.2.34
One-line Summary: interpolation method changed in waccm chemistry
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: Yes reduced memory profile of waccm/chemistry
Requires change in run script: no
Ran test_driver.shl script: yes
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail: none
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None (one-line description)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
plat and plon dependence were removed from waccm/chemistry.F90
following the same method as cam1/chemistry.F90
The method in cam1 was modified to read only on the masterproc, this does
not change the memory profile.
===============================================================
===============================================================
cam3_2_32
Originator: mvr ( Mathew Rothstein)
Date: Fri Oct 28 16:58:34 MDT 2005
Model: CAM
Version: CAM3.2.32
One-line Summary: Update csm_share to share3_051025; update remaining clm files to clm3_expa_45 that were missed with previous tag
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Ran test_driver.sh script: yes (does all of the following tests)
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail: none
Tested to work coupled with CCSM : yes
Tested to work with SCAM : yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: yes, updated doc files to clm3_expa_45
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
M models/csm_share/ChangeLog
M models/csm_share/ChangeSum
M models/csm_share/cpl/cpl_contract_mod.F90
M models/csm_share/cpl/cpl_domain_mod.F90
M models/csm_share/cpl/cpl_fields_mod.F90
M models/csm_share/cpl/cpl_fields_mod.F90.CO2A
M models/csm_share/cpl/cpl_fields_mod.F90.CO2B
M models/csm_share/cpl/cpl_fields_mod.F90.CO2_DMSA
M models/csm_share/cpl/cpl_interface_mod.F90
M models/csm_share/shr/shr_const_mod.F90
M models/csm_share/shr/shr_map_mod.F90
M models/csm_share/shr/shr_ncread_mod.F90
M models/csm_share/shr/shr_stream_mod.F90
M models/csm_share/shr/shr_timer_mod.F90
M models/csm_share/unit_testers/test_shr_orb.F90
M models/csm_share/unit_testers/test_shr_sys.F90
- these are the files that changed between share3_050909 and share3_051025
M models/lnd/clm2/doc/ChangeLog
M models/lnd/clm2/doc/ChangeSum
- these are the files missed in updating between clm3_expa_29 and clm3_expa_45
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_2_31
Originator: mvertens ( Mariana Vertenstein)
Date: Wed Oct 26 15:24:06 MDT 2005
Model: CAM
Version: CAM3.2.31
One-line Summary: updated clm to clm3_expa_45 (removed condition that clm and cam grid had to be bfb)
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: Yes (describe)
Requires change in run script: no
Ran test_driver.shl script: No (does all of the following tests)
Machines tested: IBM (ran 2 climate simulations on IBM - see below)
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): no
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: yes
Changes answers: Yes (same-climate)
Changes to CLM land-model: None (one-line description)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
No cam physics code was modified. Only clm interface code was modified.
Modified models/lnd/clm2/src/main/clm_camMod.F90 (only several lines of clm code
were commented out) so that cam and clm grid do not have to be bfb. The
tolerance is that the grid latitudes/longitudes grids match to 1.e-12.
The fractional land however, must be identical. The clm code base has
now been updated to clm3_expa_45.
Two 10 year integration for the entire set of physics changes using climatological SSTs
were performed. See:
/MVERTENS/csm/cam3_2_29_fv4x5dom (cam3_2_29 with changes to clm_camMod.F90)
vs
MVERTENs/csm/cam3_2_29_cam3_2_29_fv4x5dom_0 (cam3_2_29 with no changes)
Diagnostic output is at:
http://www.cgd.ucar.edu/~mvertens/AMWG_diag/cam3229_fv4x5dom-cam3229_fv4x5dom_0/
===============================================================
===============================================================
cam3_2_30
Originator: mvr ( Mathew Rothstein)
Date: Thu Oct 20 09:49:40 MDT 2005
Model: CAM
Version: CAM3.2.30
One-line Summary: mods / new scripts added to cam's test suite for automated testing of new tags
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Ran test_driver.shl script: yes (does all of the following tests)
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail: none
Tested to work coupled with CCSM : yes
Tested to work with SCAM : yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
A models/atm/cam/test/system/input_tests_aix_nightly
- test ids to run as part of nightly testing of new cam tags on bluesky (aix)
A models/atm/cam/test/system/input_tests_linux_nightly
- test ids to run as part of nightly testing of new cam tags on bangkok (linux)
A models/atm/cam/test/system/td_nightly_aix.sh
- cron script for overnight testing of new cam tag on bluesky (aix)
A models/atm/cam/test/system/td_nightly_linux.sh
- cron script for overnight testing of new cam tag on bangkok (linux)
A models/atm/cam/test/system/update_collections.sh
- cron script for overnight loading of latest cam tag on machine sanitas
M models/atm/cam/test/system/TCB.ccsm.sh
- added command to modify directory path to finite volume code in ccsm scripts
M models/atm/cam/test/system/input_tests_master
- added ccsm tests for running coupled model with finite volume dycore
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_2_29
Originator: eaton ( Brian Eaton)
Date: Tue Oct 18 10:52:05 MDT 2005
Model: CAM
Version: CAM3.2.29
One-line Summary: add LANDFRAC file, move inti.F90 into phys_init
cam-bugs Requests resolved: bugzilla #19
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Ran test_driver.shl script: yes
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail: none
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Added the ability to specify the file that contains the LANDFRAC field.
The namelist variable bnd_landfrac is used to specify the path of this
file. If not specified the default is to use the file that the topography
fields are read from.
Moved the physics initialization routines into phys_init. Removed
subroutine inti. Also add initialization of grid information to the
physics_state objects that are allocated in phys_init. This information is
needed to be passed to the physics init routines that do spatial
interpolation of boundary data.
Misc:
. Fix for problem with WACCM_MOZART on SGI: was failing to write abs/ems
restart file. Resolves bugzilla #19.
. Bugfix in lf95 section of Makefile (reference to apex_subs.F changed to
apex_subs.F90).
===============================================================
===============================================================
cam3_2_28
Originator: mvertens ( Mariana Vertenstein)
Date: Mon Oct 17 09:35:36 MDT 2005
Model: CAM
Version: CAM3.2.28
One-line Summary: updated clm code clm3_expa_44
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Ran test_driver.shl script: yes (does all of the following tests)
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail: none
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: yes (all clm restarts, other than history restarts, are now NetCDF)
Changes answers: no (bit-for-bit)
Changes to CLM land-model: yes
Updated clm code to clm3_expa_44
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Updated clm code to clm3_expa_44. Interfaces between cam and clm
(in clm_camMod.F90) have been modified to be consistent with
making clm a gridded component. The clm driver routine has now
also been split into two phases as part of this process.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_2_27
Originator: jedwards ( James Edwards)
Date: Fri Oct 14 10:20:58 MDT 2005
Model: CAM
Version: CAM3.2.27
One-line Summary: consolidated linear interpolation schemes
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
(one-line description of changes: Makefile, include files, directories, etc.)
(Detailed description below)
Substantial timing or memory changes: no (describe)
Requires change in run script: no
(one-line description of changes: data files, run-scripts, namelist etc.)
(Detailed description below)
Ran test_driver.shl script: yes (does all of the following tests)
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail: none
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
(Note: test_driver.sh only does checks this -- if you run it with "env BL_ROOT=")
Changes to CLM land-model: None (one-line description)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Merged lininterp.F90 bilin.F90 vertinterp.F90 and timinterp.F90 into a single module. added use statements for this module where appropriate
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_2_26
Originator: pworley ( Patrick H Worley)
Date: Wed Oct 12 21:20:24 MDT 2005
Model: CAM
Version: CAM3.2.26
One-line Summary: Cray X1(E) optimizations for FV
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: Yes - much faster on the X1E when using FV
Requires change in run script: no
Ran test_driver.shl script: yes
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail: failed all baseline comparisons tests for FV on bluesky.
019 bl331 TBL.sh f4c11dh pghgsul 9 ................................FAIL! rc= 7
022 bl332 TBL.sh f4sdh ghgrmp 9 ...................................FAIL! rc= 7
024 bl333 TBL.sh f4pdh aqpgro 3 ...................................FAIL! rc= 7
027 bl334 TBL.sh f4dh adia 9 ......................................FAIL! rc= 7
030 bl336 TBL.sh wm4h outfrq3s 9 ..................................FAIL! rc= 7
034 bl351 TBL.sh f2h outfrq24h -2 .................................FAIL! rc= 7
037 bl354 TBL.sh fm2dh outfrq3s 9 .................................FAIL! rc= 7
040 bl355 TBL.sh fmo2dh off2x2.5 9 ................................FAIL! rc= 7
All other tests passed, and all tests passed on other two systems. pergro test
for FV on bluesky passed, indicating that change was within roundoff.
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Art Mirin, Will Sawyer
Restart files change: no
Changes answers: No for EUL and SLD, No for FV on SGI and Linux-Lahey, Yes for FV on IBM (same-to-roundoff)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Synchronizing FV with NASA version introduced code that does run
efficiently on the Cray vector system (and probably not on the Earth Simulator
either). These changes bring performance back to what it was before, as well
as vectorizing buffer copies in the communication layer.
dynamics/fv/cd_core.F90, dynamics/fv/sw_core.F90
eliminated unnecessary array segment specifications in subroutine calls where it
was causing array copying
dynamics/fv/mapz_module.F90
restructured recently added code to improve vectorization
dynamics/fv/trac2d.F90
replaced pointer-based implementation of double buffering algorithm with an
equivalent index-based implementation, working around a problem on the Cray X1E
utils/pilgrim/mod_comm.F90, utils/pilgrim/parutilitiesmodule.F90
restructured buffer copy loops for improved vectorization; general clean-up and optimization
Some of the timer logic was also broken in a recent check-in:
control/cam.F90, control/cam_comp.F90
fixed timer logic
utils/timing/gptl.c
fixed problems in mpi_wtime support in timing routines
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_2_25
Originator: jeff ( Yen-Huei Lee)
Date: Fri Oct 7 09:16:46 MDT 2005
Model: CAM
Version: CAM3.2.25
One-line Summary: user specified constant co2 when running in ccsm
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
(one-line description of changes: Makefile, include files, directories, etc.)
(Detailed description below)
Substantial timing or memory changes: no or Yes (describe)
Requires change in run script: no
(one-line description of changes: data files, run-scripts, namelist etc.)
(Detailed description below)
Ran test_driver.shl script: yes (does all of the following tests)
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail: none (Get clearance for failed tests, enter lines from td.*.status files that fail)
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: Brian Eaton, Mathew Rothstein
Restart files change: no
Changes answers: no (bit-for-bit)
(Note: test_driver.sh only does checks this -- if you run it with "env BL_ROOT=")
Changes to CLM land-model: None (one-line description)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
(If not abort this tag and follow the procedures in)
http://www.cgd.ucar.edu/~cam/cam_checkins/checkin_process.shtml
Changes made:
modified the following files:
models/atm/cam/src/control/runtime_opts.F90
models/atm/cam/src/physics/cam1/radiation.F90
models/atm/cam/src/physics/cam1/chem_surfvals.F90
models/atm/cam/src/chemistry/waccm_mozart/chem_surfvals.F90
- add one namelist variable to allow user to specify a constant co2 (vmr)
for radiation calculation when running in CCSM, indepedent of other settings
of co2.
models/atm/cam/src/control/ccsm_msg.F90
- mod to support 1x1.25 resolution when running in CCSM.
models/atm/cam/test/systems/TSB.ccsm.sh
- mod to run ccsm test effectively.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_2_24
Originator: erik ( Erik Kluzek)
Date: Mon Oct 3 13:21:24 MDT 2005
Model: CAM
Version: CAM3.2.24
One-line Summary: Move time-loop to top driver, make multiple run methods in cam_comp and
stepon modules to reduce code duplication
cam-bugs Requests resolved: none
Requires change in build system: no (Remove auto-promotion on Darwin)
Substantial timing or memory changes: no
Requires change in run script: no
Ran test_driver.shl script: yes (does all of the following tests)
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail: none
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, eaton, mvertens, mvr (testscript changes)
Restart files change: yes
Changes answers: no (bit-for-bit)
Changes to CLM land-model: Yes! (clm_camMod only, move allocation method inside, create final method)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
High level structure changes:
Make multiple run methods in cam_comp and stepon to reduce duplication and move
the time-loop up to the top level driver seq_ccsm -- still called cam.F90.
Remove nlend from comctl and make local data, passing down to relevant subroutines
and functions that need it. Make Gaussian weights and etamid module data
in cam_comp. Move all initialization calls for ocean or ice into one initialization
call (calls to read_restart_xxx moves to inside relevant initialize subroutine). And
move write_restart_xxx calls for ice and ocn inside of ice and ocean run method. Create
final methods for all component models (ice, ocn, land, atm). Use ncol in derived types
rather than using get_ncols_p method. Move tssub from surface_state derived type to private
data inside of camice. Move allocation of input/output derived types from driver to inside
of relevant component (ice, ocean, or land). Remove the temporary place-holder
restart_camsrf.F90 surface restart file. Remove ice_frac in camice and start using areafrac
and move merge of fractions to camhub_comp. This takes care of CCSM/ESMF Stage 1-2.6, 1.4
and 1.7, and some work on 1.9.1. This creates drivers and component modules that have the
bulk of the Pre-ESMF work complete.
See the "CCSM/ESMF Stage-1 Release Schedule" at...
http://swiki.ucar.edu/start/66
SCAM Changes:
Get SCAM and testscript to run on Darwin. Make Eulerian data dimensioned plat/2 to (plat+1)/2,
so will compile correctly with SCAM. Add "skip" and "noclean" as command-line arguments to
SCAM testscript to skip the CAM build/run and/or don't clean before the build of CAM or SCAM.
These are both useful for using testscript interactively for debugging running SCAM. Add
ability to testscript to run CAM at different resolutions and operate over any point
given (it gets the latitude and longitude needed for SCAM from the IOP file produced).
CAM Makefile change:
Get rid of auto-promotion on Darwin compile (Mac-OS-X with Absoft IBM XLF/XLC compilers).
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_2_23
Originator: pworley ( Patrick H Worley)
Date: Thu Sep 29 15:42:12 MDT 2005
Model: CAM
Version: CAM3.2.23
One-line Summary: Explicitly typed variables and constants and deleted autopromotion of reals in Makefile.
cam-bugs Requests resolved: none
Requires change in build system: Yes - removed autopromotion flags from FFLAGS
Substantial timing or memory changes: no
Requires change in run script: no
Ran test_driver.shl script: yes
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail: none
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Brian Eaton, Art Mirin, Forrest Hoffman, Michael Ham
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: Yes - Explicitly typed variables and constants
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Nearly every file in CAM standalone was modified, some extensively, but most
changes were "syntactic". Note that this includes most Fortran source in csm_share, ice,
lnd, and utils as well as in atm. Only the tool subdirectories were "spared".
The steps were as follows:
a) in atm/cam/src, the routines
chemistry/waccm_mozart/apex_subs.F
chemistry/waccm_mozart/svdcmp.F
chemistry/waccm_mozart/wei96.F
physics/waccm/msise00.F
utils/fft99.F
utils/sgexx.F
were all changed from fixed format that used implicit typing
to free free format with "implicit none"
and explicitly typing of all variables. (The suffixes
were all changed from .F to .F90 as well.)
b) Tom Henderson's tool "addrealkind" was used to
change "real" declarations to "real(r8) and to add
"_r8" to all floating point constants that weren't already typed.
"use shr_kind_mod" was added where needed.
(For csm_share/shr. SHR_KIND_R8 was used instead of r8.
In csm_share/cpl, R8 was used. In ice/csim4, dbl_kind was used.)
(For Tom's tool to work correctly, the name of the module was appended
to "end module" in a number of files, and "end" was replaced
by "end subroutine xxx" or "end function yyy" in a number of
files as well.)
complex declarations were also changed to complex(r8)
c) float(iii) and real(iii) were replaced by real(iii,r8).
cmplx(xxx,yyy) was replaced by cmplx(xxx,yyy,r8).
dble(iii) was replaced by real(iii,r16) if r16 was really necessary,
otherwise it was replaced by real(iii,r8).
Constants of the form xxx.yyyDeee were replaced by xxx.yyyEeee_r16
if r16 was necessary, otherwise were replaced by xxx.yyy.Eeee_r8.
d) specific intrinsics were changed to generic intrinsics,
e.g. alog10 was changed to log10, etc.
e) removed local definitions of r8 and r16 when they were identical to
those in shr_kind_mod, but left the local definitions as is if they
were different.
f) compiler flags that defined real variables to be double precision (or real*8)
and/or promoted floating point constants and intrinsics to double precision
were removed from the CAM makefile. (Something similar will need to
be done in CCSM.)
Finally, a new command was added to phys_grid to indicated whether phys_grid
had been called or not. This logical function was used in camhub_comp.F90
in place of the existing error-prone tests for this condition.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_2_22
Originator: erik ( Erik Kluzek)
Date: Tue Sep 20 14:19:53 MDT 2005
Model: CAM
Version: CAM3.2.22
One-line Summary: Create cam_comp atmosphere component and make stepon a module, add Darwin (Mac-OS-X) as a platform
cam-bugs Requests resolved: none
Requires change in build system: no
(Did add changes to configure/build system so that can compile on Mac-OS-X with
Absoft XLC/XLF compilers.
Substantial timing or memory changes: no
Requires change in run script: no
Ran test_driver.shl script: yes (does all of the following tests)
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail: none
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, eaton, mvertens
Restart files change: yes
Changes answers: no (bit-for-bit)
Changes to CLM land-model: Interface with cam_comp changes
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Move the cam3_2_21_brnchT_chghilev_8 tag onto cam_dev.
Add changes to esmf, cam, scam, and cprnc to build/run on Darwin (Mac-OS-X) platform
with the ABSOFT IBM XLF/XLC compilers. On Darwin scam will compile, but not link
at this point. More work needs to be done for it to completely build and run.
Create a cam_comp component module to encapsulate the atmosphere part
of CAM. Make stepon into a module with initial, run and final methods.
Have scam interface with cam_comp rather than underlying subroutines.
Have scam interface with runtime_options in a more straightforward manner,
that won't require duplication of default settings for SCAM inside scam
initialization. Rename scam_inital to inital module a module for scam replacing
the Eulerian inital.
The next step is to make cam_comp have multiple run phases, and stepon have
multiple run phases (this is outlined in each stepon). Then bring the time-loop
out of stepon and up to the driver level. The driver is still called cam.F90, but
will soon be refered to as sequential CCSM to distinquish it from cam_comp the
atmosphere only component, of the stand-alone CAM modeling system.
This is documented as Stage-1.2-6 on the CCSM/ESMF webpages....
http://swiki.ucar.edu/start/83
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_2_21
Originator: mvr ( Mathew Rothstein)
Date: Mon Sep 12 14:34:15 MDT 2005
Model: CAM
Version: CAM3.2.21
One-line Summary: re-work of ccsm test within cam test suite; updated to latest version of share code; mods to help waccm work in ccsm mode
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Ran test_driver.shl script: yes (does all of the following tests)
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail: none
Tested to work coupled with CCSM: yes
Tested to work with SCAM: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
(If not abort this tag and follow the procedures in)
http://www.cgd.ucar.edu/~cam/cam_checkins/checkin_process.shtml
Changes made:
A models/atm/cam/test/system/gen_test_table.sh
- utility to generate html file with table of tests currently included in cam test suite
A models/csm_share/cpl/cpl_fields_mod.F90.CO2A
A models/csm_share/cpl/cpl_fields_mod.F90.CO2B
A models/csm_share/cpl/cpl_fields_mod.F90.CO2_DMSA
A models/csm_share/shr/shr_map_mod.F90
A models/csm_share/shr/shr_ncread_mod.F90
A models/csm_share/shr/shr_stream_mod.F90
A models/csm_share/shr/shr_string_mod.F90
A models/csm_share/shr/shr_tInterp_mod.F90
- added as part of update of share code from share3_0_2 to share3_050909
M models/atm/cam/bld/configure
- changed filepath for ccsm builds to correct location of waccm src
M models/atm/cam/src/physics/waccm/msise00.F
- changed the name of variable MSS to avoid ccsm pre-processor confusion
M models/atm/cam/test/system/TCB.ccsm.sh
- redirection of ccsm build test output to appropriate log file and some cleanup
M models/atm/cam/test/system/TCS.ccsm.sh
- re-working of ccsm test within cam test suite
M models/atm/cam/test/system/TMC.sh
- added some error checking to mass conservation test script
M models/atm/cam/test/system/TSB.ccsm.sh
- re-working of ccsm test within cam test suite
M models/atm/cam/test/system/test_driver.sh
- remove references to certain job log file if running interactively
M models/csm_share/ChangeLog
M models/csm_share/ChangeSum
M models/csm_share/cpl/cpl_bundle_mod.F90
M models/csm_share/cpl/cpl_comm_mod.F90
M models/csm_share/cpl/cpl_contract_mod.F90
M models/csm_share/cpl/cpl_control_mod.F90
M models/csm_share/cpl/cpl_domain_mod.F90
M models/csm_share/cpl/cpl_fields_mod.F90
M models/csm_share/cpl/cpl_infobuf_mod.F90
M models/csm_share/cpl/cpl_interface_mod.F90
M models/csm_share/cpl/cpl_iobin_mod.F90
M models/csm_share/cpl/cpl_iocdf_mod.F90
M models/csm_share/cpl/cpl_map_mod.F90
M models/csm_share/cpl/cpl_mct_mod.F90
M models/csm_share/shr/shr_cal_mod.F90
M models/csm_share/shr/shr_msg_mod.F90
- modified as part of update of share code from share3_0_2 to share3_050909
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_2_20
Originator: jedwards ( James Edwards)
Date: Wed Sep 7 14:22:57 MDT 2005
Model: CAM
Version: CAM3.2.20
One-line Summary: removed the comhyb.h commonblock, made hycoef a module.
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
(one-line description of changes: Makefile, include files, directories, etc.)
(Detailed description below)
Substantial timing or memory changes: no or Yes (describe)
Requires change in run script: no
(one-line description of changes: data files, run-scripts, namelist etc.)
(Detailed description below)
Ran test_driver.shl script: yes (does all of the following tests)
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail: 055 cs999 TCS.ccsm.sh ER.01a T31_gx3v5 K
cam3_2_19 fails ccsm build, that failure is carried forward.
Tested to work coupled with CCSM (create_ccsmcam): no - see above
Tested to work with SCAM (tools/scam/testscript): yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None (one-line description)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
changes made: removed the file comhyb.h and created a module hycoef. subroutine hycoef is renamed hycoef_init and the common block variables from comhyb
are now module variables. Replaced #include with
use hycoef, only :
this change affected a number of files across the structure.
===============================================================
===============================================================
cam3_2_19
Originator: pworley (Pat Worley), eaton ( Brian Eaton)
Date: Tue Sep 6 16:01:22 MDT 2005
Model: CAM
Version: CAM3.2.19
One-line Summary: cray performance, XT3 support, timing lib mods
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: Yes - restore Cray performance to levels before commit of cam3_2_1;
also, timing lib changes improve SGI Altix performance significantly.
Requires change in run script: no
Ran test_driver.sh script: yes
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail: The CCSM test is failing because the CCSM scripts don't use CAM's Makefile. Will fix
the test by updating the appropriate CCSM template file.
Tested to work coupled with CCSM (create_ccsmcam): fails
Tested to work with SCAM (tools/scam/testscript): yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: Worley, Eaton, Conley
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
> Makefile
> - set default C compiler to mpcc_r for AIX (was "cc")
> - added support for Cray XT3 in Linux section.
> (For XT3, set
> USER_FC := ftn
> USER_CC := cc
> and add -DCATAMOUNT to USER_CPPDEFS ).
>
> control/cam.F90
> - added four timer calls (mpi_wtime), to get basic
> performance data (total, stepon, SYPD based on stepon time)
> even when -DDISABLE_TIMERS is specified.
> - moved first t_start calls to after spmdinit is called,
> to allow mpi_wtime to be used in the timing library
> - moved trunc call to after first t_start call, so as to
> include in measured initialization phase
> - #ifdef CATAMOUNT
> ! Improve performance of writing to standard out and error on XT3
> ! (using Catamount kernel) by increasing buffer size.
> call usetlbuf()
> #endif
>
> control/get_memusage.c
> - Linux branch of get_memusage breaks the code on the XT3.
> Disabled the routine when CATAMOUNT defined.
>
> #ifdef CATAMOUNT
> return -1;
> #else ...
>
> dynamics/fv/cd_core.F90
> - added CSD call (Cray Streaming Directive, for use on Cray X1(E) )
> that was mistakenly not included in earlier check-in.
>
> physics/cam1/aerosol_radiation_interface.F90
> - promoted local 1D arrays to 2D; added CSD; moved diagnostic output
> to separate loop; in order to recover vectorization and streaming
> lost in cam3_2_1 check-in
>
> physics/cam1/radsw.F90
> - deleted hanging CSD region delimiter
>
> utils/timing/gptl.c
> - added support for using mpi_wtime instead of gettimeofday when
> SPMD specified. (This improves performance on SGI Altix significantly,
> and allows timers to be used on Cray XT3.)
> - added call to rtcrate on Cray X1(E) systems, removing hardcoded
> ticks_per_sec value (that is different on X1 and X1E).
> - increased number of digits in timing data output
>
> utils/timing/private.h
> - defined fields needed when using mpi_wtime to measure time
>
> 2 files added:
>
> control/uselbuf.c
>
> - set system buffering for output to stdout and stderr.
> Needed on Cray XT3 to workaround performance problem.
>
> utils/timing/rtcrate.F90
>
> - C-callable Fortran function that calls irtc_rate intrinsic
> and returns the value (when #ifdef UNICOSMP).
===============================================================
===============================================================
cam3_2_18
Originator: jedwards (Jim Edwards) eaton (Brian Eaton)
Date: Mon Sep 5 09:42:57 MDT 2005
Model: CAM
Version: CAM3.2.18
One-line Summary: spmd_utils module mods
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Ran test_driver.shl script: yes
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail: none
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: jedwards, eaton
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Consolidate code into spmd_utils.F90 that sets up the generic spmd
environment. In particular, move pmgrid variables masterproc and iam,
spmd_dyn variables npes, nsmps, proc_smp_map, and the subroutines from the
files spmdinit.F90 and swap_comm.F90 into the spmd_utils module. The
subroutine spmdinit has been modified to initialize these public variables
whether SPMD is defined or not. Hence the "ifdef SPMD" conditional has
been removed from the call to spmdinit in the main cam program.
Because these variables are so widely used this change touches about 150
files. Most the changes are only one or two lines of modified "use"
association statements.
===============================================================
===============================================================
cam3_2_17
Originator: erik ( Erik Kluzek)
Date: Tue Aug 30 16:32:07 MDT 2005
Model: CAM
Version: CAM3.2.17
One-line Summary: Resolve special physics modes beneath driver, move surface restart to seperate call
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Ran test_driver.shl script: yes (does all of the following tests)
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail: ideal-phys bit-for-bit test
029 bl335 TBL.sh f4dh idphys 9 ....................................FAIL! rc= 7
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, eaton, mvertens
Restart files change: no
Changes answers: no (bit-for-bit) (Except FV ideal_phys test, because was previously broken)
Changes to CLM land-model: Add noland mode for adibatic,ideal_phys, aqua
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Resolve the non-physics modes of CAM (adiabatic, ideal_phys, and aquaplanet)
below the driver level, so that components know the appropriate action, but
the driver isn't aware of it. Continue to use the common block for the data
that signals this behavior, but move the behavior completely inside of ice,
ocean or land surface models.
Also move the read of surface initial data and restart information outside
of cam_intial and read_restart to methods just before surface model
initialization. This means breaking up the restart so that intht is called
later, the read_history is called later, and the close of nrg is done at this
later step as well. Move the nrg unit numbers to wholy contained within the
relevent restart modules.
This actually allows CCSM mode to make use of ideal_phys, or adiabatic,
unlike previously where it wasn't a possiblity.
Documentation fully spelled out at:
Pre-ESMF Stage-1.2.3 checkin from
http://swiki.ucar.edu/start/82
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_2_16
Originator: jedwards ( James Edwards)
Date: Mon Aug 29 20:40:09 MDT 2005
Model: CAM
Version: CAM3.2.16
One-line Summary: added 'only' clause to some and removed other module statements
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
(one-line description of changes: Makefile, include files, directories, etc.)
(Detailed description below)
Substantial timing or memory changes: no (describe)
Requires change in run script: no
(one-line description of changes: data files, run-scripts, namelist etc.)
(Detailed description below)
Ran test_driver.sh script: yes (does all of the following tests)
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail: none (Get clearance for failed tests, enter lines from td.*.status files that fail)
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
(Note: test_driver.sh only does checks this -- if you run it with "env BL_ROOT=")
Changes to CLM land-model: None (one-line description)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
(If not abort this tag and follow the procedures in)
http://www.cgd.ucar.edu/~cam/cam_checkins/checkin_process.shtml
Changes Made:
minor changes were made to a number of files which use modules from
the dynamics directories.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_2_15
Originator: erik ( Erik Kluzek)
Date: Mon Aug 22 11:36:09 MDT 2005
Model: CAM
Version: CAM3.2.15
One-line Summary: Split comsrf into three parts move calls to surface models up to cam.F90 and stepon.F90 level
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Ran test_driver.sh script: yes (does all of the following tests)
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail: none (Get clearance for failed tests, enter lines from td.*.status files that fail)
License problems with scamtest
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, mvertens, eaton
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: Pass in new structures to it
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Put cam3_2_14_brnchT_chghilev_2 on the main trunk (with some SCAM changes made so
that CRM should operate properly). Didn't test for anything beyond testscript.
Changes defined in pre-ESMF Stage-1-2
http://swiki.ucar.edu/start/73
Make physpkg into a module with and initial and 2 run phase methods.
Move the initialization calls of the surface models to cam.F90.
Move the run calls of the surface models to stepon.F90.
Divide comsrf into 3 parts:
comsrf.F90 ------- miscellaneous 2D arrays that should be moved into relevant packages (physics,
ice, or ocean)
camhub_comp.F90 --- Responsible for merging surface models (land, ice, ocean) into a merged
surface state
camsrfexch_types -- definition of derived data types for exchange of surface information
between surface models
Move landfrac, ocnfrac and icefrac into merged surface derived type.
Move instantiation of surface exchange types to top driver level. Pass down data types
to relevant subroutines.
Make physpkg into a module with and initial and 2 run phase methods.
Move the initialization calls of the surface models to cam.F90.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_2_14
Originator: boville (Byron Boville), eaton (Brian Eaton)
Date: Fri Aug 12 16:12:13 MDT 2005
Model: CAM
Version: CAM3.2.14
One-line Summary: history module mods
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Ran test_driver.shl script: yes
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail: none
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: boville, eaton
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Some cleanup of the history module - mainly moving addfld and add_default
calls to the modules that are responsible for defining a quantity and
making outfld calls for it. This gets rid of most of bldfld and h_default.
What remains of those subroutines has been merged into bldfld, and bldfld
has been moved into history_defaults.F90 which we expect to eventually go
away. Also, the SCAM specific routines have been put into
history_scam.F90. What's left in the history module is intended to be the
generic history module interface.
The functionality of the history module has been extended so that the
defined history files don't have to be sequential, e.g., a user can now
remove fincl2 from a namelist that contains specifications for fincl2 and
fincl3 without having to rename fincl3 to fincl2. Previously the renaming
would have been necessary because having the 2nd history file empty when
the 3rd file was not empty was not allowed. However, this new
functionality was not motivated by increased user convenience which is
pretty negligible. The real reason for this change was to allow more
flexibility in where the calls are made to add fields to the initial file
which is just a special history file which is always the last one in the
list. By allowing the list to contain empty files, the index of initial
file can now be terminal list index which is known at compile time, rather
than being the index that follows the last user or code defined history
file which isn't known until after processing all the add_default calls
which can occur at any point during the physics initialization.
precc_thresh and precl_thresh have been removed from the namelist.
===============================================================
===============================================================
cam3_2_13
Originator: mvr ( Mathew Rothstein)
Date: Thu Aug 11 12:29:05 MDT 2005
Model: CAM
Version: CAM3.2.13
One-line Summary: Added tests to default CAM test suite; various bug fixes and cleanup
cam-bugs Requests resolved: none
Requires change in build system: yes
Makefile changed for pathf90; added -ftpp option to handle directive
embedded in multi-line data statement
Substantial timing or memory changes: no
Requires change in run script: no
Ran test_driver.shl script: yes (does all of the following tests)
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail: none
Tested to work coupled with CCSM: yes
Tested to work with SCAM: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
(If not abort this tag and follow the procedures in)
http://www.cgd.ucar.edu/~cam/cam_checkins/checkin_process.shtml
Changes made:
R models/atm/cam/test/system/create_ccsmcam
- most stuff was merged into TCB.ccsm.sh
A models/atm/cam/bld/run-lightning.csh
- new self-contained example run script specifically for lightning platform
A models/atm/cam/test/system/TMC.sh
- new script for the testing of mass conservation parameters
A models/atm/cam/test/system/TSB.ccsm.sh
- new generic script that will submit a specified ccsm test and include a
baseline test if appropriate
M models/atm/cam/bld/Makefile
- added pathf90 compiler option -ftpp, needed to pre-process a directive
embedded in a multi-line data statement
M models/atm/cam/src/control/history.F90
- bug fix for the setting of time_bnds for branch runs
- bug fix for the writes of flag_xyfill and flag_isccplev to the restart file
M models/atm/cam/src/control/time_manager.F90
- added method get_prev_time (analogous to get_curr_time)
M models/atm/cam/src/physics/cam1/qneg3.F90
- made mods recommened by dave parks for NEC efficiency
M models/atm/cam/test/system/TBL.sh
- will now skip entirely if not testing vs baseline
- bug fix for mkdir validation
M models/atm/cam/test/system/TBR.sh
- reordering of algorithm for efficiency
- bug fix for mkdir validation
M models/atm/cam/test/system/TCB.ccsm.sh
- complete overhaul; is now just a wrapper script for ccsm's create_test with
some sed commands
M models/atm/cam/test/system/TCB.sh
- bug fix for mkdir validation
M models/atm/cam/test/system/TCS.ccsm.sh
- major overhaul to work with new script TSB.ccsm.sh
M models/atm/cam/test/system/TEQ.sh
- bug fix for mkdir validation
M models/atm/cam/test/system/TER.sh
- reordering of algorithm for efficiency
- bug fix for mkdir validation
M models/atm/cam/test/system/TRX.sh
- reordering of algorithm for efficiency
- bug fix for mkdir validation
M models/atm/cam/test/system/TSM.sh
- bug fix for mkdir validation
M models/atm/cam/test/system/input_tests_aix
- replaced TCB.ccsm.sh with TSB.ccsm.sh in default list
M models/atm/cam/test/system/input_tests_linux
- added the tests for mass conservation to defaul list
M models/atm/cam/test/system/input_tests_master
- added mass conservatin tests, ccsm submit test, most smoke tests now 9 steps
M models/atm/cam/test/system/scamtest.sh
- bug fix for mkdir validation
M models/atm/cam/test/system/test_driver.sh
- now using newcprnc - required to recognize isccp variables
M models/atm/cam/test/system/nl_files/ghgrmp
- added namelist variable to turn on isccp simulator for testing
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_2_12
Originator: erik ( Erik Kluzek)
Date: Wed Aug 10 12:53:31 MDT 2005
Model: CAM
Version: CAM3.2.12
One-line Summary: Get rid of #ifdefs refering to CAMSURF from cam3.2.3 commit
cam-bugs Requests resolved: none
Requires change in build system: no (but you can get rid of references to CAMSURF in misc.h)
Substantial timing or memory changes: no
Requires change in run script: no
Ran test_driver.shl script: yes (does all of the following tests)
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail: Most bit-for-bit tests fail as this is an answer changing commit
On bangkok:
004 bl111 TBL.sh e8c8mdm ttrac 9 ..................................FAIL! rc= 7
007 bl112 TBL.sh e8sdm ghgrmp 9 ...................................FAIL! rc= 7
011 bl153 TBL.sh e64m outfrq24h -2 ................................FAIL! rc= 7
015 bl311 TBL.sh f10c8mdm ttrac 9 .................................FAIL! rc= 7
022 bl312 TBL.sh f10sdm ghgrmp 9 ..................................FAIL! rc= 7
026 bl353 TBL.sh f2m outfrq24h -2 .................................FAIL! rc= 7
030 bl511 TBL.sh s8c8mdm ttrac 9 ..................................FAIL! rc= 7
033 bl512 TBL.sh s8sdm ghgrmp 9 ...................................FAIL! rc= 7
037 bl553 TBL.sh s64m outfrq24h -2 ................................FAIL! rc= 7
On bluesky:
005 bl131 TBL.sh e32c11dh pghgsul 9 ...............................FAIL! rc= 7
008 bl132 TBL.sh e32sdh ghgrmp 9 ..................................FAIL! rc= 7
015 bl151 TBL.sh e64h outfrq24h -2 ................................FAIL! rc= 7
019 bl331 TBL.sh f4c11dh pghgsul 9 ................................FAIL! rc= 7
022 bl332 TBL.sh f4sdh ghgrmp 9 ...................................FAIL! rc= 7
030 bl336 TBL.sh wm4h outfrq3s 9..................................FAIL! rc= 7
034 bl351 TBL.sh f2h outfrq24h -2 .................................FAIL! rc= 7
037 bl354 TBL.sh fm2dh outfrq3s 9 .................................FAIL! rc= 7
040 bl355 TBL.sh fmo2dh off2x2.5 9 ................................FAIL! rc= 7
044 bl531 TBL.sh s32c11dh pghgsul 9 ...............................FAIL! rc= 7
047 bl532 TBL.sh s32sdh ghgrmp 9 ..................................FAIL! rc= 7
054 bl551 TBL.sh s64h outfrq24h -2 ................................FAIL! rc= 7
On tempest:
004 bl131 TBL.sh e32c11dh pghgsul 9 ...............................FAIL! rc= 7
007 bl132 TBL.sh e32sdh ghgrmp 9 ..................................FAIL! rc= 7
014 bl152 TBL.sh e64o outfrq24h -2................................FAIL! rc= 7
017 bl314 TBL.sh wg10dh outfrq3s 9 ................................FAIL! rc= 7
021 bl331 TBL.sh f4c11dh pghgsul 9 ................................FAIL! rc= 7
024 bl332 TBL.sh f4sdh ghgrmp 9 ...................................FAIL! rc= 7
031 bl352 TBL.sh f2o outfrq24h -2 .................................FAIL! rc= 7
035 bl531 TBL.sh s32c11dh pghgsul 9 ...............................FAIL! rc= 7
038 bl532 TBL.sh s32sdh ghgrmp 9 ..................................FAIL! rc= 7
045 bl552 TBL.sh s64o outfrq24h -2 ................................FAIL! rc= 7
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, mvertens, eaton
Restart files change: no
Changes answers: Yes (same-physics)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Get rid of all the #ifdef's refering to CAMSURF that went in with
the cam3.2.3 tag.
Basically, this tag gets the same answers as cam3.2.11 if you set
#define CAMSURF CLIMATE_CAMSURF
in misc.h and in configure script.
The exception to this is that in this tag newiceproperties is
still called at initialization
in camice in camice_init. The reason for this is because there is
some initializion
that's important to do in newiceproperties.
The code changes answers compared to cam3.2.11, and the changes
were validated with
simulations with FV and DOM at 4x5 resolution. Answers also change
for SOM, but this
wasn't verified.
/ERIK/csm/cam324_fv4x5dom_camsurf_1 --- Experiment based
off cam3.2.4 (10 years)
vs.
/ERIK/csm/cam321_fv4x5dom_1 -- control of cam3.2.1 (10
years)
Plots are available from
http://www.cgd.ucar.edu/~erik/AMWG_diag/cam324_fv4x5dom_camsurf_1-cam321_fv4x5dom_1/
The difference in the code is primarily to unify the merging of
surface fields at
timestep==0 with that done while time-stepping. And to unify how
the ocean-ice
diagnostic fluxes are calculated at time-step==0 with when
time-stepping. Also in
order to remove a dependency on the order that ocean and ice are
called, I also use
1-landfrac rather than ocnfrac as a mask for certain
calculations. This changes
answers for SOM at each time-step.
The tests that fail have to do with bit-for-bit comparisions with
cam3.2.11. Some
of the bit-for-bit tests pass, if they run ideal-phys, adiabatic,
or aqua-planet.
Also error-growth tests are identical as error-growth is run in
aqua-planet mode.
See:
http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_2_11
Originator: sawyer (Will Sawyer), eaton (Brian Eaton)
Date: Fri Aug 5 11:47:16 MDT 2005
Model: CAM
Version: CAM3.2.11
One-line Summary: Cleanup and performance mods for FV
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: Yes (performance improvements for FV on Cray and NEC SX)
Requires change in run script: no
Ran test_driver.shl script: yes (does all of the following tests)
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail:
on bluesky:
030 bl336 TBL.sh wm4h outfrq3s 9 ..................................FAIL! rc= 7
034 bl351 TBL.sh f2h outfrq24h -2 .................................FAIL! rc= 7
on bangkok:
026 bl353 TBL.sh f2m outfrq24h -2 .................................FAIL! rc= 7
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: sawyer, eaton
Restart files change: no
Changes answers: no, except roundoff introduce into FV
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Replace the original dynamics_vars.F90 by the new version.
Add performance mods for both Cray X1 and NEC SX (from Art Mirin and Pat Worley).
Clean up implementation of options for using platform optimized FFT
libraries (from Art Mirin and Pat Worley).
Remove SAVEd variables in cd_core. This will come close to making the code
reentrant.
Explain failed tests:
These tests all failed due to a roundoff level changed introduced in the FV dycore. Since no
tests that had debug enabled failed, we assume this is an optimization issue. The roundoff level
changes were verified by perturbation growth tests on both bluesky and bangkok, but only for
the configuration of tests bl351 and bl353. Test bl336 is a waccm_mozart configuration and
the pergro test does not currently work there.
===============================================================
===============================================================
cam3_2_10
Originator: sawyer (Will Sawyer), eaton ( Brian Eaton)
Date: Tue Aug 2 18:02:15 MDT 2005
Model: CAM
Version: CAM3.2.10
One-line Summary: Bug fix for FV in OMP-only mode
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Ran test_driver.shl script: yes (does all of the following tests)
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail: none
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: sawyer, eaton
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
dynamics/fv/p_d_adjust.F90
. add "pole" to private variables in first OMP directive
===============================================================
===============================================================
cam3_2_9
Originator: dcn (David Noone), eaton ( Brian Eaton)
Date: Tue Jul 26 17:57:28 MDT 2005
Model: CAM
Version: CAM3.2.9
One-line Summary: mods for water isotopes
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Ran test_driver.shl script: yes
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail: none (Get clearance for failed tests, enter lines from td.*.status files that fail)
on tempest:
031 bl352 TBL.sh f2o outfrq24h -2 .................................FAIL! rc= 7
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: dcn, eaton
Restart files change: no
Changes answers: no
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
This is an incremental commit of mods to incorporate water isotopes into
CAM. Added mods to manage the water isotope constituents. Added mods to
allow the cloud sedimentation and ocean evaporation processes to act on
these constituents.
The regression test that fails is due to a bug in the FV dycore introduced
in cam3_2_8. It has only been observed in pure OMP mode on the SGI.
===============================================================
===============================================================
cam3_2_8
Originator: sawyer, eaton ( Brian Eaton)
Date: Thu Jul 21 13:37:42 MDT 2005
Model: CAM
Version: CAM3.2.8
One-line Summary: update FV dycore, part 1
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no (only tested IBM)
Requires change in run script: no
Ran test_driver.sh script: yes
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail:
On tempest:
031 bl352 TBL.sh f2o outfrq24h -2 .................................FAIL! rc= 7
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: spectral dycores are bit-for-bit, fv has roundoff change
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Update the FV dycore to contain new code from the GEOS5 version. The new
code includes derived types to encapsulate grid and state information.
Currently we've only used the grid type. The dynamics state remains module
data in the prognostics module. The new grid type replaces most of the
previous uses of the pmgrid and spmd_dyn modules although there is still
cleanup work to be done. There was also a change in the polar filtering
code which introduces a roundoff level change in the FV simulations. The
change was implemented as a bug fix to the problem that answers were
depending on the X decomposition when the 2D decomp option was used.
===============================================================
===============================================================
cam3_2_7
Originator: mvr ( Mathew Rothstein)
Date: Tue Jul 19 11:31:59 MDT 2005
Model: CAM
Version: CAM3.2.7
One-line Summary: Fix for CCSM testing within CAM test suite; other test suite enhancements, cleanup
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Ran test_driver.shl script: yes (does all of the following tests)
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail: none
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, eaton
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
R models/atm/cam/bld/CAM.pm
R models/atm/cam/bld/CAM_lab.pm
R models/atm/cam/bld/CAM_run.pm
R models/atm/cam/bld/Design.tex
R models/atm/cam/bld/Requirements.tex
R models/atm/cam/bld/bldtex.csh
R models/atm/cam/bld/condense-path.pl
R models/atm/cam/bld/fvtest.sh
R models/atm/cam/bld/graphgrowth.csh
R models/atm/cam/bld/lab_default.pm
R models/atm/cam/bld/run-model.pl
- cleaned up obsolete files from bld directory
R models/atm/cam/test/system/TER.ccsm.sh
- script made more generic and renamed - see the new file TCS.ccsm.sh
A models/atm/cam/test/system/TCS.ccsm.sh
- generic script to check the status of a coupled cam run
M models/atm/cam/test/system/TBL.sh
- status file renamed for consistency, algorithm re-ordered for efficiency
M models/atm/cam/test/system/TBR.sh
- status file renamed for consistency, algorithm re-ordered for efficiency
M models/atm/cam/test/system/TCB.ccsm.sh
- will now configure build and submit ccsm run for specified test/config
M models/atm/cam/test/system/TCB.sh
- status file renamed for consistency
M models/atm/cam/test/system/TEQ.sh
- status file renamed for consistency
M models/atm/cam/test/system/TER.sh
- status file renamed for consistency, algorithm re-ordered for efficiency
M models/atm/cam/test/system/TRX.sh
- status file renamed for consistency, algorithm re-ordered for efficiency
M models/atm/cam/test/system/TSM.sh
- status file renamed for consistency
M models/atm/cam/test/system/create_ccsmcam
- modified to use CAM sandbox for coupled test rather than CAM of CCSM tag
M models/atm/cam/test/system/input_tests_aix
- new testid for ccsm check status test
M models/atm/cam/test/system/input_tests_master
- new argument lists for ccsm tests (test, resolution, component set)
M models/atm/cam/test/system/scamtest.sh
- status file renamed for consistency
M models/atm/cam/test/system/test_driver.sh
- now allows batch jobs to run from their own directory if env var CAMROOT set
M models/atm/cam/test/system/CAM_runcmnd.sh
- corrected documentation in the comments
M models/atm/cam/tools/scam/testscript
- removed unnecessary output file
M models/atm/cam/test/system/nl_files/adia
M models/atm/cam/test/system/nl_files/aqpgro
M models/atm/cam/test/system/nl_files/ghgrmp
M models/atm/cam/test/system/nl_files/idphys
M models/atm/cam/test/system/nl_files/no_ttrac
M models/atm/cam/test/system/nl_files/off2x2.5
M models/atm/cam/test/system/nl_files/outfrq24h
M models/atm/cam/test/system/nl_files/outfrq3s
M models/atm/cam/test/system/nl_files/ttrac
M models/atm/cam/test/system/nl_files/ttrac_lb1
M models/atm/cam/test/system/nl_files/ttrac_lb2
M models/atm/cam/test/system/nl_files/ttrac_lb3
- namelists now specify secondary history tape names, freqs, and use 64-bit
M models/atm/cam/bld/run-ibm.csh
M models/atm/cam/bld/run-pc.csh
M models/atm/cam/bld/run-sgi.csh
- updated the self-contained example run scripts
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_2_6
Originator: fvitt ( Francis Vitt)
Date: Wed Jul 13 09:34:52 MDT 2005
Model: CAM
Version: CAM3.2.6
One-line Summary: Fixed bug in earth-sun distance factor used in waccm
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
(one-line description of changes: Makefile, include files, directories, etc.)
(Detailed description below)
Substantial timing or memory changes: no
Requires change in run script: no
(one-line description of changes: data files, run-scripts, namelist etc.)
(Detailed description below)
Ran test_driver.shl script: yes (does all of the following tests)
Machines tested: IBM, SGI, Linux-Lahey
bluesky, tempest, bangkok
Test id's that fail:
on bluesky -- bug fix in waccm resulted in failure of base line comparison
030 bl336 TBL.sh wm4h outfrq3s 9 ..................................FAIL! rc= 6
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Stacy Walters
Restart files change: no
Changes answers: no (bit-for-bit) or Yes
bit-for-bit except for waccm
(Note: test_driver.sh only does checks this -- if you run it with "env BL_ROOT=")
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
(If not abort this tag and follow the procedures in)
http://www.cgd.ucar.edu/~cam/cam_checkins/checkin_process.shtml
Changes made:
models/atm/cam/src/chemistry/waccm_mozart/mo_gas_phase_chemdr.F90
models/atm/cam/src/chemistry/waccm_mozart/mo_waccm_hrates.F90
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_2_5
Originator: mvr ( Mathew Rothstein)
Date: Mon Jul 11 15:56:16 MDT 2005
Model: CAM
Version: CAM3.2.5
One-line Summary: Removal of params.h and misc.h (clm code excluded)
cam-bugs Requests resolved: none
Requires change in build system: yes
#defines of misc.h and params.h now passed to Makefile via cppdefs
Substantial timing or memory changes: no
Requires change in run script: no
Ran test_driver.shl script: yes (does all of the following tests)
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail:
cb999, er999 (ccsm test on bluesky)
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, eaton, mvertens
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
(If not abort this tag and follow the procedures in)
http://www.cgd.ucar.edu/~cam/cam_checkins/checkin_process.shtml
Changes made:
M models/atm/cam/bld/configure
- moved #define's of params.h and misc.h into cppdefs
- removed creation of params.h (misc.h still temporarily remains for clm code)
M models/atm/cam/bld/Makefile
- no longer looks for SPMD specification in misc.h
**also modified all fortran and c files to remove #include of misc.h and
params.h (excluding clm code)
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_2_4
Originator: jeff ( Yen-Huei Lee)
Date: Sun Jul 3 13:48:55 MDT 2005
Model: CAM
Version: CAM3.2.4
One-line Summary: fix tro_mozart to allow extra tracers, bug fix to run in ccsm, output co2 fluxes to history tape, add index for fdms
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
(one-line description of changes: Makefile, include files, directories, etc.)
(Detailed description below)
Substantial timing or memory changes: no or Yes (describe)
Requires change in run script: no
(one-line description of changes: data files, run-scripts, namelist etc.)
(Detailed description below)
Ran test_driver.shl script: yes (does all of the following tests)
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail: none (Get clearance for failed tests, enter lines from td.*.status files that fail)
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit) or Yes (same-to-roundoff, same-physics, new-physics, new-climate)
(Note: test_driver.sh only does checks this -- if you run it with "env BL_ROOT=")
Changes to CLM land-model: None (one-line description)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
(If not abort this tag and follow the procedures in)
http://www.cgd.ucar.edu/~cam/cam_checkins/checkin_process.shtml
Changes made:
M models/atm/cam/src/chemistry/trop_mozart/chemistry.F90
allow trop_mozart to run with extra tracers, i.e. CO2.
M models/atm/cam/src/physics/cam1/physpkg.F90
bug fix to allow ccsm to run
M models/atm/cam/src/physics/cam1/co2_cycle.F90
M models/atm/cam/src/control/ccsm_msg.F90
output co2 fluxes to history tape
M models/atm/cam/src/physics/cam1/comsrf.F90
add index for fdms
M models/atm/cam/src/dynamics/fv/fv_prints.F90
bug fix
M models/atm/cam/bld/DefaultCAMEXPNamelist.xml
use new data sad_file for waccm_mozart
===============================================================
===============================================================
cam3_2_3
Originator: erik ( Erik Kluzek)
Date: Thu Jun 30 00:18:39 MDT 2005
Model: CAM
Version: CAM3.2.3
One-line Summary: Move cam3_2_2_brnchT_srfsimp6 to cam_dev -- makes surface (ocean, land, sea-ice) more modular
Requires change in build system: yes
(Requires CAMSURF defines in misc.h)
Substantial timing or memory changes: no
Requires change in run script: no
Ran test_driver.shl script: yes (does all of the following tests)
Machines tested: IBM, SGI, Linux-Lahey
Test id's that fail: Bluesky: 1, 5,8,15,22,30,34,37,40,44,47,54,55
5-54 fail because diagnostic fields change
001 cb999 TCB.ccsm.sh .............................................FAIL! rc= 4
055 er999 TER.ccsm.sh .............................................FAIL! rc= 2
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, eaton, mvertens
Restart files change: yes
Changes answers: no (bit-for-bit) (some ocean/ice diagnostic fields
change to roundoff at time=0)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Add following to misc.h:
#define B4B_CAMSURF -2
#define RND_CAMSURF -1
#define CLIMATE_CAMSURF 0
#define CAMSURF B4B_CAMSURF
This allows answers to be the same as current model, while allowing
ifdefs that can provide changes to roundoff or climate-changing.
Move tsnam and plevmx to ice_constants
Shorten initext and remove tangled dependencies of surface models.
Create initialization methods for sea-ice and oceanm that are called by
initext. Move surface methods in physpkg to run methods in ocean, and ice.
Make camice and somocn, and camocn into modules with various methods.
Move ice_read from inidat to camice. Move write_restart, read_restart
of ice to camice methods.
Make comsrf more modular. Move data specific to ocean, ice, or land
out of comsrf. Create merge method, and make some methods private.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_2_2
Originator: mvr ( Mathew Rothstein) jeff, mvertens, eaton
Date: Wed Jun 15 16:37:19 MDT 2005
Model: CAM
Version: CAM3.2.2
One-line Summary: configure/build-namelist mods for calls from ccsm and a plethora of bug fixes
cam-bugs Requests resolved: none
Requires change in build system: yes - now enabled for ccsm
Substantial timing or memory changes: no
Requires change in run script: yes - now enabled for ccsm
Tested to work coupled with CCSM (create_ccsmcam): not create_ccsmcam, but other
Tested to work with SCAM (tools/scam/testscript): yes
Machines tested: IBM, SGI, Linux-Lahey,pgi
Ran test_driver.shl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself,eaton
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: yes - logical restart writes changed to ints
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
R models/atm/cam/tools/icesst/bcgen/namelist.T42
R models/atm/cam/tools/icesst/bcgen/namelist.T42.oldtt
R models/atm/cam/tools/icesst/bcgen/namelist.T5
- cleanup of files resulting from regeneration of sst datasets
A models/atm/cam/tools/icesst/bcgen/namelist
- new file required of modified method to generate sst datasets
A models/atm/cam/test/system/config_files/wm4h
- new configuration options file for waccm test without debug
M models/atm/cam/bld/CAM_namelist.pm
M models/atm/cam/bld/build-namelist
M models/atm/cam/bld/namelist.pm
- mods required for CAM's build-namelist script to be called from within CCSM
M models/atm/cam/bld/DefaultCAMEXPNamelist.xml
M models/atm/cam/bld/DefaultCLMEXPNamelist.xml
- new defaults for 1.9x2.5 topo datasets
M models/atm/cam/bld/configure
- mods required for CAM's configure script to be called from within CCSM
M models/atm/cam/src/chemistry/waccm_mozart/mo_setrxt.F90
- bug fix to achieve b4b restart of waccm_mozart
M models/atm/cam/src/control/ccsm_msg.F90
M models/atm/cam/src/control/history.F90
M models/atm/cam/src/control/restart.F90
M models/atm/cam/src/control/time_manager.F90
M models/lnd/clm2/src/main/histFileMod.F90
- all logical restart writes are now integers (workaround to pathscale bug)
M models/atm/cam/src/physics/cam1/co2_cycle.F90
M models/atm/cam/src/physics/cam1/radiation.F90
- change co2 tracer from moist to dry
M models/atm/cam/src/control/ccsm_msg.F90
- change unit of co2 flux from kgC/m2/s to kgCO2/m2/s
M models/atm/cam/src/physics/cam1/srfxfer.F90
- bug fix for computation of netsw in coupled mode
M models/atm/cam/test/system/input_tests_master
- modified to now test waccm_mozart in non-debug mode
M models/atm/cam/test/system/scamtest.sh
- bug fix for output of test results
M models/atm/cam/tools/definesurf/fmain.f90
- land fractions < .001 set to zero; metadata now complies with CF conventions
M models/atm/cam/tools/icesst/README
- describe -g option for reading grid info from file
M models/atm/cam/tools/icesst/bcgen/bcgen.f90
- pass the history attribute to setup_outfile
M models/atm/cam/tools/icesst/bcgen/driver.f90
- input & output filepaths now specified on command-line
- generate the history attribute
M models/atm/cam/tools/icesst/bcgen/setup_outfile.f90
- add history attribute to both output files
M models/atm/cam/tools/icesst/regrid/regrid.f90
- Modify so that output grid coordinates can be read from the new grid files
M models/atm/cam/tools/icesst/regrid/wrap_nf.f90
- add wrap_nf_put_var_int
M models/atm/cam/tools/scam/ui/crm.cpp
- updated aerosol optics file to be used for column radiation model test
M models/atm/cam/tools/scam/userdata/crmtest26.out
- updated validation file for column radiation model test
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_2_1
Originator: aconley ( Andrew J. Conley)
Date: Fri Jun 3 17:29:46 MDT 2005
Model: CAM
Version: CAM3.2.1
One-line Summary: Improve interface between radiation (LW and SW) and aerosols
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
(one-line description of changes: Makefile, include files, directories, etc.)
(Detailed description below)
Substantial timing or memory changes: Yes (problems for vector architecture)
Requires change in run script: Yes
Added new data to AerosolsOptics file. So CAM namelist has changed
Tested to work coupled with CCSM (create_ccsmcam): No: ccsm namelist is now out of date
Tested to work with SCAM (tools/scam/testscript): No: crm part of test fails now because of change to Aerosol Optics
Machines tested: IBM, SGI, Lahey
Ran test_driver.shl script: No: ran test_driver.sh
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: not for waccm
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, eaton
Restart files change: no
Changes answers: Yes (no change to climate, but larger than roundoff changes)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
New files: aerosol_radiation_interface.F90
aerosol_index.F90
Deleted: volcrad.F90
Changed: DefaultCAMExpNamelist
runtime_opts.F90
advnce.F90
aer_optics.F90
aerosol_intr.F90
inti.F90
prescribed_aerosols.F90
radae.F90
radiation.F90
radsw.F90
volcanicmass.F90
===============================================================
===============================================================
cam3_2_0
Originator: mvertens ( Mariana Vertenstein)
Date: Mon May 30 15:53:48 MDT 2005
Model: CAM
Version: CAM3.2.0
One-line Summary: incorporated clm3_expa_29 into stand-alone cam
cam-bugs Requests resolved: none
Requires change in build system: yes (new clm cpp variable added)
Substantial timing or memory changes: no (unless CN code is activated)
Requires change in run script: all clm surface datasets have changed
(Detailed description below)
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Machines tested: IBM, SGI, Linux-Lahey
Ran test_driver.shl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself,
Restart files change: no
Changes answers: new-climate
Changes to CLM land-model: Yes (see below)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
The following code changes accompany the incorporation
of new CLM code - clm_expa_29 - into the stand-alone CAM
development code base. As part of these changes, new CLM
raw datasets were used to create the clm model surface
datasets. All CLM surface datasets were regenerated using
the new raw datasets. In addition, clm3_expa_29 also contains the
the incorporation of the carbon-nitrogen cycle into the CLM code base
(both Peter Thornton's CN model and the CASA carbon cycle model).
A detailed summary of climate changing modifications appears below.
Finally, clm3_expa_29, contains the flexibility, for the first time,
to run dynamic land use runs using prescribed land use datasets.
Note that generating clm surface datasets is now accomplished
via an offline tool in mksrfdata/tools/mksrfdat. This is not
part of the standard cam clm/tools directory yet
due to the current structure of the cam cvs module.
This will be incorporated into cam in a subsequent cam checkin.
Validation simulations were carried out for 1979-2000 with annually varying SSTs.
The tags used for hyd_clm3_1 were cam3_1_7 and clm3_expa_29
(camroot = /fis/cgd/cseg/people/mvertens/src/cam/cam3_1_7_clm3_expa_29)
The changes between hyd_clm3_1 and hyd_con were:
o New SST dataset (amipbc_sst_T42_1949_2004.nc as opposed to sst_HadOIBl_bc_64x128_1949_2001_c020411.nc)
o Albedo calculation in CLM moved to end of second biophysics loop
o Peter Thornton's 2-leaf canopy model
o Peter Lawrence's MODIS-based surface dataset (distribution of plant
functional types, LAI, and soil color)
o David Lawrence's and Peter Thornton's hydrology modifications (includes
new formulations for interception, infiltration (surface runoff), soil
moisture effects on stomatal conductance, soil to canopy air space transfer
coefficients, soil hydraulic conductivity, and baseflow.
===============================================================
===============================================================
cam3_1_9
Originator: boville, eaton
Date: Fri May 27 09:55:54 MDT 2005
Model: CAM
Version: CAM3.1.9
One-line Summary: WACCM bugfix
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Machines tested: IBM, SGI, Linux-Lahey
Ran test_driver.shl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: boville
Restart files change: no
Changes answers: no (bit-for-bit) except WACCM_MOZART changes answers
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
chemistry/waccm_mozart/chemistry.F90
. Fix initialization of qmin.
===============================================================
===============================================================
cam3_1_8
Originator: boville, eaton ( Brian Eaton)
Date: Tue May 17 18:33:16 MDT 2005
Model: CAM
Version: CAM3.1.8
One-line Summary: Merge latest WACCM development code
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Machines tested: IBM, SGI, Linux-Lahey
Ran test_driver.shl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: boville, eaton
Restart files change: no
Changes answers: CAM is bit-for-bit, WACCM has a new climate.
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Merge latest WACCM development code.
There is a new feature in the FV tracer code (trac2d.F90) which allows the
timestep splitting to differ in different vertical layers. This feature is
currently only enabled for the WACCM_MOZART configuration since it can
introduce answer changes into the standard CAM FV configurations.
Add LAPACK_LIBDIR macro to Makefile to allow specifying the location of the
LAPACK and BLAS libraries. Optional use is made of these libraries in
WACCM specific code (on AIX only).
Add the use of the massv library to the AIX section of the Makefile. The
use of this library introduces roundoff level differences into the
simulation when changing the parallel configuration. This library is only
used in WACCM_MOZART specific code.
In the IRIX64 section of the Makefile the -cpp option was replaced with
-macro_expand. A WACCM_MOZART ifdef which was including code in a
multi-line data statement was failing to compile otherwise.
===============================================================
===============================================================
cam3_1_7
Originator: mvr ( Mathew Rothstein)
Date: Wed May 4 12:14:23 MDT 2005
Model: CAM
Version: CAM3.1.7
One-line Summary: Updating the new cam test framework with new tests and other enhancements
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Machines tested: IBM, SGI, Linux-Lahey
Ran test_driver.shl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, eaton
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None (one-line description)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
A models/atm/cam/test/system/TEQ.sh
- new generic script to test the equivalence of two cam smoke runs
A models/atm/cam/test/system/TRX.sh
- new script to test the restart of cam using fv 2d decomposition
A models/atm/cam/test/system/scamtest.sh
- new script to test scam and crm
A models/atm/cam/test/system/config_files/f10dm
A models/atm/cam/test/system/config_files/fm2dh
A models/atm/cam/test/system/config_files/fmo2dh
A models/atm/cam/test/system/config_files/wg10dh
A models/atm/cam/test/system/config_files/wm4dh
- new configuration options files for added tests
A models/atm/cam/test/system/nl_files/no_ttrac
A models/atm/cam/test/system/nl_files/off2x2.5
A models/atm/cam/test/system/nl_files/ttrac_lb1
A models/atm/cam/test/system/nl_files/ttrac_lb2
A models/atm/cam/test/system/nl_files/ttrac_lb3
- new namelist options files for added tests
M models/atm/cam/test/system/CAM_compare.sh
M models/atm/cam/test/system/TBL.sh
M models/atm/cam/test/system/TBR.sh
M models/atm/cam/test/system/TCB.ccsm.sh
M models/atm/cam/test/system/TCB.sh
M models/atm/cam/test/system/TER.ccsm.sh
M models/atm/cam/test/system/TER.sh
M models/atm/cam/test/system/TSM.sh
M models/atm/cam/test/system/input_tests_aix
M models/atm/cam/test/system/input_tests_irix
M models/atm/cam/test/system/input_tests_linux
M models/atm/cam/test/system/input_tests_master
M models/atm/cam/test/system/test_driver.sh
mods and enhancements to the new cam test framework scripts including:
- added waccm tests
- added test for load balancing
- added test for the fv offline driver
- added test for fv 2d decomposition
- moved scam/crm test into default tests
- will now stop on first failure for interactive runs by default and
added environment variable CAM_SOFF to override default behavior
- cleaned up handling of shell string variables
- now backs up directories of failed tests when reattempting
M README
- updated with the wording used for release
(Put your detailed description of the changes you made here...)
(You will get your edited copy e-mailed to you. It will also be)
(sent to cam-dev@cgd.ucar.edu.)
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_1_6
Originator: jeff ( Yen-Huei Lee)
Date: Sat Apr 30 12:25:17 MDT 2005
Model: CAM
Version: CAM3.1.6
One-line Summary: allow model year to be different from ghg data year
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
(one-line description of changes: Makefile, include files, directories, etc.)
(Detailed description below)
Substantial timing or memory changes: no or Yes (describe)
Requires change in run script: no
(one-line description of changes: data files, run-scripts, namelist etc.)
(Detailed description below)
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Machines tested: IBM, SGI, Linux-Lahey
Ran test_driver.shl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: Brian Eaton
Restart files change: no
Changes answers: no (bit-for-bit)
(Note: test_driver.sh only does checks this -- if you run it with "env BL_ROOT=")
Changes to CLM land-model: None (one-line description)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
(If not abort this tag and follow the procedures in)
http://www.cgd.ucar.edu/~cam/cam_checkins/checkin_process.shtml
Changes made:
M models/atm/cam/src/control/ccsm_msg.F90
M models/atm/cam/src/control/runtime_opts.F90
M models/atm/cam/src/chemistry/waccm_mozart/chem_surfvals.F90
M models/atm/cam/src/physics/cam1/chem_surfvals.F90
M models/atm/cam/src/physics/cam1/advnce.F90
M models/atm/cam/src/physics/cam1/srfxfer.F90
M models/atm/cam/src/physics/cam1/radiation.F90
M models/atm/cam/src/physics/cam1/co2_cycle.F90
D models/atm/cam/src/physics/cam1/co2_data_scalar.F90
Function add:
use the following namelists to set model start year and ghg data start year,
ghg_yearStart_model = 1
ghg_yearStart_data = 1950
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_1_5
Originator: eaton ( Brian Eaton)
Date: Mon Apr 25 18:17:11 MDT 2005
Model: CAM
Version: CAM3.1.5
One-line Summary: fix CCSM test, update definesurf tool
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): no
Machines tested: IBM
Ran test_driver.shl script: Only ran CCSM test
Tested on fv dynamics: no
Tested on eul dynamics: no
Tested on sld dynamics: no
Tested that restarts are bit-for-bit: no
Tested multiple constituents: no
Tested that different domain decompositions match bit-for-bit: no
Tested in adiabatic mode: no
Code reviewed by: myself
Restart files change: no
Changes answers: no
Changes to CLM land-model: None (one-line description)
Changes made:
Fix the CCSM test.
Also add some modifications to the definesurf tool from Byron. These mods
relax the previous restriction that the same filter be applied to both SGH
and PHIS.
===============================================================
===============================================================
cam3_1_4
Originator: jeff ( Yen-Huei Lee)
Date: Sat Apr 23 11:20:36 MDT 2005
Model: CAM
Version: CAM3.1.4
One-line Summary: add co2 tracer transport and using 2D co2 for radiation
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
(one-line description of changes: Makefile, include files, directories, etc.)
(Detailed description below)
Substantial timing or memory changes: no
Requires change in run script: no
(one-line description of changes: data files, run-scripts, namelist etc.)
(Detailed description below)
Tested to work coupled with CCSM (create_ccsmcam): no
Tested to work with SCAM (tools/scam/testscript): yes
Machines tested: IBM, SGI, Linux-Lahey
Ran test_driver.shl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: Brian Eaton
Restart files change: no
Changes answers: Yes (same-to-roundoff)
(Note: test_driver.sh only does checks this -- if you run it with "env BL_ROOT=")
Changes to CLM land-model: None (one-line description)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
(If not abort this tag and follow the procedures in)
http://www.cgd.ucar.edu/~cam/cam_checkins/checkin_process.shtml
Changes made:
The constant co2 value used for radiation computation was replaced by
2D co2 distribution (column-mean). This is responsible for the roundoff.
Function added:
co2 tracer can be transported with co2 fluxes from land (through coupler),
from ocean (through data file) and from fossil fuel (through data file).
co2 scalar data can be read from data file and can be used as data-cycling
type operation.
M models/atm/cam/src/control/cam.F90
M models/atm/cam/src/control/initext.F90
M models/atm/cam/src/control/ccsm_msg.F90
M models/atm/cam/src/control/restart.F90
M models/atm/cam/src/control/runtime_opts.F90
M models/atm/cam/src/dynamics/fv/inidat.F90
M models/atm/cam/src/dynamics/sld/inidat.F90
M models/atm/cam/src/dynamics/eul/inidat.F90
M models/atm/cam/src/physics/cam1/advnce.F90
M models/atm/cam/src/physics/cam1/initindx.F90
M models/atm/cam/src/physics/cam1/restart_physics.F90
M models/atm/cam/src/physics/cam1/comsrf.F90
M models/atm/cam/src/physics/cam1/inti.F90
M models/atm/cam/src/physics/cam1/srfxfer.F90
M models/atm/cam/src/physics/cam1/rad_constituents.F90
M models/atm/cam/src/physics/cam1/radiation.F90
M models/atm/cam/src/physics/cam1/radsw.F90
M models/atm/cam/src/physics/cam1/radlw.F90
M models/atm/cam/src/physics/cam1/radae.F90
A models/atm/cam/src/physics/cam1/co2_cycle.F90
A models/atm/cam/src/physics/cam1/co2_data_flux.F90
A models/atm/cam/src/physics/cam1/co2_data_scalar.F90
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_1_3
Originator: eaton ( Brian Eaton)
Date: Thu Mar 31 08:51:32 MST 2005
Model: CAM
Version: CAM3.1.3
One-line Summary: move control of rad calcs into rad module. pathscale & g95 builds
cam-bugs Requests resolved: bugzilla IDs 4, 10, 13
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM (create_ccsmcam): no, currently broken
Tested to work with SCAM (tools/scam/testscript): yes
Machines tested: IBM, SGI, Linux-Lahey
Ran test-model.pl script: no
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Consolidate control of radiation calculations in the radiation module.
This includes resposibility for maintaining the namelist variables iradsw,
iradlw, iradae, and irad_always, and providing a function for the values
dosw, dolw, doabsems, and aeres. These variables have all been removed
from comctl.h.
Add bugfix for turbulent mountain stress code.
Add files to DefaultCAMEXPNamelist.xml needed for WACCM runs at 1.9x2.5
resolution.
Add Makefile sections and ESMF build to support pathscale and g95 compilers
on linux systems.
Replace the PGF90 CPP macro with NO_R16. The PGF90 macro was being used to
indicate that the compiler doesn't support quad precision. But that's also
the case for the pathscale and g95 compilers. The use of a generic CPP
token allows new compilers to be supported by modifying the the Makefile to
specify the compiler properties without requiring source code modification.
The g95 compiler doesn't recognize the cray pointer syntax in fv/geopk.F90.
The use of cray pointers there can't be trivially replaced by F90 pointers.
Since the cray pointers only appear in the geopk16 routine which is not
used by default, a CPP token NO_CRAY_POINTERS is defined for g95 and used
to stub out geopk16. endrun is called if the user sets the namelist
variable (geopktrans) to use geopk16 when NO_CRAY_POINTERS is defined.
Add bugfix to cmfmca pointed out by David Noone. This only affects the
simulation of tracers using dry mixing ratio (none by default). The set of
test tracers enabled by setting the namelist variable
tracers_flag=.true. use a dry mixing ratio and are affected by this fix.
===============================================================
===============================================================
cam3_1_2
Originator: mvr ( Mathew Rothstein)
Date: Fri Mar 25 16:41:11 MST 2005
Model: CAM
Version: CAM3.1.2
One-line Summary: New testing framework for CAM - replaces test-model; and a few minor bug fixes
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Machines tested: IBM, SGI, Linux-Lahey, Linux-PGI
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
...and much, much more!
Code reviewed by: myself, eaton
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
(If not abort this tag and follow the procedures in)
http://www.cgd.ucar.edu/~cam/cam_checkins/checkin_process.shtml
Changes made:
R models/atm/cam/test/system/babyblue.test-model.pl.log
R models/atm/cam/test/system/CAM_test.pm
R models/atm/cam/test/system/cam_timing.pm
R models/atm/cam/test/system/check-timing.pl
R models/atm/cam/test/system/dao_batch.csh
R models/atm/cam/test/system/default_tests.test-model.xml
R models/atm/cam/test/system/llnl_blue.csh
R models/atm/cam/test/system/llnl_compass.csh
R models/atm/cam/test/system/llnl_frost.csh
R models/atm/cam/test/system/namelist-config.test-model.dbg.log
R models/atm/cam/test/system/ncar_batch.csh
R models/atm/cam/test/system/nersc_batch.csh
R models/atm/cam/test/system/ornl_batch.csh
R models/atm/cam/test/system/specs-testslist.test-model.xml
R models/atm/cam/test/system/test-examples.pl
R models/atm/cam/test/system/test-model.pl
R models/atm/cam/test/system/test-production.pl
- cleanup of old test-model files
A models/atm/cam/test/system/CAM_compare.sh
- utility script for comparison of results
A models/atm/cam/test/system/CAM_runcmnd.sh
- utility script for constructing platform dependent run command
A models/atm/cam/test/system/input_tests_aix
- required aix tests for cam commit
A models/atm/cam/test/system/input_tests_irix
- required irix tests for cam commit
A models/atm/cam/test/system/input_tests_linux
- required linux tests for cam commit
A models/atm/cam/test/system/input_tests_master
- master list of tests
A models/atm/cam/test/system/TBL.sh
- generic script for baseline tests
A models/atm/cam/test/system/TBR.sh
- generic script for branch tests
A models/atm/cam/test/system/TCB.ccsm.sh
- script for configure and build test of cam in coupled model
A models/atm/cam/test/system/TCB.sh
- generic script for configure and build tests
A models/atm/cam/test/system/TER.ccsm.sh
- script for exact restart test of cam in coupled model
A models/atm/cam/test/system/TER.sh
- generic script for exact restart tests
A models/atm/cam/test/system/test_driver.sh
- driver script for all tests
A models/atm/cam/test/system/TSM.sh
- generic script for smoke tests
A models/atm/cam/test/system/nl_files/adia
A models/atm/cam/test/system/nl_files/aqpgro
A models/atm/cam/test/system/nl_files/ghgrmp
A models/atm/cam/test/system/nl_files/idphys
A models/atm/cam/test/system/nl_files/outfrq24h
A models/atm/cam/test/system/nl_files/outfrq3s
A models/atm/cam/test/system/nl_files/pghgsul
A models/atm/cam/test/system/nl_files/ttrac
- files containing non-default namelist options being tested
A models/atm/cam/test/system/config_files/e32c11dh
A models/atm/cam/test/system/config_files/e32dh
A models/atm/cam/test/system/config_files/e32pdh
A models/atm/cam/test/system/config_files/e32sdh
A models/atm/cam/test/system/config_files/e64h
A models/atm/cam/test/system/config_files/e64m
A models/atm/cam/test/system/config_files/e64o
A models/atm/cam/test/system/config_files/e8c8mdm
A models/atm/cam/test/system/config_files/e8pdm
A models/atm/cam/test/system/config_files/e8sdm
A models/atm/cam/test/system/config_files/f10c8mdm
A models/atm/cam/test/system/config_files/f10pdm
A models/atm/cam/test/system/config_files/f10sdm
A models/atm/cam/test/system/config_files/f2h
A models/atm/cam/test/system/config_files/f2m
A models/atm/cam/test/system/config_files/f2o
A models/atm/cam/test/system/config_files/f4c11dh
A models/atm/cam/test/system/config_files/f4dh
A models/atm/cam/test/system/config_files/f4pdh
A models/atm/cam/test/system/config_files/f4sdh
A models/atm/cam/test/system/config_files/s32c11dh
A models/atm/cam/test/system/config_files/s32dh
A models/atm/cam/test/system/config_files/s32pdh
A models/atm/cam/test/system/config_files/s32sdh
A models/atm/cam/test/system/config_files/s64h
A models/atm/cam/test/system/config_files/s64m
A models/atm/cam/test/system/config_files/s64o
A models/atm/cam/test/system/config_files/s8c8mdm
A models/atm/cam/test/system/config_files/s8pdm
A models/atm/cam/test/system/config_files/s8sdm
- files containing configuration options being tested
M models/atm/cam/src/physics/cam1/sulfur_intr.F90
- fixed declaration and setting of array totcond
M models/atm/cam/src/physics/cam1/dmsbnd.F90
- initialized arrays dmsin and dms to 0._r8
M models/atm/cam/src/dynamics/eul/inidat.F90
- added explicit setting of phis_hires
M models/atm/cam/src/dynamics/fv/inidat.F90
- modified the array assignments where it depended upon ncol
M models/atm/cam/src/dynamics/sld/inidat.F90
- added explicit setting of phis_hires
M models/atm/cam/src/control/initext.F90
- modified the array assignments where it depended upon ncol
M models/atm/cam/bld/DefaultCAMEXPNamelist.xml
- modified the default settings for fv 4x5 sulpher datasets
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_1_1
Originator: fvitt ( Francis Vitt)
Date: Mon Mar 21 12:39:28 MST 2005
Model: CAM
Version: CAM3.1.1
One-line Summary: Added a tropospheric chemistry package.
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
(one-line description of changes: Makefile, include files, directories, etc.)
(Detailed description below)
Substantial timing or memory changes: no
Requires change in run script: no
(one-line description of changes: data files, run-scripts, namelist etc.)
(Detailed description below)
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Machines tested: IBM, SGI, Linux-Lahey
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: Brian Eaton
Restart files change: no
Changes answers: no (bit-for-bit) or Yes (same-to-roundoff, same-physics, new-physics, new-climate)
(Note: test-model.pl only does checks this -- if you run it with the "-compare" option)
Changes to CLM land-model: None (one-line description)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
(If not abort this tag and follow the procedures in)
http://www.cgd.ucar.edu/~cam/cam_checkins/checkin_process.shtml
Changes made:
Files added:
models/atm/cam/bld/config_trop_chem_mozart_defaults.xml
models/atm/cam/src/chemistry/trop_mozart/chem_mods.F90
models/atm/cam/src/chemistry/trop_mozart/chemistry.F90
models/atm/cam/src/chemistry/trop_mozart/m_het_id.F90
models/atm/cam/src/chemistry/trop_mozart/m_rxt_id.F90
models/atm/cam/src/chemistry/trop_mozart/m_spc_id.F90
models/atm/cam/src/chemistry/trop_mozart/mo_adjrxt.F90
models/atm/cam/src/chemistry/trop_mozart/mo_aerosols.F90
models/atm/cam/src/chemistry/trop_mozart/mo_airplane.F90
models/atm/cam/src/chemistry/trop_mozart/mo_ch4_lbc.F90
models/atm/cam/src/chemistry/trop_mozart/mo_chemini.F90
models/atm/cam/src/chemistry/trop_mozart/mo_drydep.F90
models/atm/cam/src/chemistry/trop_mozart/mo_exp_sol.F90
models/atm/cam/src/chemistry/trop_mozart/mo_gas_phase_chemdr.F90
models/atm/cam/src/chemistry/trop_mozart/mo_imp_sol.F90
models/atm/cam/src/chemistry/trop_mozart/mo_indprd.F90
models/atm/cam/src/chemistry/trop_mozart/mo_jlong.F90
models/atm/cam/src/chemistry/trop_mozart/mo_jpl.F90
models/atm/cam/src/chemistry/trop_mozart/mo_jshort.F90
models/atm/cam/src/chemistry/trop_mozart/mo_lb_vals.F90
models/atm/cam/src/chemistry/trop_mozart/mo_lightning.F90
models/atm/cam/src/chemistry/trop_mozart/mo_lin_matrix.F90
models/atm/cam/src/chemistry/trop_mozart/mo_lu_factor.F90
models/atm/cam/src/chemistry/trop_mozart/mo_lu_solve.F90
models/atm/cam/src/chemistry/trop_mozart/mo_mass_xforms.F90
models/atm/cam/src/chemistry/trop_mozart/mo_mean_mass.F90
models/atm/cam/src/chemistry/trop_mozart/mo_negtrc.F90
models/atm/cam/src/chemistry/trop_mozart/mo_nln_matrix.F90
models/atm/cam/src/chemistry/trop_mozart/mo_prod_loss.F90
models/atm/cam/src/chemistry/trop_mozart/mo_read_chm_sim.F90
models/atm/cam/src/chemistry/trop_mozart/mo_regrider.F90
models/atm/cam/src/chemistry/trop_mozart/mo_seasalt.F90
models/atm/cam/src/chemistry/trop_mozart/mo_setext.F90
models/atm/cam/src/chemistry/trop_mozart/mo_setinv.F90
models/atm/cam/src/chemistry/trop_mozart/mo_setrxt.F90
models/atm/cam/src/chemistry/trop_mozart/mo_setsoa.F90
models/atm/cam/src/chemistry/trop_mozart/mo_setsox.F90
models/atm/cam/src/chemistry/trop_mozart/mo_srf_emissions.F90
models/atm/cam/src/chemistry/trop_mozart/mo_sulf.F90
models/atm/cam/src/chemistry/trop_mozart/mo_tracname.F90
models/atm/cam/src/chemistry/trop_mozart/mo_ub_vals.F90
models/atm/cam/src/chemistry/trop_mozart/mo_usrrxt.F90
models/atm/cam/src/chemistry/trop_mozart/mo_waccm_photo.F90
models/atm/cam/src/chemistry/trop_mozart/mo_wetdep.F90
models/atm/cam/src/chemistry/trop_mozart/phtadj.F90
models/atm/cam/src/chemistry/trop_mozart/rxtmod.F90
Files modified:
models/atm/cam/bld/DefaultCAMEXPNamelist.xml
- added files needed for trop_mozart chemistry package
models/atm/cam/bld/camexp.pm
- added trop_mozart input files to the namelist
models/atm/cam/bld/configure
- added trop_mozart chemistry package option
models/atm/cam/src/chemistry/waccm_mozart/chemistry.F90
- changed interface to be consistent with trop_mozart chemistry
models/atm/cam/src/control/history.F90
- increased tbl_hash_oflow_percent
models/atm/cam/src/control/runtime_opts.F90
- added input files to namelist for trop_mozart chemistry
models/atm/cam/src/physics/cam1/aerosol_intr.F90
- added aerosol feedback capability for trop_mozart chemistry
models/atm/cam/src/physics/cam1/chemistry.F90
- changed interface to be consistent with trop_mozart chemistry
models/atm/cam/src/physics/cam1/physpkg.F90
- enabled lightning production of NO for trop_mozart chemistry
models/atm/cam/src/physics/cam1/stratiform.F90
- added total precip/evap fields to physics buffer needed by trop_mozart
models/atm/cam/src/physics/waccm/chemistry.F90
- changed interface to be consistent with trop_mozart chemistry
models/atm/cam/tools/scam/testscript
- removed "-test" option from configure invocation
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_1
Originator: eaton ( Brian Eaton)
Date: Tue Mar 15 20:14:48 MST 2005
Model: CAM
Version: CAM3.1
One-line Summary: cam3_1 same as cam3_0_34
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM (create_ccsmcam): no
Tested to work with SCAM (tools/scam/testscript): no
Machines tested: none
Ran test-model.pl script: no
Tested on fv dynamics: no
Tested on eul dynamics: no
Tested on sld dynamics: no
Tested that restarts are bit-for-bit: no
Tested multiple constituents: no
Tested that different domain decompositions match bit-for-bit: no
Tested in adiabatic mode: no
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? no
Changes made: none.
Add tag cam3_1. This is the branch point for the CAM3.1 release
===============================================================
===============================================================
cam3_0_34
Originator: pworley ( Patrick H Worley)
Date: Tue Mar 15 18:01:47 MST 2005
Model: CAM
Version: CAM3.0.34
One-line Summary: Cray X1 optimizations
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: Yes on X1: recovers performance lost due to changes since cam3.0
Requires change in run script: no
Tested to work coupled with CCSM (create_ccsmcam): no - broken in earlier check-in
Tested to work with SCAM (tools/scam/testscript): yes
Machines tested: IBM, SGI, Linux-Lahey
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: Yes - added support for SSP execution mode
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
added models/utils/esmf/cray_x1_ssp/
Identical to cray_x1 except that compiler options in the file base_variables were modified
so that esmf routines could be used in SSP mode.
models/atm/cam/bld/Makefile
- in UNICOSMP section: added support for SSP mode and higher optimization levels for 4 routines,
recovering performance lost in earlier check-in that lowered optimization
models/atm/cam/src/control/history.F90, wrap_nf.F90
- deleted CRAY ifdefs
models/atm/cam/src/physics/cam1/convect_deep.F90
- modified calls to convtran, replacing ":" array indices with "1", to improve performance
models/atm/cam/src/physics/cam1/qneg3.F90
- modified to improve vectorization and streaming
models/atm/cam/src/physics/cam1/radiation.F90
- modified calls to radcswmx and radclwmx,
replacing ":" array indices with "1", to improve performance
- added Cray compiler directives
models/atm/cam/src/physics/cam1/zm_conv.F90
- added Cray compiler directives
models/lnd/clm2/src/main/controlMod.F90
- added ifdef to define default clump_pproc value for SSP mode
models/lnd/clm2/src/main/driver.F90
- added ifdef to disable OpenMP directive around "loop2" for SSP mode (a bug work around)
models/utils/timing/gptl.c
- added ifdef to not disable timers unnecessarily in SSP mode
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_0_33
Originator: rosinski ( Jim Rosinski)
Date: Thu Mar 10 14:05:47 MST 2005
Model: CAM
Version: CAM3.0.33
One-line Summary: Enable new timing lib.
cam-bugs Requests resolved: none
Requires change in build system: Maybe for CCSM.
Substantial timing or memory changes: Yes: on IBM timer overhead decreases
by as much as 10x. On SGI by 2x. On
Cray X1 by 3x. No speedup on Linux.
Requires change in run script: no
Tested to work coupled with CCSM (create_ccsmcam): yes but it failed. Told
to go ahead with commit.
Tested to work with SCAM (tools/scam/testscript): yes
Machines tested: IBM, SGI, Linux-Lahey
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: timer init calls and fortran include file changed.
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
o Replace timer code in utils/timing with new more effecient version. Better
performance results from replacing a linked list search for timer names
with a simple hash table. The new library now prints per-timer overhead
estimates. It can also interface with the PAPI low level performance
counter library. A brief description of how to do so is included in
utils/timing/README.
o The old interface names (t_xxx) were retained. But these should be changed
to the new names (gptlxxx) as soon as possible.
o Modify CAM Makefile for optimal performance on Cray X1.
===============================================================
===============================================================
cam3_0_32
Originator: eaton ( Brian Eaton)
Date: Tue Mar 8 10:08:10 MST 2005
Model: CAM
Version: CAM3.0.32
One-line Summary: topo fields in separate file
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: yes
namelist includes bnd_topo -- build-namelist provides defaults
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Machines tested: IBM, SGI, Linux-Lahey
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Topography fields (mods from Byron Boville)
===========================================
These mods are to read the topo fields (PHIS, SGH, SGH30, LANDFRAC,
LANDM_COSLAT) from their own file. They are no longer written to the
output initial files. Backwards compatibility is maintained by looking for
the topo fields on the initial file if a topo file is not specified using
the new namelist variable bnd_topo. There are also mods to write the names
of the initial, topo, and sst files into global attributes of the history
files.
An initial set of topo files have been created by extracting the topo
fields from the initial files currently specified in
DefaultCAMEXPNamelist.xml. The topo files are in the data directory
$CSMDATA/atm/cam/topo/. These files will eventually be replaced by a set
of standard files that contain detailed metadata describing how the fields
were produced.
There are also modifications and cleanup in the definehires and definesurf
utilities. Added alot of information to the README files for both utils.
Misc
====
. Re-insert a fix in phys_grid.F90 (transpose_chunk_to_block)
changing the chunk_buffer arg to intent(in).
. In history.F90 increase pflds to 2000 and tbl_hash_oflow_percent to 10.
===============================================================
===============================================================
cam3_0_31
Originator: bundy ( Dani Bundy Coleman)
Date: Wed Mar 2 13:56:13 MST 2005
Model: CAM
Version: CAM3.0.31
One-line Summary: Add mass conservation for dry tracers in Eularian dycore; restore same in SLD
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM (create_ccsmcam): yes (bluesky)
Tested to work with SCAM (tools/scam/testscript): yes (bangkok)
Machines tested: IBM (bluesky), SGI (tempest), Linux-Lahey (bangkok)
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Brian Eaton
Restart files change: no (pdeld added only when run with dry-type constituents)
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
A models/atm/cam/src/dynamics/eul/massfix.F90
Currently a module that just holds hw1, hw2, hw3, alpha. Could facilitate
a separation between mass fix and the timefilter.
M slt/qmassa.F90
Now a module, old subroutine qmassa is now qmassarun. Added an
optional argument pdeld to be used to calculate hw1lat for dry
mixing-ratio constituents.
M control/history.F90
moved addfld and add_defualt for PDELDRY from here to physics/cam1/diagnostics.F90
M eul/dp_coupling.F90
Added transfer of pdeld from p->d,d->p. Calculation of pdeld moved to
tfilt_massfix.F90 instead of call to set_state_pdry.
M eul/inital.F90
added initialization of dynamics variable pdeld
M eul/prognostics.F
added dynamics variable pdeld, only allocated when dry constituents
are present
M eul/restart_dynamics.F90
added pdeld to restart files
M eul/scan2.F90
moved public hw1, hw2, hw3, alpha to new module massfix.F90
pass pdeld to tfilt_massfixrun()
M eul/scandyn.F90
pass pdeld to qmassarun
M eul/scanslt.F90
pass pdeld to qmassarun
M eul/stepon.F90
pass pdeld to d_p_coupling
M eul/tfilt_massfix.F90
Calculate time filtered value of pdeldry; use this to time-filter dry
constituents conservatively.
M sld/scanslt.F90
change qmassa call to new name: qmassarun
M sld/tfilt_massfix.F90
save water vapor for use in next timestep; this was inadvertently
removed in cam3_0_28 and is necessary for dry mass conservation
M physics/constituents.F90
new public logical cnst_need_pdeldry. Default is flase, set to true if
any constituents are dry.
M physics/convect_deep.F90
only use pdeldry when needed
M physics/diagnostics.F90
addfld and outfld calls for PSDRY and PDELDRY only when needed
M physics/icarus_scops.F90
added r8 to reals
M physics/physics_types.F90
added state%pdeldry and related to methods only when needed
added optional argument pdeld_calc (default=true) to subroutine
set_state_pdry, setting false still calculates related fields but
doesn't change state%pdeldry itself.
M physics/physpkg.F90
diagnostic subroutine gavglook only calculates dry quantities when
needed
M physics/tracers_suite.F90
added missing r8
Now a module, old subroutine qmassa is now qmassarun. Added an
optional argument pdeld to be used to calculate hw1lat for dry
mixing-ratio constituents.
M control/history.F90
moved addfld and add_defualt for PDELDRY from here to physics/cam1/diagnostics.F90
M eul/dp_coupling.F90
Added transfer of pdeld from p->d,d->p. Calculation of pdeld moved to
tfilt_massfix.F90 instead of call to set_state_pdry.
M eul/inital.F90
added initialization of dynamics variable pdeld
M eul/prognostics.F
added dynamics variable pdeld, only allocated when dry constituents
are present
M eul/restart_dynamics.F90
added pdeld to restart files
M eul/scan2.F90
moved public hw1, hw2, hw3, alpha to new module massfix.F90
pass pdeld to tfilt_massfixrun()
M eul/scandyn.F90
pass pdeld to qmassarun
M eul/scanslt.F90
pass pdeld to qmassarun
M eul/stepon.F90
pass pdeld to d_p_coupling
M eul/tfilt_massfix.F90
Calculate time filtered value of pdeldry; use this to time-filter dry
constituents conservatively.
M sld/scanslt.F90
change qmassa call to new name: qmassarun
M sld/tfilt_massfix.F90
save water vapor for use in next timestep; this was inadvertently
removed in cam3_0_28 and is necessary for dry mass conservation
M physics/constituents.F90
new public logical cnst_need_pdeldry. Default is flase, set to true if
any constituents are dry.
M physics/convect_deep.F90
only use pdeldry when needed
M physics/diagnostics.F90
addfld and outfld calls for PSDRY and PDELDRY only when needed
M physics/icarus_scops.F90
added r8 to reals
M physics/physics_types.F90
added state%pdeldry and related to methods only when needed
added optional argument pdeld_calc (default=true) to subroutine
set_state_pdry, setting false still calculates related fields but
doesn't change state%pdeldry itself.
M physics/physpkg.F90
diagnostic subroutine gavglook only calculates dry quantities when
needed
M physics/tracers_suite.F90
added missing r8
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_0_30
Originator: mvr ( Mathew Rothstein)
Date: Wed Feb 9 16:34:12 MST 2005
Model: CAM
Version: CAM3.0.30
One-line Summary: Cleanup of copyright info; minor bug fixes
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Machines tested: IBM, SGI, Linux-Lahey
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None (one-line description)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
R LICENSE
- update to the copyright info and renaming of the file for
consistency with ccsm
A Copyright
- new file to replace LICENSE file that was removed
M models/atm/cam/src/control/cam.F90
- slight mod to the info on where to find copyright notice
M models/atm/cam/src/control/history.F90
- bug fix for the variable "vid" whose intent needed to be out rather than in
M models/atm/cam/src/physics/cam1/icarus_scops.F90
- removed copyright info already contained in Copyright file
M models/atm/cam/test/system/create_ccsmcam
- replaced blackforest with bluesky as default machine
M models/atm/cam/test/system/ncar_batch.csh
- default load-leveller queue now for bluesky rather than blackforest
M models/atm/cam/tools/scam/testscript
- bug fix for how the TESTSCRIPTDIR was determined
M models/atm/cam/tools/scam/ui/Platform.Notes
- removed copyright info already contained in Copyright file
M models/atm/cam/tools/scam/ui/c.h
- removed copyright info already contained in Copyright file
M models/atm/cam/tools/scam/ui/ncarg/c.h
- removed copyright info already contained in Copyright file
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_0_29
Originator: rosinski ( Jim Rosinski)
Date: Sun Feb 6 12:50:13 MST 2005
Model: CAM
Version: CAM3.0.29
One-line Summary: Reduced grid mods for CAM and interpaerosols boundary dataset generation code.
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Machines tested: IBM, SGI, Linux-Lahey
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Brian Eaton
Restart files change: no
Changes answers: no (bit-for-bit)
(Note: test-model.pl only does checks this -- if you run it with the "-compare" option)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
o Modify eul/initcom.F90 to abort in SCAM mode if reduced grid enabled.
o Enable "mkrgrid" utility to work under lf95, ifort, and Cray X1.
o Enable "interpaerosols" utility for reduced grid, and Cray X1.
o Add new utility "mkrgridnew", which can replace "mkrgrid" when IC files
are xyz. The new utility is simpler, and unlike mkrgrid works on ISCCP
history files and on Cray X1.
o Check for reduced grid in optional boundary dataset reading routines
for which reduced grid capability is not yet enabled (e.g. dust module)
and abort in reduced grid mode.
o When CAM reads SST and aerosol data, ensure that boundary dataset grids
match cam grid.
o Delete threading directives from ccsm_msg.F90 per mvertens.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_0_28
Originator: olson ( Jerry Olson)
Date: Thu Jan 27 11:11:47 MST 2005
Model: CAM
Version: CAM3.0.28
One-line Summary: Add new diagnostics to history file. Other misc mods.
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Machines tested: IBM, SGI, Linux-Lahey
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers:
- Changes are round-off for EUL (except for aqua-planet which has a new climate
because of a change in the dry airmass constant)
Changes are BFB for SLD (except for aqua-planet which has a new climate)
Changes are BFB for FV (except for aqua-planet which has a new climate)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
- Add fields to the master field list which complete the heat, moisture, cloud
liquid water, cloud liquid ice, and precip budgets
(no new *default* fields are added to the History file)
- Add an attribute ("sampling_seq") to those fields on a NetCDF History File
which are *not* based on an every-time-step sampling frequency. The
attribute would have a descriptive value indicating its sampling sequence.
Currently only one value is added: "rad_lwsw" (for fields that are only
sampled at LW/SW radiation time steps). No code is currently being added to
fill these fields with "missing" values.
- Replace the "RAD_FIRST_HOUR" namelist variable with "IRAD_ALWAYS".
"RAD_FIRST_HOUR" was a logical that was too inflexible. "IRAD_ALWAYS" is an
integer that specifies length of time in timesteps (positive) or hours
(negative) that the SW/LW radiation will be run continuously from the start
of an initial run.
- For aqua-planet runs, dry mass is hard-wired to 101325.-245. ( = 101080.) Pa
- Bug fixes:
- Fix a potential bug in the history module that would end up clobbering
the creation of the IC file if *all* the fields added to an auxiliary
history file through "add_default" calls are then excluded via FEXCL in
the namelist.
- Fix bug that corrupts high resolution runs: promote "n" to I8 when
doing n**4 and n**3 math in hdinti, hordif1, and phcs.
- Correct the logic in the aerosol data interpolator as reported by
Jim Boyle on the CGD forum
- make "jstrt_p" private in SCAN2 and REALLOC7 loops
- bugfix for adiabatic restarts
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_0_27
Originator: mvr ( Mathew Rothstein) and mirin
Date: Thu Jan 20 16:16:39 MST 2005
Model: CAM
Version: CAM3.0.27
One-line Summary: Vectorization mods for finite volume; clarification of pgf90 optimization level
cam-bugs Requests resolved: none
Requires change in build system: yes
set pgf90 optimization explicitly to -O1; vectorization mods (see below)
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Machines tested: IBM, SGI, Linux-Lahey, Linux-pgi, cray x1
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, mirin, eaton
Restart files change: no
Changes answers: b4b except for fv on cray x1, ibm, and linux-pgi, which are
round-off (pergro test passed on ibm)
Changes to CLM land-model: None (one-line description)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
M models/atm/cam/bld/Makefile
M models/atm/cam/src/control/cam.F90
M models/atm/cam/src/control/spmdinit.F90
M models/atm/cam/src/dynamics/fv/cd_core.F90
M models/atm/cam/src/dynamics/fv/dp_coupling.F90
M models/atm/cam/src/dynamics/fv/dynamics_vars.F90
M models/atm/cam/src/dynamics/fv/dynpkg.F90
M models/atm/cam/src/dynamics/fv/fill_module.F90
M models/atm/cam/src/dynamics/fv/fv_prints.F90
M models/atm/cam/src/dynamics/fv/mapz_module.F90
M models/atm/cam/src/dynamics/fv/pft_module.F90
M models/atm/cam/src/dynamics/fv/spmd_dyn.F90
M models/atm/cam/src/dynamics/fv/stepon.F90
M models/atm/cam/src/dynamics/fv/sw_core.F90
M models/atm/cam/src/dynamics/fv/te_map.F90
M models/atm/cam/src/dynamics/fv/tp_core.F90
M models/atm/cam/src/physics/cam1/check_energy.F90
M models/atm/cam/src/physics/cam1/constituents.F90
M models/csm_share/shr/shr_msg_mod.F90
M models/utils/pilgrim/mod_comm.F90
please reference corresponding documentation in cam checkin list for details
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_0_26
Originator: mvr ( Mathew Rothstein) and eaton
Date: Tue Jan 18 13:58:51 MST 2005
Model: CAM
Version: CAM3.0.26
One-line Summary: Collection of bug fixes and enhancements for Cray X1
cam-bugs Requests resolved: none
Requires change in build system: yes
pgi fortran optimization set to O by default; Cray X1 mods; dependency cleanup
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Machines tested: IBM, SGI, Linux-Lahey
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, eaton
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: yes
removal of equivalences; bug fix for offline mode
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
M cam1/models/atm/cam/bld/Makefile
- changed default optimization of pgi fortran compiler to O from O2
- added Cray X1 mods and cleaned up some dependencies
M cam1/models/atm/cam/src/control/ccsm_msg.F90
- added mods for Cray X1
M cam1/models/atm/cam/src/control/history.F90
- removed equivalences; removed addfld for unused var; added gregorian
calendar type to output; removed some unneeded date checking
M cam1/models/atm/cam/src/control/runtime_opts.F90
- added code required with removal of equivalence in history.F90;
added phys_buffer mods
M cam1/models/atm/cam/src/dynamics/fv/dp_coupling.F90
- removed beglev from declaration of var ps
M cam1/models/atm/cam/src/dynamics/fv/fv_prints.F90
- fixed bug in write statements
M cam1/models/atm/cam/src/dynamics/fv/stepon.F90
- add call to shr_sys_flush
M cam1/models/atm/cam/src/physics/cam1/phys_buffer.F90
- added mods to physics buffer
M cam1/models/atm/cam/src/physics/cam1/prescribed_aerosols.F90
- fixed bug in date comparison
M cam1/models/atm/cam/tools/scam/configure
- fixed bug with input file location for scam (created with cam3_0_25)
M cam1/models/lnd/clm2/src/main/areaMod.F90
- bug fix for offline mode
M cam1/models/lnd/clm2/src/main/histFileMod.F90
- removal of equivalences
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_0_25
Originator: jet ( John Truesdale)
Date: Mon Jan 10 16:43:43 MST 2005
Model: CAM
Version: CAM3.0.25
One-line Summary: Updated SCAM INSTALL text and added data directory to fix bug with scam configure script.
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
(one-line description of changes: Makefile, include files, directories, etc.)
(Detailed description below)
Substantial timing or memory changes: no or Yes (describe)
Requires change in run script: no
(one-line description of changes: data files, run-scripts, namelist etc.)
(Detailed description below)
Tested to work coupled with CCSM (create_ccsmcam): no
Tested to work with SCAM (tools/scam/testscript): no
Machines tested: IBM, SGI, Linux-Lahey
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: no
Tested on eul dynamics: no
Tested on sld dynamics: no
Tested that restarts are bit-for-bit: n/a
Tested multiple constituents: n/a
Tested that different domain decompositions match bit-for-bit: n/a
Tested in adiabatic mode: no
Code reviewed by: myself
Restart files change: no
Changes answers: no
(Note: test-model.pl only does checks this -- if you run it with the "-compare" option)
Changes to CLM land-model: None (one-line description)
Have you filled out the pre-check-in documentation as required by the CAM CRB? no this is a nontesting tag, only edited text files (INSTALL and REAME) no code
modified.
Changes made:
Added words to the INSTALL file for SCAM to let the user know to set the
CSMDATA environment variable. Also added a directory under scam (/data).
The scam configure script assumes this directory exists and links all
scam data files here. The previous release didn't have this directory.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_0_24
Originator: pworley ( Patrick H Worley)
Date: Thu Jan 6 11:20:57 MST 2005
Model: CAM
Version: CAM3.0.24
One-line Summary: reduced grid support for spectral dycores and miscellaneous bug fixes
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no for full grid, yes (improved) for reduced grid
Requires change in run script: no
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Machines tested: IBM, SGI, Linux-Lahey
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: fixed reduced grid support in one routine
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Fixed support for reduced grid (Jim Rosinski):
- control/readinitial.F90
- physics/cam1/prescribed_aerosols.F90
- dynamics/eul/initcom.F90
- lnd/clm2/src/main/surfFileMod.F90
Additional reduced grid support modifications (for the X1
only?). Zeroing out unused locations in the FFT buffers is required
on the X1.
- dynamics/eul/linemsdyn.F90
- dynamics/eul/spegrd.F90
- dynamics/sld/spegrd.F90
- dynamics/sld/scanslt.F90
Added new runtime option, dyn_equi_by_col, to improve performance of
spectral dycores when using reduced grid. When set to .true., it
attempts to allocate the same number of columns to each process in the
dynamics rather than the same number of latitude lines. For a full
grid, these goals are identical. The option is on by default. However,
it is not necessarily the best choice when using a reduced grid on a
vector system. Setting it to .false. executes the old code,
equidistributing the number of latitude lines whether using a full or
a reduced grid.
1) Moved decompinit out of cam, into readinitial.F90 and restart.F90,
so that reduced grid option could be identifed before computing
dynamics decomposition
- control/cam.F90
- control/readinitial.F90
- control/restart.F90
2) Added dyn_equi_by_col runtime option
- control/runtime_opts.F90
- dynamics/eul/spmd_dyn.F90
- dynamics/sld/spmd_dyn.F90
- dynamics/fv/spmd_dyn.F90
3) Added dyn_equi_by_col domain decomposition
- dynamics/eul/spmd_dyn.F90
- dynamics/sld/spmd_dyn.F90
The SLD dycore does not have all of the modifications made to EUL in
Jan.-March, 2004 to, for example, decrease memory requirements in the
spectral domain. The following files were modified as a step toward
implementing these changes. The motivation is that including these
changes allows the SLD version of spmd_dyn.F90 to be almost identical
to that used in EUL.
- dynamics/sld/comspe.F90
- dynamics/sld/scanslt.F90
The current Co-Array Fortran logic in CAM requires that the two
communication buffers allocated in the spectral dycores be the same
size. This is currently true when the number of processors divides the
latitudes, but not otherwise. Since this may be important for other
one-sided messaging systems as well, the following files have been
modified to require that all processes allocate the same size
communcation buffers. Note that this has no performance impact in
normal CAM usage (and probably never has an impact).
- control/mpishorthand.F
- control/spmdinit.F90
- control/wrap_mpi.F90
- dynamics/eul/spmd_dyn.F90
- dynamics/sld/spmd_dyn.F90
Reinsert compiler directive fixing Cray X1 bug that was lost in
a previous check-in
- control/history.F90
The routine sort_chunks is called in create_chunks and
orders the columns assigned to a chunk.
It was added to phys_grid.F90 back when we were looking for
alternatives to the buffer copies introduced into radcswmx.F90 for
vectorization. A bug in this code was recently identified. Rather than
fix it, we have decided that sort_chunks does nothing useful even when
working correctly. Thus the routine sort_chunks has been removed.
- physics/cam1/phys_grid.F90
Miscellaneous bug fixes:
1) A typo in the existing "equidistribute by latitudes"
decomposition logic was fixed. Fortunately, this bug
was never seen in practice.
- dynamics/eul/spmd_dyn.F90
- dynamics/sld/spmd_dyn.F90
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_0_23
Originator: fvitt ( Francis Vitt)
Date: Tue Dec 21 10:13:50 MST 2004
Model: CAM
Version: CAM3.0.23
One-line Summary: Add ability to use offline data to drive finite volume dynamics core.
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: yse
Requires OFFLINE_DYN C preprocessor flag to be set.
Substantial timing or memory changes: no
Requires change in run script: yes
Namelist varibles are added. See below.
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Machines tested: IBM, SGI, Linux-Lahey
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: Brian Eaton
Restart files change: yes
Offline data file names added to restart file.
Changes answers: no (bit-for-bit)
(Note: test-model.pl only does checks this -- if you run it with the "-compare" option)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
(If not abort this tag and follow the procedures in)
http://www.cgd.ucar.edu/~cam/cam_checkins/checkin_process.shtml
Changes made:
Added files:
models/atm/cam/src/dynamics/fv/metdata.F90
models/atm/cam/src/dynamics/fv/pfixer.F90
Modified files:
models/atm/cam/src/dynamics/fv/cd_core.F90
models/atm/cam/src/dynamics/fv/dp_coupling.F90
models/atm/cam/src/dynamics/fv/dryairm.F90
models/atm/cam/src/dynamics/fv/dynpkg.F90
models/atm/cam/src/dynamics/fv/inidat.F90
models/atm/cam/src/dynamics/fv/restart_dynamics.F90
models/atm/cam/src/dynamics/fv/stepon.F90
models/atm/cam/src/dynamics/fv/sw_core.F90
models/atm/cam/src/dynamics/fv/uv3s_update.F90
models/atm/cam/src/physics/cam1/inti.F90
models/atm/cam/src/physics/cam1/physpkg.F90
models/atm/cam/src/control/runtime_opts.F90
models/atm/cam/src/control/string_utils.F90
Modifications are effective only if CAM is configured to use
offline meteorology data. To configure to use offline data
set OFFLINE_DYN C preprocessor flag.
Namelist variables added:
met_data_file The filepath of the netCDF file
containing the meteorology data
met_cell_wall_winds Set to true if the meteorology winds are
defined on model cell walls.
Default is false.
met_remove_file Set to true to have the offline data file be
removed from the local file system.
Default is false.
The offline meteorology data can be read from a sequence of files
provided the files are named stringNNN.nc, where NNN is an integer
of any number of digits. If the data is contained in a series of
files, the files need to be named sequentially, e.g., metdata001.nc,
metdata002.nc, metdata003.nc, etc.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_0_22
Originator: jeff ( Yen-Huei Lee)
Date: Fri Dec 10 10:20:33 MST 2004
Model: CAM
Version: CAM3.0.22
One-line Summary: add WACCM physics and chemistry
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
(one-line description of changes: Makefile, include files, directories, etc.)
(Detailed description below)
Substantial timing or memory changes: no or Yes (describe)
Requires change in run script: no
(one-line description of changes: data files, run-scripts, namelist etc.)
(Detailed description below)
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Machines tested: IBM, SGI, Linux-Lahey
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: Brian Eaton
Restart files change: no
Changes answers: no (bit-for-bit) or Yes (same-to-roundoff, same-physics, new-physics, new-climate)
(Note: test-model.pl only does checks this -- if you run it with the "-compare" option)
Changes to CLM land-model: None (one-line description)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
(If not abort this tag and follow the procedures in)
http://www.cgd.ucar.edu/~cam/cam_checkins/checkin_process.shtml
Changes made:
Add two new direstories for WACCM:
models/atm/cam/src/physics/waccm/
models/atm/cam/src/chemistry/waccm_mozart/
Change the following files:
A models/atm/cam/bld/config_cam_eul_defaults.xml
A models/atm/cam/bld/config_cam_fv_defaults.xml
A models/atm/cam/bld/config_cam_sld_defaults.xml
A models/atm/cam/bld/config_waccm_ghg_defaults.xml
A models/atm/cam/bld/config_waccm_mozart_defaults.xml
D models/atm/cam/bld/config_cache_defaults.xml
M models/atm/cam/bld/CAM_config.pmonfig.pm
M models/atm/cam/bld/DefaultCAMEXPNamelist.xml
M models/atm/cam/bld/DefaultCLMEXPNamelist.xml
M models/atm/cam/bld/camexp.pm
M models/atm/cam/bld/clm2exp.pm
M models/atm/cam/bld/configure
M models/atm/cam/bld/resolution_parameters.xml
M models/atm/cam/src/dynamics/eul/dp_coupling.F90
M models/atm/cam/src/dynamics/eul/inital.F90
M models/atm/cam/src/dynamics/eul/stepon.F90
M models/atm/cam/src/dynamics/sld/dp_coupling.F90
M models/atm/cam/src/dynamics/sld/inital.F90
M models/atm/cam/src/dynamics/fv/dp_coupling.F90
M models/atm/cam/src/dynamics/fv/dycore.F90
M models/atm/cam/src/dynamics/fv/dynamics_vars.F90
M models/atm/cam/src/dynamics/fv/inital.F90
D models/atm/cam/src/physics/cam1/ghg_surfvals.F90
A models/atm/cam/src/physica/cam1/chem_surfvals.F90
M models/atm/cam/src/physics/cam1/advnce.F90
M models/atm/cam/src/physics/cam1/aerosol_intr.F90
M models/atm/cam/src/physics/cam1/check_energy.F90
M models/atm/cam/src/physics/cam1/chemistry.F90
M models/atm/cam/src/physics/cam1/cloud_fraction.F90
M models/atm/cam/src/physics/cam1/constituents.F90
M models/atm/cam/src/physics/cam1/convect_deep.F90
M models/atm/cam/src/physics/cam1/convect_shallow.F90
M models/atm/cam/src/physics/cam1/diagnostics.F90
M models/atm/cam/src/physics/cam1/diffusion_solver.F90
M models/atm/cam/src/physics/cam1/ghg_defaults.F90
M models/atm/cam/src/physics/cam1/gw_drag.F90
M models/atm/cam/src/physics/cam1/initindx.F90
M models/atm/cam/src/physics/cam1/inti.F90
M models/atm/cam/src/physics/cam1/molec_diff.F90
M models/atm/cam/src/physics/cam1/physics_types.F90
M models/atm/cam/src/physics/cam1/physpkg.F90
M models/atm/cam/src/physics/cam1/qneg3.F90
M models/atm/cam/src/physics/cam1/rad_constituents.F90
M models/atm/cam/src/physics/cam1/radae.F90
M models/atm/cam/src/physics/cam1/radheat.F90
M models/atm/cam/src/physics/cam1/radiation.F90
M models/atm/cam/src/physics/cam1/radlw.F90
M models/atm/cam/src/physics/cam1/radsw.F90
M models/atm/cam/src/physics/cam1/stratiform.F90
M models/atm/cam/src/physics/cam1/sulfur_intr.F90
M models/atm/cam/src/physics/cam1/tphysac.F90
M models/atm/cam/src/physics/cam1/tphysbc.F90
M models/atm/cam/src/physics/cam1/tracers.F90
M models/atm/cam/src/physics/cam1/upper_bc.F90
M models/atm/cam/src/physics/cam1/vertical_diffusion.F90
M models/atm/cam/src/control/history.F90
M models/atm/cam/src/control/restart.F90
M models/atm/cam/src/control/runtime_opts.F90
M models/atm/cam/src/control/time_manager.F90
M models/atm/cam/tools/scam/scm_init/scam_inital.F90
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_0_21
Originator: pworley ( Patrick H Worley)
Date: Mon Nov 1 07:40:57 MST 2004
Model: CAM
Version: CAM3.0.21
One-line Summary: Communication optimizations (MPI1, MPI2, Co-Array Fortran) for spectral dycores and dp_coupling; physics chunking bug fix.
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: Yes, improves performance on Earth Simulator when using MPI-2 one-sided option; may improve performance on Cray X1
Requires change in run script: no
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): no
Machines tested: IBM, SGI, Linux-Lahey
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
1) Fixed bug in assign_chunks identified by Steve Ghan and Tim Shippert. This
bug did not affect execution in the standard modes of operation.
2) Fixed UNICOS/MP Makefile option for cam3_0_20 changes.
3) Communication optimizations:
- eliminated overlap implementations of realloc and dp_coupling
communications. All that was being overlapped was buffer copies,
which did not appear to be useful on the current target platforms.
- added "alternative" implementations of alltoall and allgather
(used in realloc routines and dp_coupling), using
1) one of a large number of MPI-2 point-to-point commands,
controlled by swap_comm_order, swap_comm_protocol, and
swap_comm_maxreq. The first two parameters have the same
meaning as before. swap_comm_maxreq controls the maximum number
of outstanding send/receive requests allowed in the
implementation of alltoall/allgather. This can be used to tune
performance or work around bugs on systems that do
not tolerate large numbers of outstanding requests well.
2) MPI-2 put commands
3) Co-Array Fortran remote writes
#2 is important for performance on the Earth Simulator. The others
represent defensive programming, for situations in which the MPI
collectives are not implemented efficiently. #2 also seems to be
fastest on the Cray X1, but I haven't not done a careful study yet.
- To support the MPI-2 implementation, replaced original
communication buffer logic
* call spmdbuf in spmdinit_dyn to allocate communication buffers;
call spmdbuf_resize in d_p_coupling or in p-d_coupling to
reallocate communication buffers is they are not large enough
with
* call spmdbuf in inital or in restart, after phys_grid_init, to
allocate communication buffers of the appropriate size the
first time, then use MPI_WINCREATE to enable remote
read/write from/to these buffers.
- For the SLD dycore, added minor modifications originally
introduced into the EUL dycore to improve vectorization. This
makes realloc5, realloc5, and realloc7 almost identical between
the EUL and SLD dycores again, and should improve vectorization
for the SLD dycore.
The new runtime options for dyn_alltoall, dyn_allgather, and
phys_alltoall are:
0: use collectives
1: use MPI-2 point-to-point, as specified by swap_comm_order,
swap_comm_protocol, swap_comm_maxreq. (swap_comm_maxreq default
is -1, meaning no limit; >0 sets the upper bound)
2: if compiled with -DMPI2, then use MPI-2 put implementation, else
use an MPI_SENDRECV implementation
3: if compiled with -DCAF, then use Co-Array remote write
implementation, else use an MPI_SENDRECV implementation
Files modified are as follows:
control/restart.F90
added call to spmdbuf
control/runtime_opts.F90
added swap_comm_maxreq parameter
control/swap_comm.F90
added swapm, swap_comm_maxreq;
removed swap1, swap2, swap3, swap1m, swap3m, do_swap1, do_swap3,
delayed_swap_recv
control/wrap_mpi.F90
added mpialltoallint, altalltoallv, mpiwincreate
eul/dp_coupling.F90, sld/dp_coupling.F90
added MPI2 window parameter to transpose call;
replaced spmdbuf_resize call with an spmdbuf_size test
eul/inital.F90, sld/inital.F90
added smpdbuf call
eul/realloc4.F90, sld/realloc4.F90
added altalltoall, removed swap routine calls
eul/realloc7.F90, sld/realloc7.F90, eul/scan2.F90, sld/scan2.F90
added altalltoall, removed swap routine calls
eul/spmd_dyn.F90, sld/spmd_dyn.F90
removed spmdbuf_resize and modified spmdbuf;
moved local_dp_map, block_buf_nrecs, and chunk_buf_nrecs from phys_grid;
added support for MPI2 and CAF.
fv/dp_coupling.F90
side effect of moving local_dp_map, block_buf_nrecs, and
chunk_buf_nrecs from phys_grid to spmd_dyn
fv/spmd_dyn.F90
moved local_dp_map, block_buf_nrecs, and chunk_buf_nrecs from phys_grid
sld/comspe.F90, scanslt.F90, spegrd.F90, trunc.F90
vectorization modifications ported from EUL
physics/cam1/phys_grid.F90
added altalltoall, removed swap routine calls
fixed assign_chunk bug.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_0_20
Originator: jmccaa ( James McCaa)
Date: Mon Oct 18 11:58:59 MDT 2004
Model: CAM
Version: CAM3.0.20
One-line Summary: Refactored vertical diffusion module.
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Machines tested: IBM, SGI, Linux-Lahey
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself and Brian Eaton
Restart files change: no
Changes answers: Yes (same-to-roundoff)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Refactored the vertical diffusion module:
- new interface names
- new molecular diffusion module
- new turbulent mountain stress module
- new eddy diffusion module
- new diffusion solver module
- removed turbulence module
Cleaned up unused variables from interfaces to convect_deep,
convect_shallow, cloud_fraction, stratiform modules.
Added a T42 30-level ic file to the CAM defaults file.
Added a function form of vqsatd to the wv_saturation module.
Assorted small fixed to scam, plus modified the scam
configure and test scripts to allow running from a working
directory outside the distribution.
Files added/deleted/modified:
M models/atm/cam/bld/DefaultCAMEXPNamelist.xml
M models/atm/cam/doc/ChangeLog
M models/atm/cam/src/physics/cam1/advnce.F90
M models/atm/cam/src/physics/cam1/cloud_fraction.F90
M models/atm/cam/src/physics/cam1/convect_deep.F90
M models/atm/cam/src/physics/cam1/convect_shallow.F90
A models/atm/cam/src/physics/cam1/diffusion_solver.F90
A models/atm/cam/src/physics/cam1/hb_diff.F90
M models/atm/cam/src/physics/cam1/initindx.F90
M models/atm/cam/src/physics/cam1/inti.F90
A models/atm/cam/src/physics/cam1/molec_diff.F90
M models/atm/cam/src/physics/cam1/physconst.F90
M models/atm/cam/src/physics/cam1/stratiform.F90
M models/atm/cam/src/physics/cam1/tphysac.F90
M models/atm/cam/src/physics/cam1/tphysbc.F90
M models/atm/cam/src/physics/cam1/tphysidl.F90
A models/atm/cam/src/physics/cam1/trb_mtn_stress.F90
R models/atm/cam/src/physics/cam1/turbulence.F90
M models/atm/cam/src/physics/cam1/upper_bc.F90
M models/atm/cam/src/physics/cam1/vertical_diffusion.F90
M models/atm/cam/src/physics/cam1/wv_saturation.F90
M models/atm/cam/tools/scam/GNUmakefile
M models/atm/cam/tools/scam/configure
M models/atm/cam/tools/scam/testscript
M models/atm/cam/tools/scam/obj/Makefile
M models/atm/cam/tools/scam/scm_init/c_outfld.c
M models/atm/cam/tools/scam/scm_init/init_model.F90
M models/atm/cam/tools/scam/scm_init/scam.c
M models/atm/cam/tools/scam/ui/GNUmakefile
M models/atm/cam/tools/scam/ui/configure
M models/atm/cam/tools/scam/ui/ncfile.h
===============================================================
===============================================================
cam3_0_19
Originator: erik ( Erik Kluzek)
Date: Thu Sep 23 11:49:07 MDT 2004
Model: CAM
Version: CAM3.0.19
One-line Summary: Isolate extended grid information to just spectral dycores and just inside of SLT advection inside SLD and Eul dycores.
cam-bugs Requests resolved: 147
(See http://wreq.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
Substantial timing or memory changes: see below
Requires change in run script: no
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Machines tested: IBM, Linux-Lahey
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, eaton, olson, mvr
Restart files change: yes (Eul and SLD)
Changes answers: bit-for-bit with Eul and FV dycores
same to roundoff for SLD dycore (on IBM and Linux/Lahey,
answers identical if you run with -debug compiler options)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Move cam3_0_18_brnchT_nohalo9 to cam_dev.
Isolate extended grid information to SLT advection to only
within Spectral dy-cores.
Use plon(regular grid) instead of plond(extended grid) except for
SLT variables, inside of relevent dynamics.
Eulerian:
scanslt becomes a module with alloc, initial, run and final phases.
Remove comslt and move data to inside of scanslt or scan2.
scanslt has a advection_state data-type to keep track of all extended
grid information. All data on the extended grid are kept inside of this
state. Dynamics data iscopied into and then out of this state for advection.
New dimension for FFT -- plondfft.
spetru has new temporary variables inside of it so it can be dimensioned
the right size for FFT (plondfft).
linemsdyn becomes a module and scan2 becomes a module
Make grdini, grdxy, sltb1, and sltini module subroutines of scanslt.
SLD:
scanslt becomes a module with: alloc, initial, run_setup, run, and
final phases. Extended grid parameters moved from pmgrid and rgrid to
inside of this module. Much of the data becomes private inside of
scanslt module. Create advection_state data type to keep track of all
extended grid data, copy data into this structure and then out of
for the rest of the dynamics
comslt removed and data moved to inside of scanslt. grmult becomes a module.
Use inf instead of nan for initialization.
Spmd_dyn data only used for bndexch moved as private data inside of scanslt.
Make bndexch, grdini, grdxy, sltlinint, slttraj, sltwgts, and trajdp
module subroutines of scanslt.
test-model.pl
Fix problem where couldn't change resolution on comand line
with -skip option (cam-bug 147).
Add -cppdefs option
Restart tests done at default compiler option rather than explicitly
set to production compiler settings. This allows the use of the "-debug"
option to override the default setting for testing where this is
useful.
SCAM:
Improve readability of testscript, send output to log files
and have the script clean up after itself better. Use plon instead
of plond.
Performance changes: (Following tests were done with earlier versions
on no-halo branch based off of cam3_0_9 for eul and
cam3_0_15 for sld test)
Eul@T42 on IBM same
Eul@T85 on IBM with 11 constituents 3% slower
Sld@T42 on IBM 2% faster
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_0_18
Originator: mvr ( Mathew Rothstein), eaton, mirin
Date: Fri Sep 17 11:01:58 MDT 2004
Model: CAM
Version: CAM3.0.18
One-line Summary: A collection of bug fixes and portability enhancements
cam-bugs Requests resolved: none
Requires change in build system: yes
Makefile changed for pgi compile defaults
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Machines tested: IBM, Linux-Lahey
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, eaton
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: yes
mss system calls now go through csm_share code
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
M models/atm/cam/bld/Makefile
- mods to pgi compiler defaults: -O2 optimization (non-debug), -Mrecursive
only in debug mode
- mod to ifort name for intel builds
M models/atm/cam/src/control/abortutils.F90
- fix for ibm jobs hanging in xl__trbk routine
M models/atm/cam/src/control/ccsm_msg.F90
- fix for ccsm with FV 2D composition
M models/atm/cam/src/control/ioFileMod.F90
- now calls csm_share code for mss system calls
M models/atm/cam/src/control/runtime_opts.F90
- fix to properly communicate the input vars kmxhdc, dif2, dif4
R models/atm/cam/src/control/system_cmd.c
- no longer needed - now using csm_share code
M models/atm/cam/src/physics/cam1/cldwat.F90
- fix for non-default vertical resolution extending top of model above 1mb
M models/atm/cam/src/physics/cam1/phys_grid.F90
- corrected routine argument from intent out to intent in
M models/atm/cam/test/system/CAM_test.pm
- bug fix for "-skip" options in test-model
M models/ice/csim4/ice_data.F90
- fixed out of bounds array reference
M models/lnd/clm2/src/main/fileutils.F90
- now call csm_share code for mss system calls
R models/lnd/clm2/src/main/system_cmd.c
- no longer needed - now using csm_share code
M models/utils/esmf/README
M models/utils/esmf/build/common
A models/utils/esmf/build/Darwin_absoft/ESMF_conf.h
A models/utils/esmf/build/Darwin_absoft/base
A models/utils/esmf/build/Darwin_absoft/base.site
A models/utils/esmf/build/Darwin_absoft/base_variables
A models/utils/esmf/build/Darwin_absoft/fix.h
A models/utils/esmf/build/Darwin_xlf/ESMF_conf.h
A models/utils/esmf/build/Darwin_xlf/base
A models/utils/esmf/build/Darwin_xlf/base.site
A models/utils/esmf/build/Darwin_xlf/base_variables
M models/utils/esmf/src/Infrastructure/Error/ESMC_Error.c
M models/utils/esmf/src/Infrastructure/TimeMgmt/ESMC_Alarm.c
M models/utils/esmf/src/Infrastructure/TimeMgmt/ESMC_Calendar.c
M models/utils/esmf/src/Infrastructure/TimeMgmt/ESMC_Date.c
M models/utils/esmf/src/Infrastructure/TimeMgmt/ESMC_TOD.c
M models/utils/esmf/src/Infrastructure/TimeMgmt/ESMC_Time.c
- upgrade to esmf_0_0_8
M models/utils/timing/t_pclstr.c
M models/utils/timing/t_stop.c
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_0_17
Originator: jeff ( Yen-Huei Lee)
Date: Mon Sep 13 14:41:57 MDT 2004
Model: CAM
Version: CAM3.0.17
One-line Summary: bug fix
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
(one-line description of changes: Makefile, include files, directories, etc.)
(Detailed description below)
Substantial timing or memory changes: no or Yes (describe)
Requires change in run script: no
(one-line description of changes: data files, run-scripts, namelist etc.)
(Detailed description below)
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Machines tested: IBM, Linux-Lahey
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit) or Yes (same-to-roundoff, same-physics, new-physics, new-climate)
(Note: test-model.pl only does checks this -- if you run it with the "-compare" option)
Changes to CLM land-model: None (one-line description)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
(If not abort this tag and follow the procedures in)
http://www.cgd.ucar.edu/~cam/cam_checkins/checkin_process.shtml
Changes made:
Add one file
A models/atm/cam/src/physics/cam1/upper_bc.F90
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_0_16
Originator: jeff ( Yen-Huei Lee)
Date: Mon Sep 13 13:22:44 MDT 2004
Model: CAM
Version: CAM3.0.16
One-line Summary: modified vertical diffusion
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
(one-line description of changes: Makefile, include files, directories, etc.)
(Detailed description below)
Substantial timing or memory changes: no or Yes (describe)
Requires change in run script: no
(one-line description of changes: data files, run-scripts, namelist etc.)
(Detailed description below)
Tested to work coupled with CCSM (create_ccsmcam): yes
Tested to work with SCAM (tools/scam/testscript): yes
Machines tested: IBM, Linux-Lahey
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit) or Yes (same-to-roundoff, same-physics, new-physics, new-climate)
(Note: test-model.pl only does checks this -- if you run it with the "-compare" option)
Changes to CLM land-model: None (one-line description)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
(If not abort this tag and follow the procedures in)
http://www.cgd.ucar.edu/~cam/cam_checkins/checkin_process.shtml
Changes made:
Modified the following files:
M models/atm/cam/src/physics/cam1/vertical_diffusion.F90
M models/atm/cam/src/physics/cam1/tphysac.F90
M models/atm/cam/src/physics/cam1/rad_constituents.F90
M models/atm/cam/src/physics/cam1/ghg_defaults.F90
M models/atm/cam/src/physics/cam1/constitueadvnce.F90
M models/atm/cam/src/physics/cam1/advnce.F90
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_0_15
Originator: eaton ( Brian Eaton)
Date: Thu Sep 2 08:14:58 MDT 2004
Model: CAM
Version: CAM3.0.15
One-line Summary: new radiation/constituents interface
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM (create_ccsmcam): yes
Machines tested: IBM, Linux-Lahey
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, boville, conley
Restart files change: no
Changes answers: no
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Create an new radiation/constituent interface. This functionality (for gas
phase constituents) previously resided in the chemistry module. It has
been reimplemented in it's own module (rad_constituents.F90) to facilitate
swapping in new chemistry code without having to duplicate the control
logic in another chemistry module. Also, the new module can be extended to
coordinate control of the aerosol constituents.
Modified the constituents module:
1) Add an optional argument to cnst_add to allow specifying that the
default outfld calls for that constituent should not be made. This
allows, for example, the chemistry module to output volume mixing
ratios rather than the default mass mixing ratios.
2) Add an optional argument to cnst_get_ind to allow it to return an error
status rather than just aborting if the requested name is not found in
the constituent array.
Add a bugfix to cldwat.F90. Some variables needed initializers. Only
caused a problem if the top of the model was above 1mb.
Create a new module (ghg_defaults.F90) for the default ghg distributions
computed by trcmix.
Added lat and lon (radians) to the physics_state data type.
Add a method to the ozone_data module to return a pointer to the
interpolated data.
Returned the chemistry module to its previous functionality of just
providing prognostic ghg distributions using a simple production/loss
chemistry.
===============================================================
===============================================================
cam3_0_14
Originator: jmccaa ( James Mccaa)
Date: Wed Aug 25 19:58:09 MDT 2004
Model: CAM
Version: CAM3.0.14
One-line Summary: Introduction of definehires utility and use of 30 second topographic variance (SGH30)
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM (create_ccsmcam): yes
Machines tested: IBM, SGI, Linux-Lahey
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: yes
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
This checkin introduces a new utility, definehires, that can be used to process
USGS 30 second topographic data and produce a 10-minute dataset suitable for
use with definesurf.
It is located in the directory:
cam1/models/atm/cam/tools/definehires/README
The following files were modified to accomodate a new input dataset
field (SGH30) that represents the variance of the 30-second
height data on the 10-minute grid. This supplements the field
SGH, which represents the variance of the 10-minute mean heights
on the model grid. Both definesurf and CAM can still use the
old datasets with no change in answers from the current code.
In the case of CAM, if SGH30 is not present, the model will use
SGH in its place (as it currently does), but SGH30 will be used
if found.
>> cam1/models/atm/cam/src/dynamics/eul/inidat.F90
>> cam1/models/atm/cam/src/dynamics/sld/inidat.F90
>> cam1/models/atm/cam/src/dynamics/fv/inidat.F90
>> cam1/models/atm/cam/src/physics/cam1/comsrf.F90
>> cam1/models/atm/cam/src/physics/cam1/diagnostics.F90
>> cam1/models/atm/cam/src/physics/cam1/physpkg.F90
>> cam1/models/atm/cam/src/physics/cam1/restart_physics.F90
>> cam1/models/atm/cam/src/physics/cam1/tphysac.F90
>> cam1/models/atm/cam/src/control/history.F90
>> cam1/models/atm/cam/tools/definesurf/Makefile
>> cam1/models/atm/cam/tools/definesurf/fmain.f90
>> cam1/models/atm/cam/tools/definesurf/sghphis.f90
===============================================================
===============================================================
cam3_0_13
Originator: eaton ( Brian Eaton)
Date: Tue Aug 24 15:28:21 MDT 2004
Model: CAM
Version: CAM3.0.13
One-line Summary: add single interface for radiation to access trace gases
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM (create_ccsmcam): yes
Machines tested: IBM, SGI, Linux-Lahey
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: Yes, same-to-roundoff
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
The non-water gas phase constituents that are required by the radiation
codes are O3, CO2, CH4, N2O, CFC11, CFC12. All these constituents, except
CO2, are now accessed only via the chemistry module. A single method,
chem_get_cnst, returns constituent values regardless of whether they are
prognostic variables from the constituent array, or are diagnosed or
interpolated from a dataset. So all the logic for determining where a
constituent comes from now lives in the top level chemistry interface and
not in the top level radiation interface.
CO2 is a special case because it is assumed by the radiation algorithms to
be a single global value. For now it is still accessed directly from the
ghg_surfvals module. We expect this to change in the future.
These changes are roundoff due to the decision that chem_get_cnst should
return values as mass mixing ratio. Previously O3 values were accessed
from the interpolation routine as volume mixing ratio and converted to
mass mixing ratio as needed. The change in where this conversion happens
introduced a roundoff level difference.
===============================================================
===============================================================
cam3_0_12
Originator: jmccaa ( James McCaa)
Date: Thu Aug 19 09:06:13 MDT 2004
Model: CAM
Version: CAM3.0.12
One-line Summary: Addition of a column radiation model and various fixes to SCAM.
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM (create_ccsmcam): yes
Machines tested: IBM, SGI, Linux-Lahey
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
This checkin adds a column radiation model to the SCAM, and also
contains a variety of general fixes/enhancements to SCAM, mostly
centered around allowing it to work with nonstandard numbers of model
levels. Details follow...
Addition of the column radiation model:
---------------------------------------
The CRM can operate on old-style text CRM input files, or on a model
state initialized via any of scam's existing methods.
The CRM has both non-gui and gui modes. The gui version is accessed
via a new page under scam's "Options" tab. The non-gui version is
accessed via a shell script named crm.
I modified testscript to test the crm.
The following output from the test script indicates a pass (both lines
must be present):
scam passed its bfb test.
CRM passed.
General scam modifications:
---------------------------
Scam now honors the use_srfprop flag.
To facilitate running at nonstandard numbers of levels:
scam now guesses the appropriate level-dependent pressure file to use.
ncdio_atm.F90 has been modified to interpolate/extrapolate data to
the scam levels if necessary.
Handling of defaults files and quickstart files is slightly more robust.
Scam now plots interface level fields correctly.
Buffer sizes have been increased to allow scam to use up to 300 levels.
The gui's configure script has been modified to remove an unnecessary
dependency on libg2c when not using ncar graphics
I fixed an unitialized value problem in readiopdata.F90.
I modified init_model.F90 to handle C-to-fortran string passing
according to the unix standard.
I modified the scam testscript to retry (up to 10 times) on
compilation failures to try to overcome our persisent license server
problems.
I added a sanity check on precab(i) to cldwat.F90.
Updated init_model.F90 to reference ozone_data.F90 instead of comozp.F90.
Lastly, I removed 7 Mb of apparently unneeded ncar graphics files from
the repository.
===============================================================
===============================================================
cam3_0_11
Originator: eaton ( Brian Eaton)
Date: Fri Aug 13 12:06:12 MDT 2004
Model: CAM
Version: CAM3.0.11
One-line Summary: add fv 1.9x2.5 resolution; sst dataset generation tools
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM (create_ccsmcam): yes
Machines tested: IBM, SGI, Linux-Lahey
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Add 1.9x2.5 (96 lats, 144 lons) resolution for finite volume. Configure
using:
configure -dyn fv -res 1.9x2.5
Add sst dataset generation tools from Jim Rosinski.
===============================================================
===============================================================
cam3_0_10
Originator: eaton ( Brian Eaton)
Date: Mon Aug 9 12:11:12 MDT 2004
Model: CAM
Version: CAM3.0.10
One-line Summary: add new modules for radiation and ozone data
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM (create_ccsmcam): yes
Machines tested: IBM, SGI, Linux-Lahey
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
. Created modules for ozone datasets (ozone_data.F90), longwave radiation
(radlw.F90), and shortwave radiation (radsw.F90).
The old radini which initialized the old common block crdcon contained a
couple of constants which differed from the share constants. In order to
have a bit-for-bit commit the old values have been maintained.
. Add pointer variables inside radiation_tend to provide flexible method of
passing either prescribed or prognostic constituents to the radiation
calcs.
. Removed the common block ptrrgrid.h which was not being used.
. Added some FV 10x15 datasets to the Default*Namelist.xml files.
. Added back 13 files that were lost during the cam3_0_9 commit.
===============================================================
===============================================================
cam3_0_9
Originator: rosinski ( Jim Rosinski)
Date: Sun Jul 25 14:14:34 MDT 2004
Model: CAM
Version: CAM3.0.9
One-line Summary: Enable CAM to build and run using Intel compiler suite
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM (create_ccsmcam): no
Machines tested: IBM, SGI, Linux-Lahey
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, CLM changes by Vertenstein
Restart files change: no
Changes answers: no (bit-for-bit)
(Note: test-model.pl only does checks this -- if you run it with the "-compare" option)
Changes to CLM land-model: None (one-line description)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
(If not abort this tag and follow the procedures in)
http://www.cgd.ucar.edu/~cam/cam_checkins/checkin_process.shtml
Changes made:
CVS says:
M atm/cam/bld/Makefile
M lnd/clm2/src/biogeophys/Hydrology2Mod.F90
M lnd/clm2/src/main/decompMod.F90
M lnd/clm2/src/main/histFileMod.F90
A utils/esmf/build/linux_altix/ESMF_conf.h
A utils/esmf/build/linux_altix/base
A utils/esmf/build/linux_altix/base.site
A utils/esmf/build/linux_altix/base_variables
A utils/esmf/build/linux_altix/fix.h
A utils/esmf/build/linux_intel/ESMF_conf.h
A utils/esmf/build/linux_intel/base
A utils/esmf/build/linux_intel/base.site
A utils/esmf/build/linux_intel/base_variables
A utils/esmf/build/linux_intel/fix.h
M utils/timing/t_pr.c
Minor mods (listed above) to CAM Makefile, timing print routine, CLM
routines, and ESMF to enable CAM to build and run on Linux machines using
Intel compiler suite. 2-day error growth test using ifort and icc passed at
T42 in EUL configuration.
ESMF upgraded to ESMF_0_0_6 per Erik Kluzek. No code changes, but support
added for linux_intel and linux_altix. Note that linux_intel assumes that
Fortran and C compilers are both Intel (ifort and icc respectively). Though
CAM can be configured (and will run) with gcc, there is little point in doing
this as of this commit since ESMF will still require icc.
Minor mods to CLM routines to allow compilation with ifort.
Addition of an "fclose" to timing print routine t_pr.c to force a buffer
flush. Without it icc was truncating its printed output of model timers.
IMPORTANT: the only configurations of Intel compiler suite and underlying
glibc library that proved successful were:
ifort >= 8.0.046
icc >= 8.0.055
glibc >= 2.3.x
Tests in which not all these conditions were met resulted in various
failures, including general compiler errors, internal compiler errors, and
segmentation faults.
Notes:
o In general the Intel compilers create code that runs fast. Compile times
for optimized code (-O2) can be fairly long.
o The debugger (idb) is line-based. Hopefully it can be used in conjunction
with ddd to give a window-based option.
o Hybrid OpenMP/MPI configuration with Intel compilers was tested and does
work. Little timing difference compared with pure-MPI was observed.
Depending on model conifiguration, manipulation of environment variable
KMP_STACKSIZE may be required to avoid seg faults when threading is enabled.
o Tests enabling floating point traps (-fpe0) did not work in SPMD mode, so
this option was not included in Makefile mods. Hopefully the code changes
required to enable this option will be minor.
===============================================================
===============================================================
cam3_0_8
Originator: jeff ( Yen-Huei Lee)
Date: Wed Jul 21 11:53:53 MDT 2004
Model: CAM
Version: CAM3.0.8
One-line Summary: ntoplw changes to radiation arrays
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
(one-line description of changes: Makefile, include files, directories, etc.)
(Detailed description below)
Substantial timing or memory changes: no or Yes (describe)no
Requires change in run script: no
(one-line description of changes: data files, run-scripts, namelist etc.)
(Detailed description below)
Tested to work coupled with CCSM (create_ccsmcam): yes
Machines tested: IBM, SGI, Linux-Lahey
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit) or Yes (same-to-roundoff, same-physics, new-physics, new-climate)no
(Note: test-model.pl only does checks this -- if you run it with the "-compare" option)
Changes to CLM land-model: None (one-line description)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
(If not abort this tag and follow the procedures in)
http://www.cgd.ucar.edu/~cam/cam_checkins/checkin_process.shtml
Changes made:
M models/atm/cam/src/physics/cam1/crdcon.h
M models/atm/cam/src/physics/cam1/radae.F90
M models/atm/cam/src/physics/cam1/radclwmx.F90
M models/atm/cam/src/physics/cam1/radini.F90
M models/atm/cam/src/physics/cam1/restart_physics.F90
M models/atm/cam/src/physics/cam1/trcplk.F90
M models/atm/cam/src/physics/cam1/trcpth.F90
See: http://www.cgd.ucar.edu/cgi-binradae.F90/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_0_7
Originator: mvr ( Mathew Rothstein), eaton, bundy
Date: Tue Jul 20 10:03:47 MDT 2004
Model: CAM
Version: CAM3.0.7
One-line Summary: Added new physics interface (deep & shallow convection, stratiform, radiation)
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: to be determined
Requires change in run script: no
Tested to work coupled with CCSM (create_ccsmcam): yes
Machines tested: IBM, SGI, Linux-Lahey
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, eaton, bundy
Restart files change: no
Changes answers: Yes (same-to-roundoff)
Changes to CLM land-model: None (one-line description)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
R models/atm/cam/src/physics/cam1/cldcond.F90
R models/atm/cam/src/physics/cam1/cldnrh.F90
R models/atm/cam/src/physics/cam1/convtran.F90
R models/atm/cam/src/physics/cam1/moistconvection.F90
R models/atm/cam/src/physics/cam1/radctl.F90
A models/atm/cam/src/physics/cam1/convect_deep.F90
A models/atm/cam/src/physics/cam1/convect_shallow.F90
A models/atm/cam/src/physics/cam1/hk_conv.F90
A models/atm/cam/src/physics/cam1/radiation.F90
A models/atm/cam/src/physics/cam1/stratiform.F90
M models/atm/cam/src/control/history.F90
M models/atm/cam/src/dynamics/eul/inidat.F90
M models/atm/cam/src/dynamics/fv/inidat.F90
M models/atm/cam/src/dynamics/sld/inidat.F90
M models/atm/cam/src/physics/cam1/aerosol_intr.F90
M models/atm/cam/src/physics/cam1/buffer.F90
M models/atm/cam/src/physics/cam1/carbon_intr.F90
M models/atm/cam/src/physics/cam1/diagnostics.F90
M models/atm/cam/src/physics/cam1/dust_intr.F90
M models/atm/cam/src/physics/cam1/initindx.F90
M models/atm/cam/src/physics/cam1/inti.F90
M models/atm/cam/src/physics/cam1/param_cldoptics.F90
M models/atm/cam/src/physics/cam1/phys_adiabatic.F90
M models/atm/cam/src/physics/cam1/phys_idealized.F90
M models/atm/cam/src/physics/cam1/physics_types.F90
M models/atm/cam/src/physics/cam1/physpkg.F90
M models/atm/cam/src/physics/cam1/restart_physics.F90
M models/atm/cam/src/physics/cam1/srfxfer.F90
M models/atm/cam/src/physics/cam1/sulfur_intr.F90
M models/atm/cam/src/physics/cam1/tphysbc.F90
M models/atm/cam/src/physics/cam1/zm_conv.F90
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_0_6
Originator: mvr ( Mathew Rothstein)
Date: Fri Jul 16 16:54:46 MDT 2004
Model: CAM
Version: CAM3.0.6
One-line Summary: Added SOM tools; Modified CLM log output to version 3.0; Bug fix for system command strings
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM (create_ccsmcam): yes
Machines tested: IBM, SGI, Linux-Lahey
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: yes...updated log output to version 3.0
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
---added files needed to create SOM tools
A models/atm/cam/tools/definemld/Makefile
A models/atm/cam/tools/definemld/README
A models/atm/cam/tools/definemld/binf2c.f90
A models/atm/cam/tools/definemld/definemld1x1.c
A models/atm/cam/tools/definemld/definemld1x1.h
A models/atm/cam/tools/definemld/definemldbdy.f90
A models/atm/cam/tools/definemld/err_exit.c
A models/atm/cam/tools/definemld/neighborfill.f90
A models/atm/cam/tools/definemld/precision.f90
A models/atm/cam/tools/definemld/sm121.f90
A models/atm/cam/tools/definemld/wrap_nc.c
A models/atm/cam/tools/definemld/wrap_nf.f90
A models/atm/cam/tools/defineqflux/Makefile
A models/atm/cam/tools/defineqflux/README
A models/atm/cam/tools/defineqflux/backsolve.c
A models/atm/cam/tools/defineqflux/check_consistent.c
A models/atm/cam/tools/defineqflux/defineqflux.c
A models/atm/cam/tools/defineqflux/defineqflux.h
A models/atm/cam/tools/defineqflux/definesomic.h
A models/atm/cam/tools/defineqflux/err_exit.c
A models/atm/cam/tools/defineqflux/gepp.c
A models/atm/cam/tools/defineqflux/mksith.c
A models/atm/cam/tools/defineqflux/printeq.c
A models/atm/cam/tools/defineqflux/timediddle_mavg.c
A models/atm/cam/tools/defineqflux/wrap_nc.c
M models/lnd/clm2/src/main/initializeMod.F90
- updated log output to version 3.0
M models/atm/cam/src/control/system_cmd.c
- bug fix for handling of system command strings
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_0_5
Originator: jet ( John Truesdale)
Date: Thu Jun 24 13:17:43 MDT 2004
Model: CAM
Version: CAM3.0.5
One-line Summary: Updated SCAM files to match cam3.0 distribution
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
(one-line description of changes: Makefile, include files, directories, etc.)
(Detailed description below)
Substantial timing or memory changes: no
Requires change in run script: no
(one-line description of changes: mostly documentation)
(Detailed description below)
Tested to work coupled with CCSM (create_ccsmcam): no
Machines tested: none
Ran test-model.pl script: no
Tested on fv dynamics: na
Tested on eul dynamics:na
Tested on sld dynamics:na
Tested that restarts are bit-for-bit: na
Tested multiple constituents: na
Tested that different domain decompositions match bit-for-bit: na
Tested in adiabatic mode: na
Code reviewed by: myself
Restart files change: no
Changes answers: no
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM
CRB? yes
Changes made:
Added data directory and README file
deleted duplicate INSTALL and scam.html file in html directory
updated userguide.html
corrected but in testscript
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_0_4
Originator: mvr ( Mathew Rothstein), truesdale, vertenstein
Date: Mon Jun 21 10:19:32 MDT 2004
Model: CAM
Version: CAM3.0.4
One-line Summary: Cleanup of default settings of CAM/CLM input datasets; SCAM updated; Fixed two CLM bugs
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM (create_ccsmcam): yes
Machines tested: IBM, SGI, Linux-Lahey
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, truesdale, vertenstein
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: yes
fixed bug in code to exclude fields from history tape; fixed bug in code that
generates land surface datasets for fv
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
R models/atm/cam/tools/scam/scm_init/scam_rpc.h
A models/atm/cam/tools/scam/testscript
- new validation test script
M models/atm/cam/bld/DefaultCAMEXPNamelist.xml
- modified entry for fv2x2.5 input dataset, removed references to linked
files, modified some comments
M models/atm/cam/bld/DefaultCLMEXPNamelist.xml
- new entry for fv2x2.5 input dataset, modified some comments
M models/atm/cam/tools/scam/.scam_defaults
M models/atm/cam/tools/scam/configure
M models/atm/cam/tools/scam/ui/configure
M models/atm/cam/tools/scam/ui/manager.cpp
M models/lnd/clm2/src/main/histFileMod.F90
- fixed bug in code that allows user to exclude fields from primary
history tape
M models/lnd/clm2/src/mksrfdata/mksrfdatMod.F90
- fixed bug in generation of land surface datasets for fv
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_0_3
Originator: olson ( Jerry Olson)
Date: Thu Jun 17 14:33:44 MDT 2004
Model: CAM
Version: CAM3.0.3
One-line Summary: new initial conditions input/output
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM (create_ccsmcam): yes
Machines tested: IBM, SGI, Linux-Lahey
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Brian Eaton
Restart files change: no
Changes answers: no (bit-for-bit) except:
a) answers change in aqua-planet mode for all dycores
(new climate)
b) round-off for SLD on IBMs
Changes to CLM land-model: yes. trivial mods to comments
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
--------------
Description of changes:
- The history.F90 module is modified to include the Initial Conditions file as
part of the general history file build process. Details:
- the IC file is formatted exactly like a history file.
- just as in other history files, the IC file is filled via calls to
"outfld".
- no time averaging; fields are always instantaneous
- only one time sample written to file
- accumulation/output is in double precision
- "ptapes" increased from 6 to 7
- IC file is indexed "mtapes" in the history module (always at end of
active history file list)
- name of file includes ".i." rather than ".hX."
- two namelist parameters control the building of the IC file:
- INITHIST: controls frequency of IC file writes (for both land and
atmosphere): '6-HOURLY, 'DAILY', 'MONTHLY', 'YEARLY' or
'NONE'
- INITHIST_ALL: .false.: dump ONLY required fields to IC file
.true. : dump required AND optional fields to IC
file (DEFAULT)
- setting "EMPTY_HTAPES = .true." does NOT prevent the IC file from being
created.
- only setting INITHIST = 'NONE' prevents the IC file from being created.
- no space allocated on history buffer if INITHIST = 'NONE'
- the old IC file build routine, "write_inithist" is removed
- Note: - In an Eulerian run, the new code does not produce identical IC
files as the old code. The values for QCWAT, TCWAT, LCWAT, and
CLD will be different.
- In adiabatic mode, the FV dycore *still* does not create an IC
file with the proper temperature on it. In adiabatic mode, FV
does not update temperature. Therefore, the IC file will
contain the wrong temperature.
- History (and IC) fieldnames can be up to 16 characters long
- inidat.F90 re-written for all dycores. All fields are now read and processed
one at a time, reducing memory costs. All reads are done via a call to a
single netCDF interface (modified version of module written for the Land
Model by Mariana Vertenstein).
- inidat.F90 is backward compatible with OLD IC files.
- EUL and SLD spetru.F90 modules modified to be identical to eachother
- another new namelist variable:
RAD_FIRST_HOUR: Flag to force model to compute full radiation every
timestep in the first model hour (for use in
"forecast/analysis" runs)
Default: .false.
- restart files will change (because of extra characters for each field)
- Removed code associated with HADVTEST and VADVTEST cpp tokens ("blob" test)
- Bugfixes in aqua-planet mode:
- the field "landm" is now set to 0. Changes answers; new climate for
aqua-planet.
- fixed bug in analytical SST specification (in a branch of code not run
by default model)
- Bugfix to test-model.pm provided by Erik.
--------------
Files modified:
M models/atm/cam/bld/CAM_run.pm
- Bugfix provided by Erik
M models/atm/cam/src/control/comctl.h
- added namelist variable, "rad_first_hour"
M models/atm/cam/src/control/filenames.F90
- deleted old IC file logic
M models/atm/cam/src/control/history.F90
- extensive changes to add IC file as
part of the history file set (ptapes = 7)
- history file field names can be up to
16 characters long
M models/atm/cam/src/control/readinitial.F90
- added logic to trap possible model and IC file inconsistencies in the
grid/spectral resolutions
M models/atm/cam/src/control/runtime_opts.F90
- added new namelist variables
M models/atm/cam/src/dynamics/eul/dyndrv.F90
- removed blob test code
M models/atm/cam/src/dynamics/eul/grmult.F90
- removed blob test code
M models/atm/cam/src/dynamics/eul/inidat.F90
- cleanup
M models/atm/cam/src/dynamics/eul/linemsdyn.F90
- removed blob test code
M models/atm/cam/src/dynamics/eul/spetru.F90
- modified so that "eul" and "sld" spetru routines are identical
M models/atm/cam/src/dynamics/eul/stepon.F90
- removed logic associated with old IC file code
- added call to dump state field values to IC file buffer
M models/atm/cam/src/dynamics/eul/tfilt_massfix.F90
- removed blob test code
- fix to qneg3 diagnostic print
M models/atm/cam/src/dynamics/fv/inidat.F90
- cleanup
M models/atm/cam/src/dynamics/fv/inital.F90
- modified a "use" statement
M models/atm/cam/src/dynamics/fv/stepon.F90
- removed logic associated with old IC file code
- added call to dump state fields to IC file buffer
M models/atm/cam/src/dynamics/sld/dyndrv.F90
- removed blob test code
M models/atm/cam/src/dynamics/sld/grmult.F90
- removed blob test code
M models/atm/cam/src/dynamics/sld/inidat.F90
- cleanup
M models/atm/cam/src/dynamics/sld/restart_dynamics.F90
- removed unused fields from restart file
M models/atm/cam/src/dynamics/sld/scan2.F90
- removed logic associated with old IC file code
M models/atm/cam/src/dynamics/sld/scanslt.F90
- removed blob test code
M models/atm/cam/src/dynamics/sld/sltwgts.F90
- removed blob test code
M models/atm/cam/src/dynamics/sld/spetru.F90
- modified so that "eul" and "sld" spetru routines are identical
M models/atm/cam/src/dynamics/sld/stepon.F90
- added call to dump state fields to IC file buffer
M models/atm/cam/src/dynamics/sld/tfilt_massfix.F90
- removed blob test code
- fixed diagnostics bug in qneg3
M models/atm/cam/src/ocnsice/dom/sst_data.F90
- fixed bug in analytical SST specification (used in "aqua-planet" mode)
M models/atm/cam/src/physics/cam1/advnce.F90
- Do full radiation every timestep of first model hour if "rad_first_hour"
is .true.
M models/atm/cam/src/physics/cam1/diagnostics.F90
- add routine to dump physics fields to IC file buffer
M models/atm/cam/src/physics/cam1/phys_adiabatic.F90
- cleanup
M models/atm/cam/src/physics/cam1/physics_types.F90
- fix to qneg3 diagnostic print
M models/atm/cam/src/physics/cam1/phys_idealized.F90
- cleanup
M models/atm/cam/src/physics/cam1/physpkg.F90
- call routine to dump physics fields to IC file buffer
M models/atm/cam/src/physics/cam1/qneg3.F90
- fix to qneg3 diagnostic print
M models/atm/cam/src/physics/cam1/restart_physics.F90
- fix to quell compiler complaint
M models/atm/cam/src/physics/cam1/tphysbc.F90
- fix to qneg3 diagnostic print
M models/lnd/clm2/src/main/clm_varctl.F90
M models/lnd/clm2/src/main/inicFileMod.F90
--------------
Files added:
A models/atm/cam/src/control/ncdio_atm.F90
- Wrapper for reading a 2-D or 3-D field from a netCDF IC file
A models/atm/cam/src/dynamics/eul/diag_dynvar_ic.F90
- "outfld" dynamics state variables to history buffer (for IC file)
A models/atm/cam/src/dynamics/fv/diag_dynvar_ic.F90
- "outfld" dynamics state variables to history buffer (for IC file)
A models/atm/cam/src/dynamics/sld/diag_dynvar_ic.F90
- "outfld" dynamics state variables to history buffer (for IC file)
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_0_2
Originator: erik ( Erik Kluzek)
Date: Wed Jun 9 10:29:42 MDT 2004
Model: CAM
Version: CAM3.0.2
One-line Summary: Major test-model.pl update, bug fix for DGVM in clm code
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM (create_ccsmcam): yes
Machines tested: IBM, SGI, Linux-Lahey
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Brian Eaton, Matt Rothstein, and Tom Henderson
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: DGVMMod.F90, and driver.F90 from Mariana for DGVM bug-fix
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Synopsis:
Overall this is a major refactoring of test-model.pl. This is the version
of test-model.pl that was on the subroutinization branch for the last year.
This improves the default behavior of test-model.pl and makes it easier
and more flexible to use. It also provides for the ability to test the
test-model.pl test harness.
Files changed and a description of the changes:
M models/atm/cam/bld/CAM.pm
M models/atm/cam/bld/CAM_config.pm
M models/atm/cam/bld/CAM_namelist.pm
M models/atm/cam/bld/CAM_run.pm
R models/atm/cam/bld/check-namelist.pl
R models/atm/cam/bld/lsmexp.pm
M models/atm/cam/bld/run-model.pl
Add abilities needed by new test-model.pl. Don't delete *.f90 files.
Delete lsmexp.pm file. Make run options more clear. Add printlev and test
option to namelist. Build_namelist always does a new on a namelist
object. Get rid of check-namelist.pl as not used.
M models/atm/cam/test/system/CAM_test.pm
M models/atm/cam/test/system/README
A models/atm/cam/test/system/babyblue.test-model.pl.log ! testonly mode results
M models/atm/cam/test/system/dao_batch.csh
A models/atm/cam/test/system/default_tests.test-model.xml ! XML default tests
A models/atm/cam/test/system/namelist-config.test-model.dbg.log ! file of
configuration/namelists from testonly mode
M models/atm/cam/test/system/ncar_batch.csh
A models/atm/cam/test/system/specs-testslist.test-model.xml ! XML file of
specifications of fields expected in testslist XML files
M models/atm/cam/test/system/test-model.pl
New version of test-model.pl as described below.
M models/atm/cam/tools/newcprnc/Makefile
Allow use of env variables needed by test-model.pl. Still not used in
test-model.pl as newcprnc doesn't do a proper mass weighting of the
differences.
M models/lnd/clm2/src/biogeochem/DGVMMod.F90
M models/lnd/clm2/src/main/driver.F90
Bug-fixes from Mariana for DGVM.
Details of changes to test-model.pl
test-model.pl -- new version with XML files to control tests that
are run. This is the version of test-model.pl that was
used on the subroutinization branch. Modified batch scripts
to correspond to new command line options.
Default is changed so that if you are comparing and a test is
NOT bit-for-bit to the comparision code -- test-model.pl will
stop.
New command line options to test-model.pl
-cleanonly = only clean files out
-help_defaults = Create file with configuration/namelist information
for all tests -allow_nonb4bcompare = Allow comparisions not bit-for-bit to continue.
-[no]spmd = Turn on [off] SPMD mode
-[no]smp = Turn on [off] SMP mode
-nocompareifskipped = If you are skipping tests a comparision will NOT
be run if a given test is skipped. -namelist "namelist" = Reads in namelist and adds these options to the
namelists that are used.
-testonly = Only test test-model.pl running configure and
build-namelist for all tests to demonstrate what
test-model.pl would have produced.
-testslist filename = Use the given XML file of tests rather
than the default file.
Information on all command-line options is given with
test-model.pl -help
newcprnc -- Add ability to use the env variables needed for test-model.pl
VPATH, EXEDIR. To use this in test-model.pl change the path
in test-model.pl and change CAM_test.pm to NOT use the "-m" option
as the behavior of this flag was reversed in newcprnc from cprnc.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_0_1
Originator: erik ( Erik Kluzek)
Date: Tue May 25 16:40:56 MDT 2004
Model: CAM
Version: CAM3.0.1
One-line Summary: Get rid of ncdata_vers stuff, and fix SOM comparision bug in test-model.pl
cam-bugs Requests resolved: 118
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: yes
(ncdata_vers is no longer needed)
Tested to work coupled with CCSM (create_ccsmcam): yes
Machines tested: IBM, SGI, Linux-Lahey
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Get rid of ncdata_vers stuff in the build-namelist scripts.
Fix bug in SOM comparisions, so test-model.pl will flag a difference
if SOM code changes. Previously this aspect was broken and the comparisions
files were mislabled. Now the files are properly labeled, SOM control
tests are made in a seperate directory, and namelist and configuration
variables are consistent such that the comparision is valid. This problem
was documented as cam-bug #118.
No code was changed only build and configuration and test scripts.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam3_0
Originator: eaton ( Brian Eaton)
Date: Mon May 24 15:33:43 MDT 2004
Model: CAM
Version: CAM3.0
One-line Summary: remove old CAM2 documentation
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: none
Ran test-model.pl script: no
Tested on fv dynamics: no
Tested on eul dynamics: no
Tested on sld dynamics: no
Tested that restarts are bit-for-bit: no
Tested multiple constituents: no
Tested that different domain decompositions match bit-for-bit: no
Tested in adiabatic mode: no
Code reviewed by: myself
Restart files change: no
Changes answers: no
Changes made:
Remove old CAM2 documentation.
===============================================================
===============================================================
cam2_0_2_dev86
Originator: mvertens ( Mariana Vertenstein)
Date: Thu May 20 10:56:00 MDT 2004
Model: CAM
Version: CAM2.0_2.dev86
One-line Summary: updated clm code clm3_deva_14
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: yes (TER.01a.T31_gx3v5.K.blackforest)
Machines tested: IBM, SGI, Linux(lf95)
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
1) Updated clm code to clm3_deva_14
2) Updated clm documentation to clm3_deva_14
3) One line modification to
atm/src/control/ioFileMod.F90
clm2/src/main/fileutils.F90
to fix file removal problem associated with mss writes
===============================================================
===============================================================
cam2_0_2_dev85
Originator: jet ( John Truesdale)
Date: Tue May 18 14:53:15 MDT 2004
Model: CAM
Version: CAM2.0_2.dev85
One-line Summary: Added SCAM to cam_dev
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
(one-line description of changes: Makefile, include files, directories, etc.)
(Detailed description below)
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: yes
Machines tested: IBM, SGI, Compaq
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: Added cpp delimeted code to allow clm to work
in single column mode.
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
My check-in includes 3 sets of modifications to CAM code.
1) modifications to allow it to work in single column mode
2) additional source code and configuration files to create GUI
for the Single-Column CAM (SCAM).
3) modification to CAM to allow it to produce a history file that
can be used by SCAM to run in a pseudo-iop mode using model
data.
Most of the modifications to CAM source have been delimited by cpp
tokens (#defined SCAM and #defined BFB_CAM_SCAM_IOP)
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam2_0_2_dev84
Originator: mvr ( Mathew Rothstein)
Date: Fri May 14 09:43:26 MDT 2004
Model: CAM
Version: CAM2.0_2.dev84
One-line Summary: Replaced numerical recipes routines for random# generation and findvalue; modified Makefile to eliminate double comma problem; other small bug fixes
cam-bugs Requests resolved: none
Requires change in build system: yes
- replaced the use of function subst in Makefile with patsubst to eliminate
double comma problem
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: yes
Machines tested: IBM, SGI, Compaq
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, pjr
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
R models/atm/cam/src/utils/findvalue.F90
- removing old find routine based on numerical recipe algorithm
A models/atm/cam/src/utils/marsaglia.F90
- new random number generator for cloud simulator
M models/atm/cam/src/physics/cam1/cloudsimulator.F90
- modified to seed new random number generator
M models/atm/cam/src/physics/cam1/icarus_scops.F90
- now uses new random number generator rather than algorithm from
numerical recipes
M models/atm/cam/src/physics/cam1/radclwmx.F90
- added r8 declaration to array aer_trn_ttl
M models/atm/cam/src/physics/cam1/radcswmx.F90
- now uses fortran intrinsic for finding an array minimum rather
than a routine from numerical recipes
M models/atm/cam/bld/Makefile
- replaced the use of function subst with patsubst to eliminate ,,
M models/atm/cam/test/system/CAM_test.pm
- modified error message in test-model
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam2_0_2_dev83
Originator: mvr ( Mathew Rothstein)
Date: Thu May 13 09:56:08 MDT 2004
Model: CAM
Version: CAM2.0_2.dev83
One-line Summary: Replaced sort routine; modified handling of namelist vars for computing run stop date; modifed input dataset path strings
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: yes
Machines tested: IBM, SGI, Compaq
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: same-to-roundoff
Changes to CLM land-model: yes
- modified strings of default input datasets to include full path from csmdata
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
R models/atm/cam/src/utils/sortarray.F90
- removing old sort routine based on numerical recipe algorithm
A models/atm/cam/src/utils/quicksort.F90
- replacing sort routine with new quick-sort algorithm
M models/atm/cam/src/physics/cam1/radclwmx.F90
- modified to call new sort routine
M models/atm/cam/src/physics/cam1/radcswmx.F90
- modified to call new sort routine
M models/atm/cam/src/control/time_manager.F90
- modified handling of namelist parameters for computing stop date
M models/atm/cam/bld/DefaultCAMEXPNamelist.xml
- modified input dataset strings to include full path from csmdata
M models/atm/cam/bld/DefaultCLMEXPNamelist.xml
- modified input dataset strings to include full path from csmdata
M models/atm/cam/bld/camexp.pm
- modified to use new input dataset strings
M models/atm/cam/bld/clm2exp.pm
- modified to use new input dataset strings
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam2_0_2_dev82
Originator: jmccaa ( James Mccaa)
Date: Mon May 10 15:40:39 MDT 2004
Model: CAM
Version: CAM2.0_2.dev82
One-line Summary: Tuning modifications for the finite volume dycore.
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: yes
Machines tested: IBM, SGI, intel
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: Yes, for FV -- bit-for-bit for EUL and SLD
Changes to CLM land-model: None (one-line description)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Modified tuning for FV 2x2.5, and introduced tuning settings for FV 4x5
and FV 1x1.25.
A new i.c. file for FV 4x5 was introduced.
These files were modified:
M cam1/models/atm/cam/bld/DefaultCAMEXPNamelist.xml
M cam1/models/atm/cam/src/dynamics/fv/dycore.F90
M cam1/models/atm/cam/src/physics/cam1/cldwat.F90
M cam1/models/atm/cam/src/physics/cam1/cloud_fraction.F90
M cam1/models/atm/cam/src/physics/cam1/moistconvection.F90
M cam1/models/atm/cam/src/physics/cam1/zm_conv.F90
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam2_0_2_dev81
Originator: mvertens ( Mariana Vertenstein)
Date: Tue May 4 10:40:20 MDT 2004
Model: CAM
Version: CAM2.0_2.dev81
One-line Summary: updated clm code to clm3_deva_10
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: yes (TER.01a.T31_gx3v5.K.blackforest)
Machines tested: IBM, SGI, Linux(lf95)
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Forrest Hoffman
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: Yes (see below)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
1) Updated clm code to clm3_deva_10
This primarily includes the addition of CrayX1 directives to the code
as well as changes to DGVM to make it effectively work on the Cray X1
Also changed the name of the clm file changed in cam2_0_2_dev80 back
to its original name for the sake of cvs history continuity
2) Modified history.F90 to use (ieor,iand) rather than (xor,and)
===============================================================
===============================================================
cam2_0_2_dev80
Originator: eaton ( Brian Eaton) and Pat Worley
Date: Wed Apr 28 17:44:37 MDT 2004
Model: CAM
Version: CAM2.0_2.dev80
One-line Summary: merge vector branch
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: Yes
Requires change in run script: no
Tested to work coupled with CCSM: yes
Machines tested: IBM, SGI, Linux/lf95, Linux/pgi
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Pat Worley
Restart files change: no
Changes answers: Yes (same-to-roundoff)
Changes to CLM land-model: rename STATICEcosysDynMod.F90 --> STATICEcosystemDynMOD.F90
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
1. Merge of vector and scalar code branches:
The mods for the NEC that didn't change answers (bit-for-bit) were
committed previously. This set of mods introduces a roundoff level
change into the simulation. The main modifications are:
. A new version of radcswmx. This code runs a little slower on the IBM
(about 4% of total run time). It introduces roundoff level changes
which affects all dycores.
. Large sections of code have been changed which reorder the spectral
calculations in the EUL dycore. We observed no performance degradation
on the IBM from this change.
. Most of the Cray optimizations have been included as well. They are mainly
compiler directives.
2. Rename a CLM file which didn't match the module name it contained (this
breaks the dependency generator)
3. Some cleanup in ncar_batch.csh
===============================================================
===============================================================
cam2_0_2_dev79
Originator: eaton ( Brian Eaton)
Date: Fri Apr 23 07:56:47 MDT 2004
Model: CAM
Version: CAM2.0_2.dev79
One-line Summary: T31 tuning; T31 and T85 datasets; diagnostics
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: yes
Machines tested: IBM, Linux/lf95
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
. Tuning mods for T31
. New datasets for T31 and T85
. Diagnostics:
- AIX specific traceback call added to abortutils
- Check for NaNs in timeinterp
===============================================================
===============================================================
cam2_0_2_dev78
Originator: eaton ( Brian Eaton) and Mat Rothstein
Date: Tue Apr 20 10:52:58 MDT 2004
Model: CAM
Version: CAM2.0_2.dev78
One-line Summary: misc configure/Makefile/test-model changes
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: yes, on Blackforest
Machines tested: IBM, Linux/lf95
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, mvr
Restart files change: no
Changes answers: no
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
. Fix hack in test-model that set F_OPTIMIZATION_OVERRIDE for certain PGI
tests. The hack was broken by the inclusion of a commandline option to
configure for setting the F_OPTIMIZATION_OVERRIDE macro in the Makefile.
. Add print in test-model output to tells which log contains NOT BFB
message.
. Various configure/Makefile changes:
- Temporarily remove -Mbounds from PGI debug options. CLM2 is not working
with this option.
- Add -cppdefs option to configure for defining CPP tokens. These cpp
definitions are appended to the Makefile defaults. Remove the -precomp
option which was adding #define lines to misc.h.
- Add capability to configure to determine the name of the MPI library
being used. This is currently most useful for Linux which may use
either mpi or mpich as the library name.
- Clean up Linux section of Makefile. Fixed a bug which broke the build
if the user specified the option '-fc pgf90' to configure. Remove
options for compilers that don't have corresponding builds of the ESMF
library.
. Add missing addfld call for H2O2DRY in sulfur_intr.F90
===============================================================
===============================================================
cam2_0_2_dev77
Originator: mvr ( Mathew Rothstein) and Brian Eaton
Date: Tue Apr 6 10:38:51 MDT 2004
Model: CAM
Version: CAM2.0_2.dev77
One-line Summary: Update cloud simulator; T31 mods; Enable cloud overlap scenarios; Bug fix in masterlist lookups; Re-commit of infnan mods
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: yes
Machines tested: IBM, SGI, Compaq
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, eaton
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: yes
replaced infnan mods of dev65, mistakenly removed
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
R models/atm/cam/src/physics/cam1/cloudsimulatorparms.F90
- put data in icarus_scops.h, put doisccp into cloudsimulator.F90
A models/atm/cam/src/physics/cam1/icarus_scops.F90
- new module containing subroutine isccp_cloud_types (converted to free
format) and subroutine ran0_vec
- change output types to r8
- replace "STOP" with endrun calls
- add diagnostic for meanttop
- add icarus_scops_init routine to read in tautab and invtau arrays
M models/atm/cam/bld/DefaultCAMEXPNamelist.xml
- added default value for dataset isccpdata, needed for isccp simulator
M models/atm/cam/bld/camexp.pm
- modified to handle namelist parameters related to the isccp simulator
M models/atm/cam/src/control/cam.F90
- remove call to initialization routine
M models/atm/cam/src/control/history.F90
- use icarus_scops instead of use cloudsimulatorparms
- remove addfld and add_default calls for diagnostic fields (move to
cloudsimulator_init)
- remove unnecessary addfld for aermmr
- added endrun calls for hash table lookups of fields not on masterlist
M models/atm/cam/src/control/runtime_opts.F90
- use cloudsimulator instead of cloudsimulatorparms
M models/atm/cam/src/dynamics/eul/tfilt_massfix.F90
- removed outfield calls and related code for fields not on masterlist
M models/atm/cam/src/dynamics/sld/tfilt_massfix.F90
- removed outfield calls and related code for fields not on masterlist
M models/atm/cam/src/physics/cam1/check_energy.F90
- removed unneeded code to fix bug identified by SX6 machine
M models/atm/cam/src/physics/cam1/cldwat.F90
- added special setting of physical constants for T31 runs
M models/atm/cam/src/physics/cam1/cloud_fraction.F90
- added special setting of physical constants for T31 runs
M models/atm/cam/src/physics/cam1/cloudsimulator.F90
- rename isccptab --> cloudsimulator_init
- rename ccm_isccp --> cloudsimulator_run
- add state to cloudsimulator_run args
- add addfld and add_default calls for diagnostic fields
- change arguments to isccp_cloud_types from single columns to chunks
- use the totalcldarea diagnostic from isccp_cloud_types rather than the
cltot diagnostic from cam to decide whether it's cloudy or not. This
avoids assigning fillvalues inside the isccp_cloud_types code
M models/atm/cam/src/physics/cam1/constituents.F90
- commented out code for fields no longer in use
M models/atm/cam/src/physics/cam1/initindx.F90
- commented out code for fields no longer in use
M models/atm/cam/src/physics/cam1/inti.F90
- call cloudsimulator_init
M models/atm/cam/src/physics/cam1/moistconvection.F90
- added special setting of physical constants for T31 runs
M models/atm/cam/src/physics/cam1/param_cldoptics.F90
- add_default call for icldiwp
- modify long name attribute for icldlwp
M models/atm/cam/src/physics/cam1/radclwmx.F90
- uncommented line and modified associated array declaration in LW
cloud-overlap code (critical for certain cloud overlap assumptions)
M models/atm/cam/src/physics/cam1/tphysbc.F90
- use cloudsimulator instead of cloudsimulatorparms
- update cloudsimulator_run args
- removed outfield calls and related code for fields not on masterlist
M models/atm/cam/src/physics/cam1/zm_conv.F90
- added special setting of physical constants for T31 runs
M models/lnd/clm2/src/main/nanMod.F90
- replaced mods for quiet/signalling NaN's of dev65 that were
mistakenly overwritten with dev75
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam2_0_2_dev76
Originator: rosinski ( Jim Rosinski)
Date: Tue Mar 30 15:11:23 MST 2004
Model: CAM
Version: CAM2.0_2.dev76
One-line Summary: bit-for-bit NEC vector branch changes.
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: yes
Machines tested: IBM, SGI, PC
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit). PC is only roundoff
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
o Merged cam2.0.2.dev67.brnchT.nec15 (vector branch) changes that do not affect
answers onto cam_dev. Opted for Pat Worley's dev68 communication mods over NEC
communication mods where conflicts occurred. A few NEC_SX ifdefs still
exist in the committed code. Most involve reversing "do k" and "do i"
loops in dynamics routines for vectorization. An inner loop over k results
in a non-unit-stride situation on scalar machines and should be avoided.
That is why these code portions were left ifdef'd
o Note 1: The code appears to be good to roundoff on IBM machines when NEC_SX is
#define'd. To test this on an IBM, "! defined AIX" needs to be added in 2
obvious places in cfort.h and abortutils.F90
o Note 2: An odd behavior was discovered whereby inconsistent chunck indices are
printed from ice_srf.F90 at times when the ice temperature profile is reset to
linear. This behavior appears to only manifest when threads * tasks exceeds
the number of physical CPUs available. It was decided to commit without
resolving this discrepancy because the behavior occurs in the base model and
in earlier libraries, i.e. independent of the NEC vector mods.
===============================================================
===============================================================
cam2_0_2_dev75
Originator: mvertens ( Mariana Vertenstein)
Date: Fri Mar 26 15:07:40 MST 2004
Model: CAM
Version: CAM2.0_2.dev75
One-line Summary: updated clm to clm3_deva_08
cam-bugs Requests resolved: none
Requires change in build system: no (can removed atmlnd_share from Filepath howeer)
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: yes
Machines tested: IBM, Linux
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, (Brian Eaton revied cam time_manager.F90)
Restart files change: yes
Changes answers: no (bit-for-bit)
Changes to CLM land-model: Yes - updated to clm3_deva_08
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
1) added perpetual model to cam-clm
- updated clm/src to clm3_deva_09 which had perpetual mode changes.
- modified models/atm/cam/src/control/time_manager.F90 to add offset argument to
subroutine get_perp_date(..)
2) changes to models/atm/cam/bld/:
- removed setting of reset_csim_iceprops in DefaultCAMEXPNamelist.xml
- removed atmlnd_share from configure
3) chages to atmlnd_share/:
atmlnd_share no longer needed
atmlnd_share/cfort.h updated and moved to models/atm/cam/src/control and models/lnd/clm2/src/main
atmlnd_share/datetime.F90 moved to models/atm/cam/src/control
atmlnd_share/infnan.F90 moved to models/atm/cam/src/control
atmlnd_share/linebuf_stdout.c moved to models/atm/cam/src/control
atmlnd_share/mpishorthand.F moved to models/atm/cam/src/control
atmlnd_share/string_utils.F90 moved to models/atm/cam/src/control
atmlnd_share/wrap_nf.F90 moved to models/atm/cam/src/control
4) other changes:
removed all files in following directories - they are now obsolete:following obsolete files:
models/bld/
scripts/gui/
scripts/gui_run/
scripts/system_test/
scripts/test.a1/
scripts/test.a2/
scripts/tests/cpl_check/
scripts/tests/gprof/
scripts/tests/sub_sgi/
scripts/tests/test_scripts/
scripts/tests/timing/
===============================================================
===============================================================
cam2_0_2_dev74
Originator: mvr ( Mathew Rothstein)
Date: Thu Mar 18 11:51:59 MST 2004
Model: CAM
Version: CAM2.0_2.dev74
One-line Summary: Add fields to master list for radiation diagnostics; fixes to build-namelist for "naked decimal" and handling of raw land surface datasets
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: yes
Machines tested: IBM, SGI, Compaq
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, eaton
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: yes
clm2exp.pm: now properly handles user-specified raw land surface datasets
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
M cam1/models/atm/cam/bld/clm2exp.pm
- now properly handles user-specified raw land surface datasets
M cam1/models/atm/cam/bld/namelist.pm
- added ability to read "naked" decimal entries
M cam1/models/atm/cam/bld/script_tests/namelist.pl
- added namelist tests for "naked" decimal entries and more
M cam1/models/atm/cam/src/control/history.F90
- add addfld calls for ASDIR, ASDIF, ALDIR, ALDIF
- remove old addfld call for REL
- add addfld call for SST
M cam1/models/atm/cam/src/physics/cam1/comsrf.F90
- init trefmxav, trefmnav
- init lchnk and ncol in srfflx_state_2d
M cam1/models/atm/cam/src/physics/cam1/diagnostics.F90
- change diag_surf to use srfflx_state and surface_state types
- add outfld calls for ASDIR, ASDIF, ALDIR, ALDIF
- add outfld call for SST
M cam1/models/atm/cam/src/physics/cam1/param_cldoptics.F90
- add addfld and outfld calls for EMIS, CICEWP, REI, REL
M cam1/models/atm/cam/src/physics/cam1/physpkg.F90
- pass srfflx_state and surface_state types to diag_surf instead of the
individual fields
M cam1/models/atm/cam/src/physics/cam1/prescribed_aerosols.F90
- remove old outfld call for REL
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam2_0_2_dev73
Originator: mirin ( Arthur Andrew Mirin)
Date: Mon Mar 15 09:16:02 MST 2004
Model: CAM
Version: CAM2.0_2.dev73
One-line Summary: Mod_comm and related updates.
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: new test on blackforest
Machines tested: IBM, SGI, Compaq
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Sawyer
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
(A) mod_comm has been significantly streamlined to make it easier to
understand the coding; module mod_irreg has been combined into mod_comm;
option to dynamically allocate target window has been added.
More work still needs to be done. Most of the changes in the FV routines
are due to the new linkage.
(B) the utils/pilgrim subdirectory can now be included for all 3 dycores;
this requires setting the if-def MODCM_DP_TRANSPOSE. Consequently,
the mod_comm option for the dynamics/physics transposes in phys_grid
is now operational with all 3 dycores;
(C) namelist variable modcomm_method has been fissioned into 2 new
namelist variables - modcomm_transpose and modcomm_geopk; these are
applicable to the FV dycore.
(D) the namelist variable phys_alltoall is now overloaded when applicable
to mod_comm; its new value is 11 + (mod_comm method), where (mod_comm method)
can have values from 0 to 3; see code documentation in phys_grid.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam2_0_2_dev72
Originator: bundy ( Danielle R. B. Coleman)
Date: Fri Mar 12 12:13:22 MST 2004
Model: CAM
Version: CAM2.0_2.dev72
One-line Summary: fix FV mass conservation problem
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Linux (Lahey)
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: Phil Rasch
Restart files change: no
Changes answers: changes answers in FV only
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
models/atm/cam/src/dynamics/fv/fill_module.F90
fix mass conservation for small mixing ratios by scaling tiny fill value
to magnitude of mixing ratio.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam2_0_2_dev71
Originator: bundy ( Danielle R. B. Coleman)
Date: Wed Mar 10 16:13:03 MST 2004
Model: CAM
Version: CAM2.0_2.dev71
One-line Summary: Add dry mixing ratio formulation, suite of test tracers
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: yes
Machines tested: IBM, SGI, Linux (Lahey)
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: Phil Rasch
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
a. A list of any subroutines eliminated;
models/atm/cam/src/physics/cam1/test_tracers.F90
models/atm/cam/src/physics/cam1/rnozunit.F90
b. A list of any subroutines added and what they do; and
models/atm/cam/src/physics/cam1/tracers.F90
replaces eliminated file test_tracers.F90
an interface for a generic suite of tracers specified in tracers_suite.F90
models/atm/cam/src/physics/cam1/tracers_suite.F90
replaces eliminated rnozunit.F90
contains all the specifications for a suite
of tracers, and is interfaced to the model via tracers.F90.
c. For existing files that have been modified, an itemized list
of the changes and where they can be found
models/atm/cam/src/physics/cam1/advnce.F90
add call to tracers_timestep_init
models/atm/cam/src/physics/cam1/check_energy.F90
add check_tracers mass capability
models/atm/cam/src/physics/cam1/constituents.F90
add mixtype (wet or dry) field to constituent description
add functions to get type by index or name
models/atm/cam/src/physics/cam1/convtran.F90
/models/atm/cam/src/physics/cam1/diagnostics.F90
outfld call for PDELDRY
models/atm/cam/src/physics/cam1/initindx.F90
models/atm/cam/src/physics/cam1/inti.F90
change namelist variables that control tracers
models/atm/cam/src/physics/cam1/moistconvection.F90
models/atm/cam/src/physics/cam1/physics_types.F90
add capability for wet or dry mixtype tracers
models/atm/cam/src/physics/cam1/physpkg.F90
calculates pdeldry for dry mixtypes
added gavglook subroutine for checking mass conservation
models/atm/cam/src/physics/cam1/tphysac.F90
change tracer namelist variables, subroutine calls
check mass of tracers
for LR dyn, convert dry to wet at end
models/atm/cam/src/physics/cam1/tphysbc.F90
calculate and write out dry surface pressure PSDRY
check mass of tracers
models/atm/cam/src/physics/cam1/vertical_diffusion.F90
calculate tend wrt dry pressure for dry mixtype tracers
by calling vdiff for again only if there are dry tracersm
models/atm/cam/src/physics/cam1/turbulence.F90:trbintr()
add optional argument to skip outfld calls
( used on 2nd call to subroutine from 2nd call to vdiff)
models/atm/cam/src/control/filenames.F90
add boundary file for test tracers: bndtvsf6
models/atm/cam/src/control/runtime_opts.F90
change namelist variables for tracers:
[new] bndtvsf6
[mod] tracers_flag [replaces trace_test{1-3}]
models/atm/cam/src/control/history.F90
addfld PSDRY
models/atm/cam/src/advection/slt/qmassa.F90
models/atm/cam/src/advection/slt/xqmass.F90
add capability for dry mixtype tracers
models/atm/cam/src/dynamics/eul/dp_coupling.F90
models/atm/cam/src/dynamics/sld/dp_coupling.F90
only change wet mixtype tracers to dry/wet mr for dyn/phys
models/atm/cam/src/dynamics/fv/dp_coupling.F90
convert dry mixtype tracers to wet/dry for dyn/phys
models/atm/cam/src/dynamics/fv/stepon.F90
before 1st call to dynamics, convert dry tracers to wet
models/atm/cam/src/dynamics/eul/inidat.F90
models/atm/cam/src/dynamics/sld/inidat.F90
models/atm/cam/src/dynamics/fv/inidat.F90
name changes test_tracers -> tracers
models/atm/cam/src/dynamics/eul/scanslt.F90
models/atm/cam/src/dynamics/sld/scanslt.F90
pass vars (n3,q3) to slt interface
models/atm/cam/src/dynamics/eul/tfilt_massfix.F90
models/atm/cam/src/dynamics/sld/tfilt_massfix.F90
modify qneg threshold (corm) for dry mixtypes to conserve mass
/models/atm/cam/src/dynamics/eul/comslt.F90
add trcavg variable
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam2_0_2_dev70
Originator: pworley ( Patrick H Worley)
Date: Tue Mar 9 08:30:43 MST 2004
Model: CAM
Version: CAM2.0_2.dev70
One-line Summary: replaced use of mpi_comm_world with mpicom in initialization of swap_comm module
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: yes
Machines tested: IBM, SGI, Linux/Lahey
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
modified control/swap_comm.F90:
replaced mpi_comm_world with mpicom when duplicating communicators (for use in
swap_comm module routines). Without this, CAM does not work with CCSM.
added test/system/create_ccsmcam:
script provided by Mariana Vertenstein to test CAM with CCSM
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam2_0_2_dev69
Originator: mvr ( Mathew Rothstein)
Date: Thu Mar 4 17:35:12 MST 2004
Model: CAM
Version: CAM2.0_2.dev69
One-line Summary: Fix for DMS bug; Fix for co2 ramping with rate<0; Added co2vmr to history output
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Compaq
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, eaton
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None (one-line description)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
M models/atm/cam/src/physics/cam1/acbnd.F90
- move allocate outside of masterproc conditional in acbndint()
M models/atm/cam/src/physics/cam1/dmsbnd.F90
- fix bug reading new time sample in dmsbndint()
- move allocate outside of masterproc conditional in dmsbndint()
M models/atm/cam/src/physics/cam1/soxbnd.F90
- move allocate outside of masterproc conditional in soxbndint()
M models/atm/cam/src/physics/cam1/ghg_surfvals.F90
- fixed the handling of co2 ramping when the rate is less than zero
M models/atm/cam/src/control/history.F90
- added co2vmr (co2 volume mixing ratio) to the history output files
===============================================================
===============================================================
cam2_0_2_dev68
Originator: pworley ( Patrick H Worley)
Date: Wed Mar 3 12:55:10 MST 2004
Model: CAM
Version: CAM2.0_2.dev68
One-line Summary: communication optimizations and removal of PVP ifdefs
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: not with default settings
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Linux cluster
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself (earlier version reviewed by Tom Henderson)
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
1) Removed PVP ifdefs (to make merge with vector branches cleaner). Files affected:
control/cam.F90, rgrid.F90
dynamics/eul/comspe.F90, dyn.F90, dyndrv.F90, grcalc.F90, hordif.F90,
inital.F90, initcom.F90, quad.F90, settau.F90, spetru.F90,
trunc.F90, tstep.F90
dynamics/fv/initcom.F90
dynamics/sld/dyn.F90, dyndrv.F90, dynpkg.F90, grcalc.F90, hordif.F90,
initcom.F90, quad.F90, spetru.F90, trunc.F90, tstep.F90,
tstep1.F90, vertnm.F90
2) Reimplemented communication algorithms in spectral dycore realloc routines
and in the physics load balancing, using either MPI collectives or
point-to-point implementations that use the swap_comm module. swap_comm
was imported from PSTSWM and PCCM, and supports 28 different options. The default
is to use collectives. When not using collectives, the default is to use the
original point-to-point algorithm (using mpi_sendrecv). Runtime options
dyn_alltoall, dyn_allgather, phys_alltoall, swap_comm_order, and swap_comm_protocol
were added to runtime_options to specify these. For example,
dyn_alltoall == 0 means to use MPI_Alltoallv in realloc4a and realloc4b.
dyn_alltoall == 1 means to use one of the point-to-point algorithms.
Physics load balancing was also changed to use the same communication buffer
as that used by the realloc routines. Finally, the physics load balancing
options were tweaked, primarily to remove a high complexity initialization algorithm
that was dominating runtime for short runs at high resolution. Files affected:
control/runtime_opts.F90, spmd_utils.F90, spmdinit.F90, swap_comm.F90, mpi_wrap.F90
dynamics/eul/dp_coupling.F90, realloc4.F90, realloc7.F90, scan2.F90, spmd_dyn.F90
dynamics/fv/spmd_dyn.F90
dynamics/sld/dp_coupling.F90, realloc4.F90, realloc7.F90, scan2.F90, spmd_dyn.F90
physics/cam1/phys_grid.F90
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam2_0_2_dev67
Originator: mvr ( Mathew Rothstein)
Date: Fri Feb 27 08:53:27 MST 2004
Model: CAM
Version: CAM2.0_2.dev67
One-line Summary: Implemented co2 ramping; Enabled lf95/debug checking with sulfates on; fix for 200MB radiation diagnostics
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI(pending), Compaq
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, eaton
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
M models/atm/cam/src/control/runtime_opts.F90
- added new option to namelist var scenario_ghg: RAMP_CO2_ONLY
- added new namelist variables: ramp_co2_annual_rate, ramp_co2_cap,
ramp_co2_start_ymd
M models/atm/cam/src/control/time_manager.F90
- added new routine to calculate #of days between dates: timemgr_datediff
M models/atm/cam/src/physics/cam1/acbnd.F90
- mods to enable lf95/debug checking with sulfates on
M models/atm/cam/src/physics/cam1/dmsbnd.F90
- mods to enable lf95/debug checking with sulfates on
M models/atm/cam/src/physics/cam1/soxbnd.F90
- mods to enable lf95/debug checking with sulfates on
M models/atm/cam/src/physics/cam1/ghg_surfvals.F90
- formally implemented ability to ramp co2
M models/atm/cam/src/physics/cam1/radcswmx.F90
- added fix for 200MB radiation diagnostics
- uses new function to retrieve co2 mass mixing ratio
M models/atm/cam/src/physics/cam1/trcpth.F90
- uses new function to retrieve co2 mass mixing ratio
===============================================================
===============================================================
cam2_0_2_dev66
Originator: eaton ( Brian Eaton)
Date: Mon Feb 23 16:40:39 MST 2004
Model: CAM
Version: CAM2.0_2.dev66
One-line Summary: misc performance improvements
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: Yes (performance improvement validated by G. Carr)
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, PC
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Carr, Edwards
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
>> models/atm/cam/src/dynamics/eul/herxin.F90
>> . "rolled" a loop for better cache performance (Edwards)
>>
>> models/atm/cam/src/dynamics/eul/limdy.F90
>> models/atm/cam/src/dynamics/eul/limdz.F90
>> . inlined scm0 by hand into limdy and limdz (Carr)
>>
>> models/atm/cam/src/dynamics/eul/sltini.F90
>> . remove SPMD ifdefs and thread a loop that was previously unthreaded
>> when SPMD was defined. (Eaton)
===============================================================
===============================================================
cam2_0_2_dev65
Originator: hender ( Tom Henderson)
Date: Thu Feb 12 16:18:09 MST 2004
Model: CAM
Version: CAM2.0_2.dev65
One-line Summary: IPCC performance upgrade in prognostic sulfate interpolation
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Intel-Linux
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Jim McCaa, Brian Eaton, Jim Rosinski, Mathew Rothstein
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: Restored signalling Nan except for pgf90
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
>> Performance fix to scale computations in linintp. Speedup = ~12% for
>> stand-alone CAM at T85 with IPCC scenario configurations for prognostic
>> sulfates and greenhouse gasses:
>> models/atm/cam/bld/run-ibm.csh
>> models/atm/cam/src/physics/cam1/acbnd.F90
>> models/atm/cam/src/physics/cam1/dmsbnd.F90
>> models/atm/cam/src/physics/cam1/soxbnd.F90
>> models/atm/cam/src/physics/cam1/sulchem.F90
>> models/atm/cam/src/physics/cam1/sulemis.F90
>> models/atm/cam/src/physics/cam1/sulfur_intr.F90
>> models/atm/cam/src/control/error_messages.F90
>> Jim Rosinski's new timers:
>> models/atm/cam/src/physics/cam1/comsrf.F90
>> models/atm/cam/src/physics/cam1/convtran.F90
>> models/atm/cam/src/physics/cam1/prescribed_aerosols.F90
>> models/atm/cam/src/physics/cam1/tphysbc.F90
>> models/atm/cam/src/control/intp_util.F90
>> Brian Eaton's change to remove broadcasts from chemistry.F90 that were
>> happening every time step:
>> models/atm/cam/bld/DefaultCAMEXPNamelist.xml
>> models/atm/cam/bld/DefaultCLMEXPNamelist.xml
>> models/atm/cam/src/physics/cam1/chemistry.F90
>> Mathew Rothstein's change to use quiet NaN for pgf90 and signalling NaN for
>> all other compilers.
>> models/lnd/clm2/src/main/nanMod.F90
>> models/atmlnd_share/infnan.F90
>> Jim McCaa's addition of 200mb fields.
>> models/atm/cam/src/physics/cam1/radclwmx.F90
>> models/atm/cam/src/physics/cam1/radcswmx.F90
>> models/atm/cam/src/physics/cam1/radctl.F90
>> models/atm/cam/src/control/history.F90
>> By default, this change produces no new output fields; however, it allows
>> one to fincl the following:
>> FSN200: Clearsky shortwave flux at 200 mb
>> FSN200C: Shortwave flux at 200 mb
>> FLN200: Clearsky longwave flux at 200 mb
>> FLN200C: Longwave flux at 200 mb
===============================================================
===============================================================
cam2_0_2_dev64
Originator: mvr ( Mathew Rothstein)
Date: Mon Feb 2 11:13:05 MST 2004
Model: CAM
Version: CAM2.0_2.dev64
One-line Summary: patch to enable runs using portland group compiler; removal of mprun2d.pm
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Compaq
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: yes
- modified definition of parameter nan to enable runs using portland group compiler
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
M cam1/models/lnd/clm2/src/main/nanMod.F90
- modified definition of parameter nan to enable run using portland
group comiler
R cam1/models/atm/cam/bld/mprun2d.pm
===============================================================
===============================================================
cam2_0_2_dev63
Originator: rosinski ( Jim Rosinski)
Date: Tue Jan 27 10:15:42 MST 2004
Model: CAM
Version: CAM2.0_2.dev63
One-line Summary: Eliminate dead code, more flexible endrun, more "intent" attributes,
better lf95 strict error checking compliancy
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: Mariana checked this.
Machines tested: IBM, SGI, PC
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Vertenstein
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None (one-line description)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
o endrun is now in a module and takes an optional character string argument.
If present, this string is printed before the model aborts.
o Eliminated dead code, including the old "dataicemodel". Mods were made to
ncar_batch.csh to eliminate the dataicemodel tests. This was done by passing
-skip all:1-17 to test-model.
o Also eliminated dead logic related to sulfur variables.
o Better strict error checking compliancy under lf95. But "lf95 -x" cannot be
used because lf95 warns that all underlying libraries must also be compiled
with this option in order for it to work properly.
o Added "intent" clause to arguments where the proper value was clear. The code
was already in pretty good shape w.r.t. all arguments having their intent
declared. There are still a few places where intent needs to be added though.
===============================================================
===============================================================
cam2_0_2_dev62
Originator: hender ( Tom Henderson)
Date: Fri Jan 23 10:49:42 MST 2004
Model: CAM
Version: CAM2.0_2.dev62
One-line Summary: IPCC scenario changes
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: yes, new namelist variables, new data files
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, PC-Linux
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Conley, Bundy, Rosinski
Restart files change: no
Changes answers: no (bit-for-bit) for runs that do not use prognostic sulfates,
Yes otherwise
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
1. Complete documentation of the changes, including:
a. A list of any subroutines eliminated;
>> cam1/models/atm/cam/src/physics/cam1/Population.F90
>> cam1/models/atm/cam/src/physics/cam1/aerosols.F90
>> cam1/models/atm/cam/src/physics/cam1/ramp_ghg_bau.h
>> cam1/models/atm/cam/src/physics/cam1/ramp_ghg_stab.h
>> cam1/models/atm/cam/src/physics/cam1/ramp_scon.h
b. A list of any subroutines added and what they do; and
>> cam1/models/atm/cam/src/physics/cam1/carbonscales.F90
>> Formerly Population.F90
>> cam1/models/atm/cam/src/physics/cam1/prescribed_aerosols.F90
>> Formerly aerosols.F90
>> cam1/models/atm/cam/src/utils/timeinterp.F90
>> Combined repeated code bits into a new subroutine.
c. For existing files that have been modified, an itemized list
of the changes and where they can be found
(NOTE: A cvs command that generates all of the above information can be
provided in place of individual answers. For example, "cvs -nq
update ..." or "cvs rdiff ...".)
>> cam1/models/atm/cam/bld/DefaultCAMEXPNamelist.xml
>> cam1/models/atm/cam/bld/camexp.pm
>> cam1/models/atm/cam/bld/run-model.pl
>> cam1/models/atm/cam/src/physics/cam1/advnce.F90
>> cam1/models/atm/cam/src/physics/cam1/aerosol_intr.F90
>> cam1/models/atm/cam/src/physics/cam1/chemistry.F90
>> cam1/models/atm/cam/src/physics/cam1/dmsbnd.F90
>> cam1/models/atm/cam/src/physics/cam1/ghg_surfvals.F90
>> cam1/models/atm/cam/src/physics/cam1/inti.F90
>> cam1/models/atm/cam/src/physics/cam1/radae.F90
>> cam1/models/atm/cam/src/physics/cam1/radcswmx.F90
>> cam1/models/atm/cam/src/physics/cam1/radctl.F90
>> cam1/models/atm/cam/src/physics/cam1/ramp.h
>> cam1/models/atm/cam/src/physics/cam1/ramp_scon.F90
>> cam1/models/atm/cam/src/physics/cam1/ramp_so4_mod.F90
>> cam1/models/atm/cam/src/physics/cam1/soxbnd.F90
>> cam1/models/atm/cam/src/physics/cam1/sulfur_intr.F90
>> cam1/models/atm/cam/src/physics/cam1/volcanicmass.F90
>> cam1/models/atm/cam/src/physics/cam1/volcrad.F90
>> cam1/models/atm/cam/src/control/filenames.F90
>> cam1/models/atm/cam/src/control/oznint.F90
>> cam1/models/atm/cam/src/control/runtime_opts.F90
>> cam1/models/atm/cam/src/control/so4bnd.F90
>> cam1/models/atm/cam/src/ocnsice/dom/sst_data.F90
>> cam1/models/atm/cam/src/ocnsice/som/somint.F90
>> cam1/models/lnd/clm2/src/main/lp_coupling.F90
>>
>> See below for details of changes.
2. Complete documentation of these changes in the code by comments.
>> DONE
3. On what machines did you successfully run test-model.pl?
a. blackforest
>> YES
b. chinook
>> YES
c. anchorage-lf95
>> YES
4. For all of the runs in #3 above, which version of cam did you
compare to (which tag)?
>> cam2_0_2_dev61, using hand-tests since test-model does not support
>> "-compare" when namelist has changed.
5. If bitwise differences were observed, how did you prove they were no
worse than roundoff?
>> bit-for-bit
6. Does the CAM Users Guide need to be updated due to these changes?
(Possible reasons include changes to the namelist, history output,
configure script, build or run process, etc.)
>> Yes
If so, has the CAM Users Guide been updated in the CVS repository?
>> No
>>
>> Source: http://www.cgd.ucar.edu/cseg/plans/ipcc_plan2.txt 01/23/04
>>
>> IPCC Features in CAM
>>
>> Requirements
>> . thoroughly tested
>> . releaseable, reproducible, community usable
>> . usable for 1990, 1870, historicals, future, paleo
>> (without changing source code, only namelist?)
>> . no hacks, no magic dates in code
>> . datasets available to support T85 resolution (T42?, T31?)
>> . implementation must be general, namelist and dataset must be
>> robust for the specified use cases below. that means making
>> sure that 1870-2000 datasets can be used robustly in the model
>> on Jan 1, 1870 and Jan 1, 2000, and that future datasets can
>> be used in the model robustly on Jan 1, 1990.
>>
>> -------------------------------
>>
>> Present 1870 control run mods relative to 1990 controls
>> Add CAM namelists f11vmr, f12vmr, co2vmr, ch4vmr, n2ovmr
>> Add CAM namelist carscl=0.3
>> Add CAM namelist sulscl=0.0
>>
>> -------------------------------
>>
>> Dust, Sea Salt
>> . seasonal cycle of monthly averages, data(x,y,z,12)
>> . dataset is resolution dependent, requires T42 and T85 datasets
>> . same datasets will be used for all runs
>> . can be scaled, scaling will always be 1
>> . this has always been on
>> . file name AerosolMass*
>> . file also includes carbon which is also always on
>> Tasks
>> + Review dataset, verify supports all resolutions; T85, T42, T31, 2x2.5
>> (Andrew, Bill, Jim R)
>> - Verify correct datasets in CAM scripts (Andrew, Tom)
>> + Implement changes to CCSM scripts as needed (Tony, Mariana)
>> Namelist
>> bndtvaer
>>
>> Carbon
>> . carbon dataset is in same file as Dust, Sea Salt, same format (x,y,z,12)
>> . in file AerosolMass
>> . always on, carscl can scale, default 1.0
>> . carscl=0.3 for 1870 run, continuous with pop dataset
>> . historical will be based on population dataset (t)
>> . time interpolation is flexible, can have varying dt in dataset
>> . dataset "easy to generate"
>> . endruns if date outside dataset
>> . dataset will cover 1870 to 2000
>> . 2000 to 2300 constant carscl at 2000 (1.0)
>> . 1870 and future will use carscl, historical will use dataset
>> . "ramp year" namelist implementation is defered
>> Tasks
>> + Rename carbon_by_pop to scenario_carbon_scale, char string input
>> (Andrew)
>> + Rename bndtvpop to bndtvcarbonscale (Andrew)
>> + Generate a 1870-2000 dataset (Andrew)
>> + Verify carbon will call endrun if date is outside dataset boundaries
>> (Andrew)
>> - Verify correct datasets and namelist in CAM scripts (Andrew, Tom)
>> + Implement changes to CCSM scripts as needed (Tony, Mariana)
>> Namelist
>> bndtvaer
>> carscl
>> scenario_carbon_scale (RAMPED, FIXED), ramped will turn on use of
>> dataset, fixed
>> means use carscl.
>> bndtvcarbonscale
>>
>> Solar
>> . normally a constant, namelist (scon) can override default
>> . can turn on ramping via scenario_scon
>> . ramping data is implemented as data statement and include file
>> . 1870-2000 dataset
>> . all runs with using ramping feature, 1870 will use rampyear_scon
>> set to 1870, future runs will use rampyear_scon set to "2005"
>> Tasks
>> - Create solar dataset, end solar cycle in middle of 11 year cycle
>> around 2005, verify dataset, (Lawrence, Caspar)
>> + Remove include file and data statement, add read of netcdf dataset (Tom)
>> - Move datasets to inputdata and verify naming convention (Lawrence, Tom)
>> + Verify solar calls endrun if using ramped feature and date is outside
>> dataset range (Tom)
>> + Implement in source code ability to use new namelist convention
>> (bndtvscon) (Tom)
>> - Verify correct datasets and namelist in CAM scripts (Lawrence, Tom)
>> + Implement changes to CCSM scripts as needed (Tony, Mariana, Lawrence)
>> - Verify functionality of scenario and rampYear via code review
>> or validation runs (Tom, Lawrence)
>> Namelist
>> scon (won't be used)
>> scenario_scon (RAMPED)
>> bndtvscon
>> rampYear_scon
>>
>> GHGs
>> . 5 tracers, f11, f12, co2, ch4, n2o
>> . normally a constant, namelist can override the constants
>> . can turn on ramping
>> . ramping data is implemented as data statement and include file,
>> . turns on some ghg chemistry, IPCC will use that feature, automatically
>> turned on when ramped feature is used
>> . future scenarios will be time varying
>> . Use scenario_ghg = RAMPED always
>> . set rampYear_ghg to 1870 for 1870 control
>> . there will be multiple datasets, 1870-2000 for historical, several
>> 1990 to future datasets.
>> . requires pcnst=pcnst+4
>> Tasks
>> + Talk to Byron or Jeff about ghg chemistry (Bill)
>> + Generate 1870-2000 netcdf ghg dataset (Lawrence)
>> - Generate 1990-future netcdf ghg datasets (Lawrence, Gary)
>> + Remove include file and data statement, add a dataset (Tom)
>> + Verify ghg code calls endrun if year outside dataset for ramped
>> usage (Tom)
>> + Implement in source code ability to use new namelist convention
>> (bndtvghg) (Tom)
>> - Move datasets to inputdata and verify naming convention (Lawrence, Tom)
>> - Verify correct datasets and namelist in CAM scripts (Lawrence, Tom)
>> + Implement changes to CCSM scripts as needed (Tony, Mariana, Lawrence)
>> Namelist
>> f11vmr (won't be used)
>> f12vmr (won't be used)
>> co2vmr (won't be used)
>> ch4vmr (won't be used)
>> n2ovmr (won't be used)
>> scenario_ghg (RAMPED)
>> rampYear_ghg
>> bndtvghg
>> trace_gas
>> read_trace
>>
>> Volcanics
>> . one field, mvolc(x,y,z,t), 1 sample per month
>> . flexible time interpolation, won't handle years outside dataset bounds
>> . turned off by default
>> . should be on for historicals only, datafile for 1870-2000
>> . will do spatial interpolation
>> Tasks
>> + Generate 1870-2000 dataset and review for historical runs, make
>> sure volcanics go to 0.0 at edges of dataset, make sure model
>> calls endrun if strat_volc=true and date is outside bounds of
>> dataset (Andrew, Caspar)
>> + Move datasets to inputdata and verify naming convention (Andrew)
>> - Verify correct datasets and namelist in CAM scripts (Andrew, Tom)
>> + Implement changes to CCSM scripts as needed (Tony, Mariana)
>> Namelist
>> strat_volc = .true.
>> bndtvvolc = 'dataset'
>>
>> Sulfate
>> . three different tracers in three datasets, dms, oxidants, sox
>> . dms is f(x,y,t), seasonal cycle with 12 time samples
>> . oxidants is f(x,y,z,t) seasonal cycle with 12 time samples
>> . dms and oxidants are seasonally repeating, not a function of "year"
>> . sox(x,y,2,t) is 4 samples/year, varying over different years
>> . sox time interpolation works fine, no extrapolation currently
>> . dataset must be generated on CAM grid
>> . REQUIRES pcnst = pcnst+4 (not 3)
>> . historical and future will use ramped feature with time varying data
>> . 1870 run will use ramped feature with rampyear_prognostic_sulfur set to
>> 1870
>> . some sort of huge(int) intrinsic value needs to be default for rampyear
>> Tasks
>> + Design new namelist, prescribed_sulfur and prognostic_sulfur
>> and usage/features (Bill, Andrew, Phil)
>> + Remove dead namelist, aero_sulfur, aero_feedback_sulfur, implement new
>> namelist and features (see below) (Andrew and Dani)
>> + Remove (3) old dataset namelist (dms_emis, oxid, sox_emis) and add
>> (3) new dataset namelists (see below) (Dani)
>> + Verify endrun is called if date outside sox datasets (Dani)
>> + Generate 1870-2000 sox dataset for T85 (Dani, Gary)
>> + Generate future datasets for sulfur (Dani, Lawrence, Gary)
>> + Move all datasets to inputdata and verify naming convention
>> (Andrew, Dani)
>> - Verify correct datasets and namelist in CAM scripts (Andrew, Dani, Tom)
>> + Implement changes to CCSM scripts as needed (Tony, Mariana)
>> Namelist
>> bndtvdms
>> bndtvoxid
>> bndtvsox
>> scenario_prescribed_sulfur (FIXED (DEFAULT), RAMPED)
>> rampyear_prescribed_sulfur
>> prescribed_sulfur (off, passive, direct (DEFAULT))
>> scenario_prognostic_sulfur (FIXED (DEFAULT), RAMPED)
>> rampyear_prognostic_sulfur
>> prognostic_sulfur (off (DEFAULT), passive, direct)
>>
>> Ozone
>> . ozone function of (x,y,z,t) with seasonal cycle
>> . use T42 dataset with monthly data, 1870-2000, (850Mb)
>> . ozone dataset not CAM grid dependent, does space interpolation
>> . for 1870, use 1870 only dataset with cycling set to true
>> . for historical, use 1870-2000 with cycling set to false
>> . for future, use 2000 dataset with cycling set to true
>> . tropopause consistency issue
>> . defer namelist changes
>> Tasks
>> - Generate 1870-2000 monthly dataset on T42, spoof to 1870 (Lawrence)
>> - Generate 1870 and 2000 cycling datasets (Lawrence)
>> - Create, review, and validate datasets, dataset continuity between
>> 1870 cycling dataset, 1870-2000 dataset, 2000 cycling dataset (Lawrence)
>> - Move datasets to inputdata and verify naming convention (Lawrence, Tom)
>> - Verify correct datasets and namelist in CAM scripts (Tom)
>> + Implement changes to CCSM scripts as needed (Tony, Mariana)
>> Namelist
>> bndtvo
>> ozncyc
>>
>> -------------------------------
>>
>> Tony will monitor overall task progress
>> Tom will collect code mods and CAM script mods and coordinate checkin
>> Tony will coordinate changes in CCSM scripts
>>
>> -------------------------------
>>
>> Use Cases:
>>
>> 1990 control - runs out of the box without above turned on,
>> this is not the same as historical run at 1990 or 1990 control
>> done similar to 1870. use pcnst = 3
>>
>> (no source code mods required to switch between 1870, 1990,
>> historical, future cases below, pcnst = 11 for all cases)
>>
>> 1870 control, pcnst = 11 in build, standard namelist plus,
>> X means not needed or current default inside model (included
>> for clarity)
>>
>> bndtvaer = AerosolMass_V_128x256_clim_c031022.nc
>> X scenario_carbon_scale = 'FIXED'
>> carscl = 0.3
>> X bndtvcarbonscale = carbonscaling_1870-1999_c040114.nc
>> scenario_scon = 'RAMPED'
>> rampYear_scon = 1870
>> bndtvscon = *.nc (1870-2005 dataset TBD)
>> scenario_ghg = 'RAMPED'
>> rampYear_ghg = 1870
>> bndtvghg = ghg_1870_2100_c040122.nc
>> trace_gas = .true.
>> read_trace = .false.
>> X strat_volcanic = .false.
>> X bndtvvolc = VolcanicMass_1870-1999_64x1_L18_c040115.nc
>> bndtvdms = DMS_emissions_128x256_clim_c040122.nc
>> bndtvoxid = oxid_128x256_L26_clim_c040112.nc
>> bndtvsox = SOx_emissions_128x256_L2_c040109.nc
>> scenario_prescribed_sulfur = 'FIXED'
>> prescribed_sulfur = 'passive'
>> scenario_prognostic_sulfur = 'RAMPED'
>> rampYear_prognostic_sulfur = 1870
>> prognostic_sulfur = 'direct'
>> X ozncyc = .true.
>> bndtvo = *.nc (1870 dataset)
>>
>> historical (1870-2000), hybrid startup at year 1870, run to 2000
>> pcnst = 11 in build, set namelist as follows, X means not needed or
>> current default, * indicate changes from 1870 control
>>
>> bndtvaer = AerosolMass_V_128x256_clim_c031022.nc
>> * scenario_carbon_scale = 'RAMPED'
>> *X carscl (remove)
>> bndtvcarbonscale = carbonscaling_1870-1999_c040114.nc
>> scenario_scon = 'RAMPED'
>> *X rampYear_scon (remove)
>> bndtvscon = *.nc (1870-2005 dataset TBD)
>> scenario_ghg = 'RAMPED'
>> *X rampYear_ghg (remove)
>> bndtvghg = ghg_1870_2100_c040122.nc
>> trace_gas = .true.
>> * strat_volcanic = .true.
>> bndtvvolc = VolcanicMass_1870-1999_64x1_L18_c040115.nc
>> bndtvdms = DMS_emissions_128x256_clim_c040122.nc
>> bndtvoxid = oxid_128x256_L26_clim_c040112.nc
>> bndtvsox = SOx_emissions_128x256_L2_c040109.nc
>> scenario_prescribed_sulfur = 'FIXED'
>> prescribed_sulfur = 'passive'
>> scenario_prognostic_sulfur = 'RAMPED'
>> * rampYear_prognostic_sulfur = (remove)
>> prognostic_sulfur = 'direct'
>> * ozncyc = .false.
>> * bndtvo = *.nc (1870-2000 dataset TBD)
>>
>> future (1990-), hybrid startup at year 1990 from historical 1990,
>> pcnst = 11 in build, set namelist as follows, X means not needed or
>> current default, * indicate change from historical run
>>
>> bndtvaer = AerosolMass_V_128x256_clim_c031022.nc
>> *X scenario_carbon_scale = 'FIXED'
>> *X carscl = 1.0
>> X bndtvcarbonscale = carbonscaling_1870-1999_c040114.nc
>> scenario_scon = 'RAMPED'
>> * rampYear_scon = 2005
>> bndtvscon = *.nc (1870-2006 dataset TBD)
>> scenario_ghg = 'RAMPED'
>> X rampYear_ghg (remove)
>> * bndtvghg = *.nc (future dataset TBD)
>> trace_gas = .true.
>> *X strat_volcanic = .false.
>> X bndtvvolc = VolcanicMass_1870-1999_64x1_L18_c040115.nc
>> bndtvdms = DMS_emissions_128x256_clim_c040122.nc
>> bndtvoxid = oxid_128x256_L26_clim_c040112.nc
>> bndtvsox = SOX_T85_future
>> scenario_prescribed_sulfur = 'FIXED'
>> prescribed_sulfur = 'passive'
>> scenario_prognostic_sulfur = 'RAMPED'
>> prognostic_sulfur = 'direct'
>> *X ozncyc = .true.
>> * bndtvo = *.nc (2000 dataset TBD)
>>
>> ---------------------------------
>>
>> Available (and validated) Datasets:
>>
>> bndtvaer: (rad dir)
>> AerosolMass_V_128x256_clim_c031022.nc T85
>> AerosolMass_V_64x128_clim_c031022.nc T42
>> AerosolMass_V_48x96_clim_c031022.nc T31
>> AerosolMass_V_2x2.5_clim_c031022.nc 2x2.5
>>
>> bndtvdms: (scyc dir)
>> DMS_emissions_64x128_c030722.nc T42
>> DMS_emissions_128x256_clim_c040122.nc T85
>>
>> bndtvoxid: (scyc dir)
>> oxid_3d_64x128_L26_c030722.nc T42
>> oxid_128x256_L26_clim_c040112.nc T85
>>
>> bndtvsox: (scyc dir)
>> SOx_emissions_64x128_L2_c031219.nc T42 (1870+historical)
>> - T42 (A1)
>> - T42 (A2)
>> - T42 (B1)
>> - T42 (B2)
>> SOx_emissions_128x256_L2_c040109.nc T85 (1870+historical)
>> SOx_emissions_A1_128x256_L26_1990-2100_c030121.nc T85 (A1)
>> SOx_emissions_A2_128x256_L26_1990-2100_c030121.nc T85 (A2)
>> SOx_emissions_B1_128x256_L26_1990-2100_c030121.nc T85 (B1)
>> SOx_emissions_B2_128x256_L26_1990-2100_c030121.nc T85 (B2)
>>
>> bndtvo: (ozone dir)
>> pcmdio3.r8.64x1_L60_clim_c970515.nc (present day cycling
>> dataset)
>> - (1870 cycling)
>> - (1870-2000)
>> - (2000 cycling)
>>
>> bndtvvolc: (rad dir)
>> VolcanicMass_1870-1999_64x1_L18_c040115.nc (1870-2000 dataset)
>>
>> bndtvcarbonscale: (rad dir)
>> carbonscaling_1870-1999_c040114.nc (1870-2000 dataset)
>>
>> bndtvscon: (rad dir)
>> scon_1870_2100_c040122.nc (CAM default, do not
>> use)
>> - (1870-2006 dataset)
>>
>> bndtvghg: (ggas dir)
>> ghg_1870_2100_c040122.nc (CAM default,
>> 1870+historical)
>> - (A1)
>> - (A2)
>> - (B1)
>> - (B2)
===============================================================
===============================================================
cam2_0_2_dev61
Originator: mirin ( Arthur Andrew Mirin)
Date: Wed Jan 14 11:51:56 MST 2004
Model: CAM
Version: CAM2.0_2.dev61
One-line Summary: Finite Volume and Transpose Improvements (2 of 2)
This is the second of two successive archive updates. The only changed
routine is dynamics/fv/geopk.F90. Due to reordering of computations,
there are roundoff differences.
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
(one-line description of changes: Makefile, include files, directories, etc.)
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Compaq
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Sawyer
Restart files change: no
Changes answers: Yes (same to roundoff for FV), EUL and SLD are bit-for-bit.
(Note: test-model.pl only does checks this -- if you run it with the "-compare" option)
Changes to CLM land-model: None (one-line description)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
(If not abort this tag and follow the procedures in)
http://www.cgd.ucar.edu/~cam/cam_checkins/checkin_process.shtml
Changes made:
Reordering of computations in FV geopk.F90 causes roundoff changes.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam2_0_2_dev60
Originator: mirin ( Arthur Andrew Mirin)
Date: Wed Jan 14 11:39:22 MST 2004
Model: CAM
Version: CAM2.0_2.dev60
One-line Summary: Finite Volume and transpose improvements (1 of 2)
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: yes
Mprun2d namelist no longer exists; relevant variables in CAMEXP namelist.
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Compaq
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Sawyer, Eaton
Restart files change: no
Changes answers: no (bit-for-bit)
(Note: test-model.pl only does checks this -- if you run it with the "-compare" option)
Changes to CLM land-model: None (one-line description)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
This is the first of two successive archive updates. These changes are
bit-fot-bit.
Summary of changes:
Removal of mprun2d namelist. This involved creating routines
spmd_dyn_defaultopts and spmd_dyn_setopts with optional arguments. I
mimicked the procedure in phys_grid.F90. In runtime_opts.F90 I put
"dycore_is" test around the calls, and I created dummy versions in
dynamics/eul and dynamics/sld. The initial dummy versions were just
subroutine and end statements, with no arguments. However, the code would
not compile; there were complaints about inaccessible routines and
needing explicit interfaces with optional arguments. I ended up putting
'public' declarations in the EUL and SLD versions and using the same
argument list as with the FV version. (I think the key problem was the
argument list more than the public declaration.) I know it is our
intention to have different versions (with different arguments) of these
routines for the different dycores, but that clearly didn't work; the
present solution is out of expediency. I suspect that the fact that the
arguments are optional is central to the problem. I used optional
arguments only because that is what was done in phys_grid.F90. Also, one
of the namelist variables (modcomm_method) will eventually apply outside
of FV once we use mod_comm transposes outside of FV, so it will
eventually have to be moved or replicated.
Single coding path for 1-D and 2-D decompositions for FV dycore. TWOD_YZ
ifdef eliminated. The "xy" variables are always invoked (e.g., ptxy). The
twod_decomp variable now refers to whether or not transposes (versus
copies) are called to go between yz and xy decompositions; naturally
copies are relevant only for 1-D decomposition; for debugging purposes,
one can force transposes even with 1-D decomposition (see new namelist
variable force_2d). Also, there is a new FV runtime diagnostic (not
input) variable (called spmd_on) that indicates SPMD versus non-SPMD.
Method for FV transposes now runtime rather than compile time (although
the choice between ordinary MPI and MPI-2 is still compile-time). For
ordinary MPI, the choice is between invoking temporary contiguous buffers
versus using MPI derived types. The relevant namelist variable is
modcomm_method (0 for contiguous buffers, 1 for derived types (default)).
MPI-2 presently has 4 options (0,1(default),2,3) (options 0,1,2 use
contiguous buffer at target; 0 is for contiguous source buffers; 1 is for
direct MPI_put's of contiguous segments, with threading over the
segments; 2 is for MPI derived types at the source with threading over
target; option 3 is for MPI derived types at source and target, with
threading over target; option 3 involves re-definition of windows and is
probably not optimal). For MPI-2, one specifies "USE_MPI2" precompile
flag when configuring.
New configure option "-precomp ", where string is a non-empty
list of precompile flags to be placed in misc.h. For example, one can
specify "-precomp USE_MPI2" for MPI-2 mod_comm transposes.
Option to use nested OpenMP constructs for FV dycore on IBM. This must be
specified at compile time using new configure option "-nested_omp"; this
has effect on misc.h and the Makefile. In sw_core.F90 and tp_core.F90
(which are already threaded with respect to z), latitude loops are
threaded. The number of inner threads is the new namelist variable
"ompnest", which must divide evenly into the total number of threads; the
number of outer threads will be the total number of threads divided by
"ompnest". IBM's implementation is non-standard.
Option to use mod_comm for dynamics/physics transposes when using
load-balanced chunking. In phys_grid.F90, the "alltoall" flag is replaced
by new namelist variable "phys_transpose" (1 for original alltoall
(default), 2 for Pat Worley's send_and_receive, 3 for mod_comm). Because
mod_comm is presently compiled and linked only for FV, the mod_comm
option is presently valid only for FV; it must be activated at compile
time with the MODCM_DP_TRANSPOSE if-def (use precomp option in
configure).
Miscellaneous transpose improvements and code cleaning.
Deletion of configure_fv.pl and script.m4.
I also created the necessary surface data and aerosol files to support
the FV 0.5x0.625 resolution. The CAM and CLM default namelists were
updated accordingly. This required updating
tools/interpaerosols/REGRID.pl as well.
*****************************************************************
With the removal of the TWOD_YZ if-def and mprun2d namelist, the
interactive configuration procedure no longer queries for the 2-D
decompositions, and test-model contains no references to 2-D
decompositions. The validity of 2-D decompositions is checked in
spmd_dyn.F90. To impart 2-D decompositions in test-model, one needs to
set the camexp namelist accordingly. A decision needs to be made on which
decompositions are to be used, and the necessary changes must be made in
the namelist specification.
*****************************************************************
Changed files:
utils/pilgrim: mod_comm.F90, mod_irreg.F90, parutilitiesmodule.F90
bld: CAM_config.pm, CAM_namelist.pm, CAM_run.pm, configure, Makefile,
DefaultCAMEXPNamelist.xml, DefaultCLMEXPNamelist.xml,
config_cache_defaults.xml, mprun2d.pm (deleted), configure_fv.pl
(deleted), script.m4 (deleted)
test/system: CAM_test.pm, test-model.pl
tools/interpaerosols: REGRID.pl
src/physics/cam1: phys_grid.F90
src/control: cam.F90, history.F90, runtime_opts.F90
src/dynamics/fv: cd_core.F90, d2a3dijk.F90, d2a3dikj.F90,
dp_coupling.F90, dyn_grid.F90, dynamics_vars.F90, dynpkg.F90, geopk.F90,
pmgrid.F90, spmd_dyn.F90, stepon.F90, sw_core.F90, te_map.F90,
tp_core.F90, trac2d.F90, uv3s_update.F90
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam2_0_2_dev59
Originator: mvertens ( Mariana Vertenstein)
Date: Thu Jan 8 13:52:37 MST 2004
Model: CAM
Version: CAM2.0_2.dev59
One-line Summary: updated clm code to clm3_deva_01 (vector version of clm)
Requires change in build system: no (see below)
ecosysdyn/ is no longer needed in clm Filepath (will not
trigger error if it is there)
Substantial timing or memory changes: Yes (describe)
3-5% performance improvement should be obtained
Requires change in run script: no
Tested to work coupled with CCSM: yes
Machines tested: IBM, SGI, Linux
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Forrest Hoffman, Dave Parks, Keith Oleson
Restart files change: yes (for clm)
Backwards compatilibity with older restart files have been incorporated
Changes answers: Yes (same-climate)
Changes to CLM land-model: Yes (see below)
Have you filled out the pre-check-in documentation as required by the CAM CRB? no, I plan to.
Changes made:
1) Updated the clm code to clm3_dev_01 (the vector version of clm).
Both offline and cam-clm stand-alone runs showed that the new code
resulted in the same climate. For the cam-clm validation diagnostic
results see:
http://www.cgd.ucar.edu/tss/clm/diagnostics/cam_clm2/cam202d39clm2d47a-cam202d39clm2d36a/sets.htm
2) Modified models/atm/cam/bld/configure to removed ecosysdyn/ from
the clm filepath.
===============================================================
===============================================================
cam2_0_2_dev58
Originator: aconley ( Andrew J. Conley)
Date: Wed Jan 7 13:53:42 MST 2004
Model: CAM
Version: CAM2.0_2.dev58
One-line Summary: added volcanic aerosols and population based carbonaceous aerosol scaling
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Compaq
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: Yes (same-to-roundoff, same-physics)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? no, I plan to.
Changes made:
New documentation:
UG-20.html
UG-20a.html
UG-27.html
UG-45.html
table_of_contents.html
addition of filenames for volcanic mass, new optics, and population data
as well as namelist variables for strat_volcanic and carbon_by_pop
filenames.F90
DefaultCAMEXPNamelist.xml
camexp.pm
runtime_opts.F90
advnce.F90 initialization for volcanics and population
aer_optics.F90 added volcanic optics
aerosols.F90 added massfields for volcanics
volcanicmass.F90 (new file) read, interpolate volcanic masses
radctl.F90 added diagnostics for volcanic masses and effects of volcanics
radcswmx.F90 added volcanic mass shortwave effects
Added effects of volcanics
radclwmx.F90
radae.F90
trcab.F90
trcabn.F90
trcems.F90
volcrad.F90 (new file)
Population.F90 (new file) read population history and provide carbonaceous aerosol scaling factor based on population
New data files
bndtvvolc: historical masses from volcanic eruptions
bndtvpop: estimates of global total population
appended new volcanic shortwave optics to aeroptics
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam2_0_2_dev57
Originator: mvr ( Mathew Rothstein)
Date: Fri Jan 2 12:03:34 MST 2004
Model: CAM
Version: CAM2.0_2.dev57
One-line Summary: Upgrade CSM share; Upgrade ESMF; Add namelist option brnch_retain_casename; Update ccsm_msg to remove chunks and knuhcs; caseid now length 80; Remove goto from aerosols.F90.
cam-bugs Requests resolved: none
Requires change in build system: yes
-references to directory csm_share now point to csm_share/shr
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: yes
Machines tested: IBM, SGI, Linux (Lahey)
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, tcraig, eaton
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
- Updated ESMF to ESMF0_0_4
- Updated CSM_SHARE to share3_0_2
- added function get_chunk_owner_p in src/physics/cam1/phys_grid.F90: returns
the owner of a chunk given the lat, lon
- modified bld/configure: references to directory csm_share now point to
csm_share/shr
- modified bld/configure_fv.pl: references to directory csm_share now point
to csm_share/shr
- modified src/control/ccsm_msg.F90: utilize new function get_chunk_owner_p
rather than vars chunks, knuhcs
- modified src/control/filenames.F90: modified length of caseid to 80 from 33;
now using shr_kind_cs and shr_kind_cl for strings of length 80 and 256;
defined new namelist var brnch_retain_casename
- modified src/control/restart.F90: modified var tcase to same length as
caseid (shr_kind_cs); implemented new namelist var brnch_retain_casename to
allow branch runs without a change in casename
- modified src/control/runtime_opts.F90: added var brnch_retain_casename to
namelist camexp
- modified src/physics/cam1/phys_grid.F90: added new function
get_chunk_owner_p
- modified src/physics/cam1/aerosols.F90: replaced "goto" statement with code
friendlier to compiler optimization
- updated users guide to reflect new namelist option
===============================================================
===============================================================
cam2_0_2_dev56
Originator: rosinski ( Jim Rosinski)
Date: Mon Dec 29 15:35:44 MST 2003
Model: CAM
Version: CAM2.0_2.dev56
One-line Summary: Enable restarting coupled model from dev51 restart file.
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: yes (done by mvertens)
Machines tested: IBM, SGI, PC
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: yes (in coupled mode they now look like dev51 restart files)
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
o Modified coupled-mode CAM restart file to be identical to dev51.
o NOTE: Standalone and SOM restart files are NOT the same.
===============================================================
===============================================================
cam2_0_2_dev55
Originator: jmccaa ( James McCaa)
Date: Mon Dec 22 14:11:07 MST 2003
Model: CAM
Version: CAM2.0_2.dev55
One-line Summary: Tuning mods for finite volume 2x2.5.
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, linux
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: Yes - new-climate for FV
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Model parameters have been adjusted for runs using the finite volume
dycore. Changes occurred in the following routines:
cldwat.F90
cloud_fraction.F90
gw_drag.F90
vertical_diffusion.F90
moistconvection.F90
zm_conv.F90
The turbulence mountain stress parameterization is now turned off. Also,
the parameters governing gravity wave drag are now the same for FV as for
the spectral dycores.
===============================================================
===============================================================
cam2_0_2_dev54
Originator: pworley ( Patrick H Worley)
Date: Thu Dec 18 09:27:25 MST 2003
Model: CAM
Version: CAM2.0_2.dev54
One-line Summary: Improved existing and added SMP-aware physics load balancing schemes
cam-bugs Requests resolved: none
Requires change in build system: maybe
To exploit alternative EUL and SLD latitude decomposition, need to add -DMIRROR_DECOMP to Makefile
Substantial timing or memory changes: Yes - benchmarks by Rory Kelley indicate that new opt=0 load balancing improves performance
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, PC cluster
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Tom Henderson
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
1) Added Henderson, et al's modification to call spmdinit_dyn after runtime_options,
so that future modifications can use namelist input to decide how to decompose
dynamics data structures.
added: control/decompinit.F90
modified: control/cam.F90, control/spmdinit.F90
2) Added Henderson, et al's modification to move duplicate code (ceil_2, pair)
to a new spmd_util module.
added: control/spmd_utils.F90
modified: dynamics/eul - realloc4.F90, realloc7.F90, scan2.F90, spmd_dyn.F90
dynamics/sld - realloc4.F90, realloc7.F90, scan2.F90, spmd_dyn.F90
physics/cam1 - phys_grid.F90
3) Determined number of SMPs and process/SMP map. For EUL and SLD, equidistributed extra
latitudes (when mod(NLAT/npes) /= 0) over the SMPs. (This is useful when using option 1
load balancing. See below.)
modified: control/spmdinit.F90, dynamics/eul/spmd_dyn.F90, dynamics/sld/spmd_dyn.F90,
dynamics/fv/spmd_dyn.F90
4) For EUL and SLD, added an option to assign latitudes to processes so that southern
hemisphere latitudes and their "mirrors" in the northern hemisphere are assigned
to consecutive processes. This is useful when using load balancing options 1 and 3 (see below).
Currently, this is a compile time option. It will be changed to a runtime option
in the near future. To enable now, the CPP token MIRROR_DECOMP must be added to the
FPPFLAGS in the Makefile.
modified: dynamics/eul - spmd_dyn.F90, bndexch.F90; dynamics/sld - spmd_dyn.F90, bndexch.F90
5) Redesigned load balancing options and implementation. The new options are
-1: each latitude line is a single chunk, same as 1D dynamics decompositions.
(UNCHANGED)
0: split local longitude/latitude blocks into chunks,
while attempting to create load-balanced chunks.
(IMPROVED)
1: split SMP-local longitude/latitude blocks into chunks,
while attempting to create load-balanced chunks.
(NEW)
2: load balance chunks with respect to diurnal and
seaonsal cycles and wth respect to latitude,
and assign chunks to processor
in a way that attempts to minimize communication costs
(UNCHANGED)
3: load balance chunks with respect to diurnal and
seasonal cycles (but not latitude), and assign chunks to
processor in a way that attempts to minimize communication
costs
(NEW)
The implementation was significantly modified. The heuristic for
load balancing is now identical for options 0-3. The only difference
between the options is which processes are allowed to communicate
when load balancing.`
Performance comparisons on the IBM indicate that the new option 0 is an improvement
over the old option 0. The other options (especially option 3) are now
competitive with option 0, but are not noticeably better. These results are for T85L26
on 128 processors of Bluesky, and different results may hold on other systems or
for other problem sizes.
modified: physics/cam1/phys_grid.F90
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam2_0_2_dev53
Originator: hender ( Tom Henderson)
Date: Mon Dec 15 11:55:00 MST 2003
Model: CAM
Version: CAM2.0_2.dev53
One-line Summary: Improve and extend prototype of new run-time options module.
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, anchorage-lf95
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Brian Eaton. Design reviewed by Eaton and Rosinski
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Files removed:
models/atm/cam/src/control/comhd.h
models/atm/cam/src/control/parse_namelist.F90
models/atm/cam/src/control/preset.F90
Files added:
models/atm/cam/src/control/comhd.F90
models/atm/cam/src/control/runtime_opts.F90
Files Modified:
models/atm/cam/src/dynamics/eul/courlim.F90
models/atm/cam/src/dynamics/eul/grcalc.F90
models/atm/cam/src/dynamics/eul/hdinti.F90
models/atm/cam/src/dynamics/eul/hordif.F90
models/atm/cam/src/dynamics/eul/spegrd.F90
models/atm/cam/src/dynamics/sld/courlim.F90
models/atm/cam/src/dynamics/sld/grcalc.F90
models/atm/cam/src/dynamics/sld/hdinti.F90
models/atm/cam/src/dynamics/sld/hordif.F90
models/atm/cam/src/dynamics/sld/spegrd.F90
models/atm/cam/src/physics/cam1/aerosol_intr.F90
models/atm/cam/src/physics/cam1/phys_grid.F90
models/atm/cam/src/control/cam.F90
models/atm/cam/src/control/history.F90
===============================================================
===============================================================
cam2_0_2_dev52
Originator: rosinski ( Jim Rosinski) and John Truesdale
Date: Fri Dec 12 15:56:44 MST 2003
Model: CAM
Version: CAM2.0_2.dev52
One-line Summary: Improve handling of surface fractions, especially in coupled mode (SOM and CCSM)
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: yes
Machines tested: IBM, SGI, PC
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: Rosinski, Truesdale
Restart files change: yes
Changes answers: no (bit-for-bit) except SOM which is roundoff. There is a #define in somoce.F90
that if defined will make SOM bfb. If the "puny" criterion of ice_srf.F90 is
ever met (aice <= puny = 1.e-12) then there is a possibility that prescribed-ice
mode could be good only to roundoff, but we were unable to generate such a case.
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
o Install a more permanent fix to erroneous surface fraction calculations
than the quick fix of dev50. The original bug which dev50 addressed was that
surface fractions were being changed in an incorrect way by calls to
update_srf_fractions in coupled mode (both SOM and CCSM). At various points
during a timestep, variable ICEFRAC contained either ice fraction of the grid
box or ice fraction of the non-land portion of the grid box. Routine
update_srf_fractions implicity assumed it was being called to convert a
non-land fraction into a grid box fraction. In coupled mode,
update_srf_fractions was being called by ccsmrcv every timestep after
receiving surface fraction information from the coupler. Since ICEFRAC held
ice fraction of the grid box at this point, an erroneous calculation thus
resulted. The dev50 quick fix was to install ifdefs around the ICEFRAC
recomputation in update_srf_fractions. In the more permanent fix, array AICE
always represents ice fraction of the non-land portion of the grid box, and
ICEFRAC always represents the ice fraction of the entire grid box.
The updating of surface fractions and verifying their validity is now
segregated into two subroutines, comsrf.F90 routines update_ocnice and
verify_fractions, respectively. ccsm_msg.F90 routines ccsmrcv and
ccsm_msg_getalb now call verify_fractions, instead of a routine which will
actually change the surface fractions. If update_ocnice is ever called in
coupled mode it will result in an endrun() call. Instead, ccsmrcv and
ccsm_msg_getalb now contain explicit code to compute new fractions
(correctly) rather than make calls to update_ocnice. This resetting should
be unnecessary, but was retained to obtain bit-for-bit reproducibility with
earlier libraries and to avoid concerns about restartability. We believe the
calls to update_srf_fractions were originally installed to address bizarre
restartability issues in the coupled model. Ultimately the recomputation of
fractional information (especially landfrac every timestep) should be removed
from these routines. These spots are currently delimited with "bit-for-bit"
comments.
For further details see details on the CAM checkin list.
===============================================================
===============================================================
cam2_0_2_dev51
Originator: mvr ( Mathew Rothstein)
Date: Mon Dec 8 11:29:27 MST 2003
Model: CAM
Version: CAM2.0_2.dev51
One-line Summary: add -Q to Makefile for IBM, added XLSMPOPTS back into run-ibm.csh
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Linux
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, hender
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
>> Makefile:
>> Add "-Q" for ~1% speedup.
>>
>> run-ibm.csh:
>> Add XLSMPOPTS do avoid thread-stack limits that cause seg-fault when
>> prognostic sulfates, carbon, sea salt, and dust are all turned on.
>>
>> ## suggestion from Jim Edwards to reintroduce XLSMPOPTS on 11/13/03
>> setenv XLSMPOPTS "stack=256000000"
===============================================================
===============================================================
cam2_0_2_dev50
Originator: jmccaa ( James Mccaa)
Date: Wed Dec 3 15:35:26 MST 2003
Model: CAM
Version: CAM2.0_2.dev50
One-line Summary: Small change to prevent CAM from incorrectly modifying sea ice fractions in SOM/CCSM mode.
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Compaq
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
A small change was made to comsrf.F90 to prevent CAM from incorrectly
modifying sea ice fractions in SOM/CCSM mode. This corrects an error
introduced in dev47, and therefore the tags dev47, dev48, and dev49
should not be used for SOM or CCSM runs.
===============================================================
===============================================================
cam2_0_2_dev49
Originator: hender ( Tom Henderson)
Date: Tue Nov 18 12:26:08 MST 2003
Model: CAM
Version: CAM2.0_2.dev49
One-line Summary: New T85 surface data set with pre-computed fields, fixed SST data sets at T85 and T170, T85 tuning.
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Compaq
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
(If not abort this tag and follow the procedures in)
http://www.cgd.ucar.edu/~cam/cam_checkins/checkin_process.shtml
Changes made:
- Modified DefaultCLMEXPNamelist.xml to point to new T85 surface data set from
Jim McCaa. This data set contains pre-computed surface fields and cuts more
than 10 minutes off of T85 initialization.
- Modified DefaultCAMEXPNamelist.xml to point to new climatological sst files
for T85 and T170 from Jim McCaa.
Details from Jim McCaa:
On Fri, 31 Oct 2003, Jim McCaa wrote:
> These two files are intended to replace broken climatological sst files for
> T85 and T170, respectively, in /fs/cgd/csm/inputdata/atm/cam2/sst.
>
> /fs/cgd/data0/jmccaa/sst_HadOIBl_bc_128x256_clim_c031031.nc
> /fs/cgd/data0/jmccaa/sst_HadOIBl_bc_256x512_clim_c031031.nc
>
>
> The defaults xml file should be updated to point to these new files,
> replacing> any entries pointing to either of
> /fs/cgd/csm/inputdata/atm/cam2/sst/sst_HadOIBl_bc_128x256_clim_c020812.nc
> /fs/cgd/csm/inputdata/atm/cam2/sst/sst_HadOIBl_bc_256x512_clim_c020812.nc
>
> This will primarily affect anyone trying to run Jim R.'s defineqflux
> program,
> i.e. anyone trying to set up a T85 or T170 SOM run.
>
> I produces these new files using the old ones and the command:
> ncks -d time,0,11
- T85 tuning in cloud_fraction.F90 from Jim McCaa:
49c49
< rhminl = .92
---
> rhminl = .91
===============================================================
===============================================================
cam2_0_2_dev48
Originator: rosinski ( Jim Rosinski)
Date: Tue Nov 11 09:12:51 MST 2003
Model: CAM
Version: CAM2_0_2_dev48
One-line Summary: Save memory in aerosol code by holding in-core only 2 time levels of data.
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: Yes. per MPI task memory saved @ T42 = 190 MB/ntask
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, PC
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit), except only roundoff under PC/pgf90
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
o Revised aerosol reading to be like SST and ozone, keeping in-core only 2
time levels of data to save memory. Initialization should speed up lots,
since 2 time levels of aerosol data are read in instead of 12.
It is now feasible from a memory use standpoint to run T85 using 2 MPI
tasks on a machine with 2 GB of memory.
BYPASS ifdef was removed since the speedup renders it unnecessary.
Minor mods were made to sst and ozone interpolation routines for clarity
and consistency.
There are still some differences between the new code and ozone/SST
handling. There is no "aercyc = .false." namelist option. Date info is
gotten from an internal Fortran "data" statement rather than read from a
boundary dataset. Existing routine "aerosol_initialize" takes the role of
what otherwise would be called "aerini" (instead of being named like
"sstini" and "oznini"), and is called from "inti" instead of "initext".
There is a new routine "aerint" in aerosols.F90 a la "sstint" and "oznint"
which is called from "advnce". Like these routines it reads in a new time
slice of data when needed. But the actual time interpolation is done in
"get_aerosol". The reasons for these differences have mainly to do with
maintaining minimal intrusiveness, and the fact that there are plans
to modify the way time and vertical interpolation of aerosols is done.
===============================================================
===============================================================
cam2_0_2_dev47
Originator: jet ( John Truesdale)
Date: Wed Nov 5 15:41:05 MST 2003
Model: CAM
Version: CAM2.0_2.dev47
One-line Summary: cam usage of boundary dataset ice fraction made consistant with HADISST data.
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Linux
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: yes. Ice extent is now reduced around land.
Changes to CLM land-model: None (one-line description)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
After computing the ice area of the original datasets and comparing it to
the ice area produced by our t42 model, it became apparent that CAM was
over estimating the ice area. The ice fraction on the boundary dataset was
intrepreted as being the fraction of ice covering an entire gridbox. A more
accurate way of determining the ice coverage is to regard the boundary
dataset ice fraction as the fraction of ice covering the *non-land* portion
of the grid box. So
gridbox ice fraction = boundary dataset ice fraction * (1-land fraction)
The ice area produced by cam with this change is now very close to the
original datasets and to the ice area produced by CSM. The overall effect
is a reduction of ice coverage in those grid boxes with some land fraction.
This change is a one line mod to comsrf.F90
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam2_0_2_dev46
Originator: pworley ( Patrick H Worley)
Date: Tue Nov 4 15:26:37 MST 2003
Model: CAM
Version: CAM2.0_2.dev46
One-line Summary: Decreased memory requirements in dynamics/eul/inidat and added new load balancing scheme to phys_grid (same as dev39).
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: Yes - decreased number of full 3D fields needed in inidat from 9+(pcnst+pnats) to 4
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, Linux cluster
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None (one-line description)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes (same as dev39)
Changes made:
The cam2_0_2_dev39 commit failed (on Sep. 29, 2003). This commit reintroduces those changes.
The following is identical to the dev39 documentation ...
1) restructured dynamics/eul/inidat.F90 to minimize the memory requirements.
In the original version, masterproc allocates (9+pcnst+pnats) 3D fields
and 16 2D fields. It then reads in or generates the initial data,
scatters the data, and finally deallocates the arrays. In the revised
version, subsets of arrays are allocated, initialized, scattered, and
deallocated together, decreasing the maximum memory requirements to 4 3D fields.
Note that I implemented the logic that Brian Eaton used in FV inidat (for WACCM)
to treat each species in q3 separately.
2) added a new load balancing algorithm in physics/cam1/phys_grid.F90.
The original global balancing algorithm (opt = 1) eliminates
imbalances due to the diurnal and seasonal cycles. However, it still
assigns all polar latitdues to the same processors, the mid-latitudes
to the same processors, and the equatorial latitudes to the same processors.
The new algorithm (opt = 2) mixes up the latitudes as well, achieving
almost perfect load balance. It comes at a cost of increased communication cost
compared to opt = 1, and may or may not be a performance enhancer for
a given system, problem size, and processor count.
3) added an option to physics/cam1/phys_grid.F90 to implement the transpose_block_to_chunk
and transpose_chunk_to_block routines using point-to-point MPI commands rather
than mpi_alltoallv. This has been important on some systems in the past,
and is included here as an aid to experimentation. This option is currently
disabled in the code.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam2_0_2_dev45
Originator: hender ( Tom Henderson)
Date: Fri Oct 31 11:24:43 MST 2003
Model: CAM
Version: CAM2.0_2.dev45
One-line Summary: Performance optimization for IBM.
cam-bugs Requests resolved: none
Requires change in build system: yes
Fortran compiler optimization for IBM "xlf90" has changed from
"-O2 -qmaxmem=-1 -Q" to "-O3 -qstrict". The "-lmass" library is now
linked in unless DEBUG is on.
Substantial timing or memory changes: yes, it runs ~5% faster.
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, Linux-lf95
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Jim McCaa
Restart files change: no
Changes answers: Yes, same-physics
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Changed optimization in Makefile for IBM.
Changed "sanity checks" in pcond() (cldwat.F90) to make IBM compiler
happy (bfb).
Changed "rliq1 sanity check" in pcond() (cldwat.F90) to use Jim McCaa's
fix (bfb).
===============================================================
===============================================================
cam2_0_2_dev44
Originator: jmccaa ( James McCaa)
Date: Fri Oct 31 09:48:06 MST 2003
Model: CAM
Version: CAM2.0_2.dev44
One-line Summary: T85 tuning, removal of many default history file fields, modification of dif4 initialization.
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, linux
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit at all resolutions except T85)
Changes to CLM land-model: None (one-line description)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
This tag supplies three changes to the model:
1) The tuning of the T85 Eulerian model has been modified through the
addition of the relative humidity cloud threshold ramp in
cloud_fraction.F90. Tuning parameters were also modified in
zm_conv.F90, moistconvection.F90, and cldwat.F90.
2) The number of fields on the default monthly history files has been
reduced substantially. All fields continue to be available via the
fincl namelist facility. Also, default daily history files have
been eliminated.
3) The initialization of the diffusion parameters dif4 and dif2 have
been moved out of build-namelist and into the model. The default
dif4 value for T85 has been modified.
The files that were modified are:
>> M models/atm/cam/bld/DefaultCAMEXPNamelist.xml
>> M models/atm/cam/src/control/history.F90
>> M models/atm/cam/src/control/preset.F90
>> M models/atm/cam/src/dynamics/eul/dycore.F90
>> M models/atm/cam/src/dynamics/sld/dycore.F90
>> M models/atm/cam/src/physics/cam1/cldcond.F90
>> M models/atm/cam/src/physics/cam1/cldwat.F90
>> M models/atm/cam/src/physics/cam1/cloud_fraction.F90
>> M models/atm/cam/src/physics/cam1/gw_drag.F90
>> M models/atm/cam/src/physics/cam1/moistconvection.F90
>> M models/atm/cam/src/physics/cam1/vertical_diffusion.F90
>> M models/atm/cam/src/physics/cam1/zm_conv.F90
===============================================================
===============================================================
cam2_0_2_dev43
Originator: rosinski ( Jim Rosinski)
Date: Wed Oct 29 14:02:46 MST 2003
Model: CAM
Version: CAM2.0_2.dev43
One-line Summary: cos(lat) now applied to LANDM field offline. Since all IC files now require this
field (LANDM_COSLAT), new IC fields were generated.
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, PC
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: Yes. Same-to-roundoff T42. Changes answers other resolutions (due to LANDM).
Changes to CLM land-model: None (except sfc datasets must be generated).
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
o Migrated multiplication of LANDM field by cos(lat) offline. One reason to do this was that
model-written IC files were invalid, due to erroneous multiple applications of the above-mentioned
cosine function. Also, for horizontal resolutions other than T42 there was an undesirable
resolution-dependent behavior of the LANDM specification algorithm. In the new approach,
linear interpolation to the target resolution is done from a "master" LANDM dataset, which was
built at T42 (per pjr). The resulting LANDM_COSLAT field now looks similar regardless of resolution.
The "master" LANDM_COSLAT dataset lives in /fs/cgd/csm/inputdata/atm/cam2/hrtopo/landm_coslat.nc
It is read in by offline tool "definesurf", which has been modified to read it.
Since new IC files are now required for all resolutions, it was decided to introduce consistency
to these files by using the newest 10-minute hi-res topography dataset when creating them. The
newest topography file is /fs/cgd/csm/inputdata/atm/cam2/hrtopo/topo10min.merged_c030506.nc
As a consequence of using this new file, existing resolution-specific surface datasets and IC
files for CLM also are invalid. Thus for many resolutions (though not T42), CLM must be spun
up for any runs done with cam2.0.2.dev43 and beyond. These changes are reflected in the xml
files which are used to build CAM and CLM namelists.
===============================================================
===============================================================
cam2_0_2_dev42
Originator: rosinski ( Jim Rosinski)
Date: Tue Oct 28 16:02:07 MST 2003
Model: CAM
Version: CAM2.0_2.dev42
One-line Summary: Eliminate global TS print from physpkg and other minor speedups.
Add T85 and 2x2.5 aerosol datasets to xml file.
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: Minor speedup (see below).
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, PC
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
o Eliminated global TS calculation and printout from physpkg. This code was left over
from long ago and the output is now more easily obtainable by other means. Minor speedup.
o Moved calculation of tsice_rad to threaded region. Minor speedup on IBM, potentially
more significant when OMP_NUM_THREADS is a larger number (e.g. SGI). Unfortunately this
calculation is needed to create v2 IC files so it cannot be eliminated entirely. But it is
only needed when an IC file is generated, not every timestep. So further speedup is
possible.
o Made srfflx_state_reset callable by chunk to enable fusing threaded loops.
o Added reference to T85 and 2x2.5 aerosol datasets to CAM namelist xml file.
===============================================================
===============================================================
cam2_0_2_dev41
Originator: jmccaa ( James Mccaa)
Date: Thu Oct 23 15:54:52 MDT 2003
Model: CAM
Version: CAM2.0_2.dev41
One-line Summary: Climate tuning mods for T42 and T85.
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Linux
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: Yes, new-climate
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Modifications have been made to allow the code to run using different
physics tuning parameters for different dycores and resolutions.
The values of these parameters have been adjusted following numerous
sensitivity experiments.
The file cldfrc.F90 has been converted to a module named cloud_fraction.F90, and
an initialization routine has been added.
The files closure.F90, guang.h, and q1q2.F90 have merged into zm_conv.F90.
Minor code modifications have been made to allow greater optimization levels on
the IBM.
Changed files:
>> M models/atm/cam/src/advection/slt/hordif1.F90
>> M models/atm/cam/src/control/cam.F90
>> M models/atm/cam/src/control/history.F90
>> M models/atm/cam/src/control/preset.F90
>> M models/atm/cam/src/control/spmdinit.F90
>> M models/atm/cam/src/dynamics/eul/dycore.F90
>> M models/atm/cam/src/dynamics/fv/dycore.F90
>> M models/atm/cam/src/dynamics/sld/dycore.F90
>> M models/atm/cam/src/physics/cam1/cldnrh.F90
>> M models/atm/cam/src/physics/cam1/cldwat.F90
>> M models/atm/cam/src/physics/cam1/geopotential.F90
>> M models/atm/cam/src/physics/cam1/inti.F90
>> M models/atm/cam/src/physics/cam1/moistconvection.F90
>> M models/atm/cam/src/physics/cam1/pkg_cld_sediment.F90
>> M models/atm/cam/src/physics/cam1/radae.F90
>> M models/atm/cam/src/physics/cam1/sulchem.F90
>> M models/atm/cam/src/physics/cam1/tphysbc.F90
>> M models/atm/cam/src/physics/cam1/zm_conv.F90
>> M models/csm_share/shr_vmath_mod.F90
Removed files:
>> R models/atm/cam/src/physics/cam1/cldfrc.F90
>> R models/atm/cam/src/physics/cam1/closure.F90
>> R models/atm/cam/src/physics/cam1/guang.h
>> R models/atm/cam/src/physics/cam1/q1q2.F90
Added files:
>> A models/atm/cam/src/physics/cam1/cloud_fraction.F90
===============================================================
===============================================================
cam2_0_2_dev40
Originator: rosinski ( Jim Rosinski)
Date: Wed Oct 22 16:54:16 MDT 2003
Model: CAM
Version: CAM2.0_2.dev40
One-line Summary: Move horizontal aerosol interpolation offline. Modify print_memusage for readability and portability.
cam-bugs Requests resolved: none
Requires change in build system: yes: resolution-specific aerosol dataset required
Substantial timing or memory changes: yes: faster, uses less memory
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, PC
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Andrew Conley
Restart files change: no
Changes answers: Yes (same-to-roundoff)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
o Moved horizontal aerosol interpolation to CAM grid offline. The main reason for
this was to speed up initialization and reduce memory usage when running CAM. At T42
using 2 MPI tasks, the memory hi-water mark for MPI task 0 is reduced by about 50%.
Memory reduction for MPI task 1 is around 20%.
Since the boundary aerosol dataset is now resolution-specific, the amount of speedup vs.
dev39 will be resolution-dependent, with lower resolutions seeing the biggest benefit.
There will also be some speedup during the time-integration phase due to reordering loops
and array structure for cache efficiency in aerosol routine vert_interpolate. Since
the aerosol initialization code still reads a full 12 months of data, the startup time
overhead and memory overheads are not insignificant. We may want to further restructure
the code to read in aerosols the same way as SST and Ozone, keeping just 2 bounding
months of data in-core at any given time. In the meantime, if time spent reading
the aerosol boundary data still proves to be an annoyance, things can be sped up dramatically
if the aerosol dataset is made local rather than read from an NFS-mounted file system.
Alternatively, #define BYPASS is still available from earlier libs.
The new boundary dataset generation code lives in cam/tools/interpaerosols. A
README in that directory explains the procedure to generate an aerosol boundary
dataset at the target model resolution. Resolution-specific aerosol datasets have
already been created for resolutions needed by test-model, and some other common
resolutions.
Mods were also made to the main CAM Makefile. LAPACK/BLAS routines are only used by the
aerosol initialization procedure, so library references to these routines now only
exist in the offline code.
o Rewrote utility routine print_memusage for readability of output and portability. It now
works on Linux as well as SGI and IBM. The point of the routine is to print
memory use information at whatever point in the code the user wishes. On most machines
the numbers printed refer to hi-water mark, but on Linux they refer to current use (i.e.
on Linux the numbers do not necessarily increase monotonically with time). Also on
most machines, the printed numbers refer to KB of memory used. But this is not necessarily
true on all machines (e.g. Linux). Refer to "man getrusage" on most machines, and "man proc"
on Linux machines for details. The output from print_memusage now goes to Fortran unit 0,
which in most cases translates to stderr. So memory use information can be segregated from
stdout by for example: (cam < namelist >! out) >&! err. This code is far from perfect, e.g.
printed output can still be jumbled between MPI tasks.
Misc. notes:
o To get bit for bit vs. dev39, created a 64-bit aerosol dataset, pasted lon and lat
coordinate variables from a dev39 run into the input/output file required by interpaerosols.
Under pgf90 this is bfb. Unfortunately this procedure did not yield bfb results on IBM or SGI.
o Input aerosol dataset variable names and output history variable names are as before, but
with an "_V" appended to their name. This reflects the fact that these variables have
undergone a vertical sum. The different names will also cause the model to crash if an
incorrect boundary dataset is used.
===============================================================
===============================================================
cam2_0_2_dev39
Originator: pworley ( Patrick H Worley)
Date: Mon Sep 29 13:44:01 MDT 2003
Model: CAM
Version: CAM2.0_2.dev39
One-line Summary: Decreased memory requirements in dynamics/eul/inidat and added new load balancing scheme to phys_grid.
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: Yes - decreased number of full 3D fields needed in inidat from 9+(pcnst+pnats) to 4
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Linux cluster
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
1) restructured dynamics/eul/inidat.F90 to minimize the memory requirements.
In the original version, masterproc allocates (9+pcnst+pnats) 3D fields
and 16 2D fields. It then reads in or generates the initial data,
scatters the data, and finally deallocates the arrays. In the revised
version, subsets of arrays are allocated, initialized, scattered, and
deallocated together, decreasing the maximum memory requirements to 4 3D fields.
Note that I implemented the logic that Brian Easton used in FV inidat (for WACCM)
to treat each species in q3 separately.
2) added a new load balancing algorithm in physics/cam1/phys_grid.F90.
The original global balancing algorithm (opt = 1) eliminates
imbalances due to the diurnal and seasonal cycles. However, it still
assigns all polar latitdues to the same processors, the mid-latitudes
to the same processors, and the equatorial latitudes to the same processors.
The new algorithm (opt = 2) mixes up the latitudes as well, achieving
almost perfect load balance. It comes at a cost of increased communication cost
compared to opt = 1, and may or may not be a performance enhancer for
a given system, problem size, and processor count.
3) added an option to physics/cam1/phys_grid.F90 to implement the transpose_block_to_chunk
and transpose_chunk_to_block routines using point-to-point MPI commands rather
than mpi_alltoallv. This has been important on some systems in the past,
and is included here as an aid to experimentation. This option is currently
disabled in the code.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam2_0_2_dev38
Originator: hender ( Tom Henderson)
Date: Mon Sep 22 11:46:03 MDT 2003
Model: CAM
Version: CAM2.0_2.dev38
One-line Summary: Bug fix for prognostic aerosols, SE improvements.
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Linux-lf95
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Phil Rasch, Brian Eaton, Erik Kluzek
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None (one-line description)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Fixed bugs and added namelist variables to control prognostic aerosols.
(Bug fix for prognostic aerosols was provided by Phil Rasch.)
Improved design of parse_namelist(), prototype implementation is included
subject to review. So far design and implementation have been reviewed
only by Brian Eaton and Erik Kluzek.
Brought physics performance tuning parameters out to CAMEXP namelist.
New namelist variables are:
PHYSICS CONTROLS
AERO_SULFUR
Type: Logical
Default: .FALSE.
Description: If true, turn on sulfur prognostic aerosols.
AERO_FEEDBACK_SULFUR
Type: Logical
Default: .FALSE.
Description: If true, turn on feedback of sulfur prognostic aerosols. May
only be true if AERO_SULFUR is also true.
AERO_CARBON
Type: Logical
Default: .FALSE.
Description: If true, turn on carbon prognostic aerosols.
AERO_FEEDBACK_CARBON
Type: Logical
Default: .FALSE.
Description: If true, turn on feedback of carbon prognostic aerosols. May
only be true if AERO_CARBON is also true.
AERO_SEA_SALT
Type: Logical
Default: .FALSE.
Description: If true, turn on sea salt prognostic aerosols.
AERO_FEEDBACK_SEA_SALT
Type: Logical
Default: .FALSE.
Description: If true, turn on feedback of sea salt prognostic aerosols. May
only be true if AERO_SEA_SALT is also true.
PERFORMANCE TUNING
PHYS_LOADBALANCE
Type: Integer
Default: 0
Description: Select different options for organization of physics chunks.
Each uses a different scheme for static load balancing.
PHYS_CHNK_PER_THD
Type: Integer
Default: 1
Description: Select target number of chunks per thread. Must be positive
CHANGED FILES:
cam1/models/atm/cam/doc/UsersGuide/UG-20.html
cam1/models/atm/cam/src/dynamics/eul/inidat.F90
cam1/models/atm/cam/src/dynamics/sld/inidat.F90
cam1/models/atm/cam/src/dynamics/fv/inidat.F90
cam1/models/atm/cam/src/physics/cam1/aerosol_intr.F90
cam1/models/atm/cam/src/physics/cam1/carbon_intr.F90
cam1/models/atm/cam/src/physics/cam1/dust_intr.F90
cam1/models/atm/cam/src/physics/cam1/phys_grid.F90
cam1/models/atm/cam/src/physics/cam1/radctl.F90
cam1/models/atm/cam/src/physics/cam1/seasalt_intr.F90
cam1/models/atm/cam/src/physics/cam1/sulchem.F90
cam1/models/atm/cam/src/physics/cam1/sulfur_intr.F90
cam1/models/atm/cam/src/physics/cam1/wetdep.F90
cam1/models/atm/cam/src/control/cam.F90
cam1/models/atm/cam/src/control/comhd.h
cam1/models/atm/cam/src/control/parse_namelist.F90
===============================================================
===============================================================
cam2_0_2_dev37
Originator: mirin ( Arthur Andrew Mirin)
Date: Wed Sep 17 16:30:30 MDT 2003
Model: CAM
Version: CAM2.0_2.dev37
One-line Summary: Transpose and other FV improvements
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Compaq
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: no (unless test_model did that)
Code reviewed by: myself, Will Sawyer
Restart files change: no
Changes answers: no (bit-for-bit)
(Note: test-model.pl only does checks this -- if you run it with the "-compare" option)
Changes to CLM land-model: None (one-line description)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
(If not abort this tag and follow the procedures in)
http://www.cgd.ucar.edu/~cam/cam_checkins/checkin_process.shtml
Changes made:
This update accomplishes improvements to the finite-volume transposes and to
the FV dycore itself, as undertaken by Art Mirin and Will Sawyer. Specifically:
(A) The irregular communication routines of the mod_comm library
have been placed in their own module (mod_irreg), to enable use of a unified
version of mod_comm in CAM2 and FVGCM (NASA Goddard). This facilitates making use
of ongoing improvements at Goddard (such as support for SHMEM).
(B) The transpose calls directly reference mod_irreg routines as opposed to
going through the Pilgrim layer. This is true for use of mpi derived types as
well as contiguous buffers. Pilgrim's main function is to set up communication
patterns; mod_irreg then accomplishes the transposes.
By default, the transposes gather/scatter data from/to contiguous buffers. We
have found mpi derived types to be very slightly superior on the IBM.
One can implement use of mpi derived types at compile time by activating the line
"FPPFLAGS += -WF,-DUSE_MPI_TYPES" in the AIX section of the Makefile.
Support for one-sided (MPI2) communications when using MPI derived types has not
yet been implemented.
(C) There have been other miscellaneous improvements to Pilgrim / mod_comm.
(D) The Held-Suarez, Rayleigh friction and high-pressure mode have been removed
from the FV dycore.
(E) The FV version of inidat.F90 has been rewritten to minimize use of global 3D
arrays.
(F) Additional sectional timing diagnostics have been added, as well as an additional
optional timing barrier in the physics.
(G) There have been other miscellaneous improvements.
The following routine was added:
cam1/models/utils/pilgrim/mod_irreg.F90
The following routines were deleted:
cam1/models/atm/cam/src/dynamics/fv: hswf.F90, highp2.F90, dry_adj.F90, rayl_fric.F90
The following routines were modified:
cam1/models/utils/pilgrim: Makefile, mod_comm.F90, parutilitiesmodule.F90
cam1/models/atm/cam/bld/Makefile
cam1/models/atm/cam/src/physics/cam1/physpkg.F90
cam1/models/atm/cam/src/dynamics/fv: cd_core.F90, d2a3dikj.F90, dp_coupling.F90,
dynamics_vars.F90, dynpkg.F90, geopk.F90, inidat.F90, inital.F90, mapz_module.F90,
p_d_adjust.F90, pft_module.F90, pmgrid.F90, restart_dynamics.F90, spmd_dyn.F90,
stepon.F90, te_map.F90, uv3s_update.F90
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam2_0_2_dev36
Originator: jmccaa ( James McCaa)
Date: Fri Sep 12 14:32:06 MDT 2003
Model: CAM
Version: CAM2.0_2.dev36
One-line Summary: Merge of physics tuning branch onto cam_dev branch.
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Linux cluster
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers:Yes -- new climate
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
The following changes apply to all dycores:
* The fall velocity of large (400 micron) ice crystals has been reduced
from 3.5 m/s to 1 m/s.
* There is a lapse-rate reduction of the SST used for stratus diagnosis
when the sea surface elevation is not 0.
* The temperature ramp used to compute the effective radius of warm
clouds over land now goes from -20 C to 0 C.
* The land fraction ramp over the ocean goes from cos(lat) near the
coast to 0 at 1000 km from the coast.
* The effective radius and CCN of warm clouds over land now depend
on snow depth (linear ramp from 0 to 0.1 m liquid water equivalent).
* The evaporative efficiency for rain falling from cumulus has been
lowered from 3.0e-6 to 2.0e-6.
* Formulae for cloud fraction associated with moist convection have
been changed.
* Adjustments have been made to the aerosol optics.
The following modifications only apply to the finite volume dycore:
(They are in addition to the changes described above.)
* A new energy fixer has been added.
* The parameterization of gravity wave drag has been modified.
* A new parameterization of orographic form drag has been added.
* The Hack scheme now detrains cloud liquid.
* Evaporation efficiency for stratiform rain has been halved.
* Evaporation efficiency for convective rain has been halved.
* Formulae for cloud fraction associated with moist convection have
been changed.
The following modififications have been made to keep the standalone
CAM consistent with recent coupled runs:
* The snow fraction on land as a function of snow depth has been
changed.
* Both visible and near-ir albedos of snow on sea ice have been
lowered by 0.02.
* Albedos of bare sea ice have been lowered by 0.10 for visible and
0.06 for near-ir.
There is one infrastructure change:
* ncar_batch.csh now overrides any previously set values of the environment
variables INC_NETCDF and LIB_NETCDF to point to 64 bit versions
===============================================================
===============================================================
cam2_0_2_dev35
Originator: mvertens ( Mariana Vertenstein)
Date: Wed Sep 10 10:37:47 MDT 2003
Model: CAM
Version: CAM2.0_2.dev35
One-line Summary: fix clm bug that caused problems at very high mpi task number
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Compaq
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Forrest Hoffman
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: Yes
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Fixed bug in clm code (lnd_grid.F90) that caused problems to occur at very
high mpi task number
===============================================================
===============================================================
cam2_0_2_dev34
Originator: hender ( Tom Henderson)
Date: Tue Sep 2 12:41:50 MDT 2003
Model: CAM
Version: CAM2.0_2.dev34
One-line Summary: AIX 64-bit addressing and a bunch of other stuff
cam-bugs Requests resolved: 98, (127), 128
Requires change in build system: AIX build now uses 64-bit addressing
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Linux-lf95
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, contributors
Restart files change: no
Changes answers: Yes (same-to-roundoff for AIX, bfb for all others)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Bunch-o-fixes II:
- Jim Edwards' performance fix for estblf
- wv_saturation.F90
- change Makefile (etc.) to use 64-bit on AIX
- Modified Makefile, run-ibm.csh, CAM_lab.pm
- From Jim McCaa
The code already exists in history.F90, but is CPPed out. So, what I'm
requesting is to remove the two CPP lines, thereby enabling the code.
They are:
line 2501: #if ( defined CNTL_FORSOM )
line 2506: #endif
This won't affect answers, but will add fields to the monthly history
tape.
- From John Truesdale:
In diagnostics.F90 the outfld call for TSICERAD needs to be removed (per
Jerry O.) In tphysbc.F90 the DQCOND outfld call needs to be removed
(duplicate of outfld call using dcconnam).
- Jim Rosinski kindly offered to add a summary message to cprnc indicating
how many fields were compared and how many passed bfb.
- only one file changed:
/fs/cgd/data0/rosinski/newcprnc/cam1/models/atm/cam/tools/cprnc/cpr.F
- Fixed bug in fv 2D decomposition (From Art Mirin)
- Changed dp_coupling.F90 for fv
- Fixed CAM bug #98 (originally reported by Woo Sun Yang)
- Francis Vitt (ACD) provided the bug fix for writing staggered 'US' grid
from history.F90.
- Attempted to remove undocumented Lahey compiler option from Makefile as
suggested by Jim Rosinski
"For some reason the default CAM Makefile includes an undocumented compiler
flag (-CcdRR8) when using lf95 on the PC. The documented flag --dbl used
to suffice for auto-promotion of real to real*8, and has been removed.
Does anyone know why this was changed? If not, I propose changing back to
the documented flag if identical answers can be obtained. Undocumented
flags are a bad idea all around, I think.
JR"
- Can't make this change because "--dbl" promotes real*4's too (which
will break MPI communications and history output).
- Corrected misleading comments about this in Makefile.
- Implemented Brian Eaton's fix for incorrect override of qmin from namelist
- Modified all three flavors of initcom.F90
- Removed ccm366 physics from Default*Namelist.xml.
- Added new data file for Jerry Olson to DefaultCAMEXPNamelist.xml.
eul/ncdata_vers=2: cami_0000-09-01_64x128_L26_c030624.nc
sld/ncdata_vers=2: cami_0000-09-01_64x128_L26_c030624.nc
eul/ncdata_vers=1: cami_0000-09-01_64x128_L26_030506.nc
sld/ncdata_vers=1: cami_0000-09-01_64x128_L26_030506.nc
- Fix CAM bug #128 (fv build problems on Compaq and SGI) with bug fixes
from Tony Craig and Wei Yu.
- Modified cd_core.F90, and ccsm_msg.F90 (physpkg.F90 already fixed)
dev34 is bfb vs. dev33 on non-AIX machines, but
roundoff on AIX machines. AIX perturbation growth of cam2_0_2_dev33 is
nearly identical to error growth of difference between cam2_0_2_dev33 and
cam2_0_2_dev34 for all three dycores. roundoff is entirely due to change
in AIX compiler options to use 64-bit addressing in Makefile and in cprnc
Makefile. This was tested by comparing dev33+(new Makefiles) vs. dev33
(roundoff differences observed) and comparing dev34-(new Makefiles) vs.
dev33 (bfb exact).
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam2_0_2_dev33
Originator: mvertens ( Mariana Vertenstein)
Date: Mon Aug 25 12:06:15 MDT 2003
Model: CAM
Version: CAM2.0_2.dev33
One-line Summary: updated clm code from clm2_deva_30 to clm2_deva_36
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Linux
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: Yes (new-physics (clm only)
Changes to CLM land-model: Yes
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Only clm changes have been made in this tag.
cam2_0_2_dev32 contains clm2_deva_30 (with some changes on the cam branch for dust
that are only invoked when the cpp variable DUST is defined).
cam_0_2_dev33 contains clm2_deva_36 (which has incorporated the above dust mods).
The following summarizes the changes between clm2_deva_30 and clm2_deva_36.
The changes that lead to greater than roundoff differences are denoted by (*).
1) Put in dust mods in cam2_0_2_dev32.
*2*) Put in bug fix to BareGroundFluxesMod.F90 so that z0hg and z0qg at the column
level would not be set by pft level quantities. This changes answers to more
than roundoff.
3) Put in bug fix in Biogeophysics2Mod.F90 so that would not divide by
zero when calculating the ratio of soil evaporation for a pft relative
to the total of all evaporating PFTs. This would occur if the weight
became 0. in the dgvm calculation. This results in bfb answers in non-dgvm mode.
4) Put in bug fix to BareGroundFluxesMod.F90 so that z0hg and z0qg at the column
level would not be set by pft level quantities. This changes answers to more
than roundoff.
5) Put in bug fix in Biogeophysics2Mod.F90 so that would not divide by
zero when calculating the ratio of soil evaporation for a pft relative
to the total of all evaporating PFTs. This would occur if the weight
became 0. in the dgvm calculation. This results in bfb answers in non-dgvm mode.
6) Put in a change to DGVMMod.F90 to fix a bug in history file output
in compete mode (the default now).
*7*) Reversion to old fraction snow cover algorithm (clm2_deva_32)
8) Made clmtype variables relating to soil hydraulic and thermal properties
column physical state properties rather than land physical state properties
9) Fixed DGVM restart problem in compete mode:
Modified restFileMod.F90 and DGVMMod.F90 so that average over all
column pfts for h2ocan (needed by begwb computation in routine
driver.F90) is done in restfileMod.F90 only for non-DGVM mode and in
resetWeightstDGVM (in module DGVMMod.F90) for DGVM mode (since this
needs to be done after the weights are reset in the DGVM case).
10) Modified tssbef dimension in clmtype.F90 to be consistent.
===============================================================
===============================================================
cam2_0_2_dev32
Originator: pworley ( Patrick H Worley)
Date: Tue Aug 19 20:38:06 MDT 2003
Model: CAM
Version: CAM2.0_2.dev32
One-line Summary: Chunking data structures and optimizing interprocessor communication in ozone routines
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: Yes (performance improves for large process counts)
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Compaq
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
The original version of oznini reads in two timelevels of ozone data
and saves them in masterproc. masterproc then interpolates the data to the
current timelevel and broadcasts the entire field to each process.
Each shortwave or longwave timestep, advnce calls oznint.
In oznint, masterproc again interpolates the ozone data
to the current timelevel and broadcasts the entire field to each process.
(Every month or so, oznint also reads a new timelevel of
ozone data into masterproc.)
In the new version, oznini reads in the two timelevels, then distributes
both timelevels to all of the processes, so that they can do the interpolation
on their own. Moreover, the ozone data has been chunked, and each process
holds only the data that it needs for the columns that it is responsible for.
The performance impact is a significant decrease in the number of times
masterproc sends ozone data to the other processes, and a decrease in the
amount of data sent in each remaining communication. Routines modified were
comozp.F90, oznini.F90, oznint.F90, and radozn.F90, all in control/ .
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam2_0_2_dev31
Originator: rosinski ( Jim Rosinski)
Date: Thu Aug 14 21:54:00 MDT 2003
Model: CAM
Version: CAM2.0_2.dev31
One-line Summary: Bugfix from Natalie M. for sulfur fields. Consistent treatment of r4 vs. r8 in function calls
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, PC
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
(If not abort this tag and follow the procedures in)
http://www.cgd.ucar.edu/~cam/cam_checkins/checkin_process.shtml
Changes made:
o Sulfur bugfix from Natalie M. addresses negative concentrations.
o Consistent treatment of constants as r8.
===============================================================
===============================================================
cam2_0_2_dev30
Originator: pworley ( Patrick H Worley)
Date: Wed Aug 13 15:23:59 MDT 2003
Model: CAM
Version: CAM2.0_2.dev30
One-line Summary: Replaced global spectral arrays and indices with local arrays in EUL dycore, to decrease memory requirements.
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes:Yes (spectral memory requirements now decrease as number of MPI processes increase)
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Linux cluster
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: No (bit-for-bit) for SGI and Linux cluster, Yes (same-to-roundoff) for IBM
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
The primary goal of this update is to decrease the memory requirements
in the spectral Eulerian dycore by allocating space only for the
spectral coefficients that each processor is responsible for. While
we earlier decomposed the work, we have still been allocating the
full spectral arrays on each processor. This modification is important
for higher horizontal resolutions (T85, T170) as it decreases the memory
required for each process and may improve performance by improving memory
locality. To this end, the following files in dynamics/eul were modified:
comspe, spmd_dyn, trunc, initcom, inital, dynpkg, grcalc, hordif,
quad, and tstep
The routine spetru was also divided into 4 routines (in a new spetru module):
spetru_u., spetru_t, spetru_ps, and spetru_phis, spectrally truncating
the fields u and v, t, ps, and phis, respectively. The old spetru used
the global spectral arrays vz, d, t, and phis, which have been eliminated.
The new routines use at most a few single level global spectral arrays to
truncate the fields, again decreasing the required memory.
Finally a few random changes were included in this update:
Makefile -
Added -lcxml to the LDFLAGS for OSF, to link in DGETRF and DGETRS
on the Pittsburgh AlphaServer SC (needed by aerosols.F90)
caerbnd.F90, dmsbnd.F90:
Moved declaration for ncol so that it appears before its use in
array declarations. (HP/Compaq compiler complained.)
physpkg.F90:
Removed redundant USE of time_manager variables. (HP/Compaq compiler
complained.)
carbon_intr.F90, seasalt_intr.F90
Removed redundant USE of shr_const_mod variables. (HP/Compaq compiler
complained.)
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam2_0_2_dev29
Originator: aconley ( Andrew J. Conley)
Date: Thu Aug 7 10:25:49 MDT 2003
Model: CAM
Version: CAM2.0_2.dev29
One-line Summary: linearly interp asym and ssa parameters, Do not extrap optics beyond rh = 1.0
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, chinooke, anchorage
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers:Yes new-physics ( < 10^-3 relative change in aerosol heating)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Aerosol optics are no longer extrapolated beyond rh = 1.0.
Asymmetry and single scattering albedo are now linearly interpolated
in rh rather than exponentially interpolated.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam2_0_2_dev28
Originator: rosinski ( Jim Rosinski)
Date: Wed Aug 6 11:49:49 MDT 2003
Model: CAM
Version: CAM2.0_2.dev28
One-line Summary: Back out nf90 interface from sulfur codes.
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, PC
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
(If not abort this tag and follow the procedures in)
http://www.cgd.ucar.edu/~cam/cam_checkins/checkin_process.shtml
Changes made:
o Backed out the Fortran 90 interface from sulfur codes per CRB.
o IMPORTANT NOTE: There is a reproducibility problem on the IBM when the sulfur routines are
enabled. This problems exists both in the previous library (2.0.2.dev27) and this commit.
Only the sulfur fields are affected, and the impact appears to be minor. The problem is
likely related to threading, since 1-day test runs on IBM machines with OMP_NUM_THREADS=1
give reproducible answers. Threaded tests on chinook (SGI) were successful. Note that
test-model does not enable the sulfur routines, so results were bit-for-bit.
===============================================================
===============================================================
cam2_0_2_dev27
Originator: hender ( Tom Henderson)
Date: Wed Jul 30 15:10:50 MDT 2003
Model: CAM
Version: CAM2.0_2.dev27
One-line Summary: First cut at getting dust to work with physics chunking.
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, anchorage-lf95
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: Hacks for dust, see "#if (defined DUST)"
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Updated Natalie Mahowald's dust package (including her land model changes) to
work with PCOLS=16. It does not crash on anchorage but does crash on
blakcforest when cpp token DUST is defined (by default, dust, sulfates,
sea salt, and carbon packages are turned off via cpp tokens). MORE WORK NEEDS
TO BE DONE.
This tag is identical to cam2_0_2_dev25_brnchT_paero7.
Note that pgf90-pgcc and pgf90-gcc are known to fail because the Makefile has
not been updated to deal with the new F90 netcdf interfaces used in the new
code. The next tag should replace the F90 interfaces with the standard CAM
wrap_ncf() calls, if possible.
===============================================================
===============================================================
cam2_0_2_dev26
Originator: hender ( Tom Henderson)
Date: Mon Jul 28 18:18:26 MDT 2003
Model: CAM
Version: CAM2.0_2.dev26
One-line Summary: Added Phil Rasch's prognostic SOx aerosols.
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Linux-lf95
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
(If not abort this tag and follow the procedures in)
http://www.cgd.ucar.edu/~cam/cam_checkins/checkin_process.shtml
Changes made:
Added prognostic aerosols. All four packages (sulfates, carbon, sea salt, and
dust) are independently #ifdef'd out in aerosol_intr.F90. Sulfates have
been tested with the other three off. Results are bfb with cam2_0_2_dev25
when all four are off. Feedback to radiation is disabled in radctl.F90 (see
commented-out calls to set_aerosol_from_prognostics).
Time constraints forced the choice of cpp and comments instead of namelist
variables...
NEW FILES:
cam1/models/atm/cam/src/physics/cam1/acbnd.F90
cam1/models/atm/cam/src/physics/cam1/acsf.F
cam1/models/atm/cam/src/physics/cam1/aerosol_intr.F90
cam1/models/atm/cam/src/physics/cam1/caerbnd.F90
cam1/models/atm/cam/src/physics/cam1/caer.F90
cam1/models/atm/cam/src/physics/cam1/carbon_intr.F90
cam1/models/atm/cam/src/physics/cam1/dmsbnd.F90
cam1/models/atm/cam/src/physics/cam1/drydep_mod.F90
cam1/models/atm/cam/src/physics/cam1/dust.F90
cam1/models/atm/cam/src/physics/cam1/dust_intr.F90
cam1/models/atm/cam/src/physics/cam1/dust_sediment_mod.F90
cam1/models/atm/cam/src/physics/cam1/scyc.F90
cam1/models/atm/cam/src/physics/cam1/seasalt_intr.F90
cam1/models/atm/cam/src/physics/cam1/soxbnd.F90
cam1/models/atm/cam/src/physics/cam1/sulbnd.F90
cam1/models/atm/cam/src/physics/cam1/sulchem.F90
cam1/models/atm/cam/src/physics/cam1/sulemis.F90
cam1/models/atm/cam/src/physics/cam1/sulfur_intr.F90
cam1/models/atm/cam/src/physics/cam1/surface.F
cam1/models/atm/cam/src/physics/cam1/volcemission.F
cam1/models/atm/cam/src/physics/cam1/wetdep.F90
CHANGED FILES:
cam1/models/atm/cam/doc/UsersGuide/UG-20.html
cam1/models/atm/cam/bld/DefaultCAMEXPNamelist.xml
cam1/models/atm/cam/bld/camexp.pm
cam1/models/atm/cam/bld/run-ibm.csh
cam1/models/atm/cam/bld/run-pc.csh
cam1/models/atm/cam/src/dynamics/eul/inidat.F90
cam1/models/atm/cam/src/physics/cam1/advnce.F90
cam1/models/atm/cam/src/physics/cam1/aerosols.F90
cam1/models/atm/cam/src/physics/cam1/cldcond.F90
cam1/models/atm/cam/src/physics/cam1/comsrf.F90
cam1/models/atm/cam/src/physics/cam1/initindx.F90
cam1/models/atm/cam/src/physics/cam1/inti.F90
cam1/models/atm/cam/src/physics/cam1/physpkg.F90
cam1/models/atm/cam/src/physics/cam1/radctl.F90
cam1/models/atm/cam/src/physics/cam1/tphysac.F90
cam1/models/atm/cam/src/physics/cam1/tphysbc.F90
cam1/models/atm/cam/src/control/filenames.F90
cam1/models/atm/cam/src/control/intp_util.F90
cam1/models/atm/cam/src/control/parse_namelist.F90
cam1/models/atm/cam/test/system/ncar_batch.csh
===============================================================
===============================================================
cam2_0_2_dev25
Originator: jmccaa ( James Mccaa)
Date: Thu Jul 17 15:38:48 MDT 2003
Model: CAM
Version: CAM2.0_2.dev25
One-line Summary: Introduction of orographic form drag and modification of gravity wave drag routines.
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
(one-line description of changes: Makefile, include files, directories, etc.)
(Detailed description below)
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, linux-lf95, linux-pgf90
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: Yes, new-physics
Changes to CLM land-model: None (one-line description)
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Modified three files:
models/atm/cam/src/physics/cam1/gw_drag.F90
models/atm/cam/src/physics/cam1/tphysac.F90
models/atm/cam/src/physics/cam1/vertical_diffusion.F90
Introduced orographic form drag in vertical_diffusion module.
Modified gravity wave drag in gw_drag module.
These changes necessitated changes to the interfaces in tphysac.
Form drag and the new gravity wave drag algorithm are implemented only
when using the finite volume dycore. However, for the spectral dycores
the gravity wave algorithm was changed to disallow the formation of
gravity waves over pure ocean points with non-zero orographic standard
deviation.
====================================================
====================================================
cam2_0_2_dev24
Originator: eaton ( Brian Eaton)
Date: Fri Jul 11 12:48:54 MDT 2003
Model: CAM
Version: CAM2.0_2.dev24
One-line Summary: dycore independent energy fixer, only active for FV
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, pc-lf95
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: roundoff changes in spectral dycores, fv has new physics
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Reinsert changes made at cam2_0_2_dev21 (dycore independent energy fixer active
for fv only).
Fix bad error growth (add PERGRO condition in cldfrc.F90)
Fix namelist problem for 64x128 sld (new dataset in DefaultCAMEXPNamelist.xml).
===============================================================
===============================================================
cam2_0_2_dev23
Originator: hender ( Tom Henderson)
Date: Sat Jun 21 09:41:17 MDT 2003
Model: CAM
Version: CAM2.0_2.dev23
One-line Summary: Fixes for coupled runs, removal of CCM366 and LSM, other minor fixes.
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: yes
Machines tested: IBM, SGI, Intel-Linux
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Tony Craig, Art Mirin, and a cast of thousands
Restart files change: no
Changes answers: no (bfb with dev22)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
- Bug fix #122. (outfld calls from Jerry Olson's commit)
Modified routine diag_surf() in diagnostics.F90 so fields TSICE
and TSICE_RAD are excluded from the history files during a coupled
run.
- bugfix for ccsm_msg from Tony for fractional land
/fs/cgd/data0/tcraig/ccsm2_runs/f22.006/src.atm/ccsm_msg.F90
- Strip out CCM366 code now that test-model tests are no longer run...
Removed all files in models/atm/cam/src/physics/ccm366/ (CCM366
physics) and models/lnd/lsm/src/ (LSM).
- Deal with huge volume of print statements
Commented out print of "TIMEFILTER" messages per CAM CRB meeting on
6/20/03. Modified tfilt_massfix.F90
- Restore fv_prints.F90 as per Art Mirin's email
Got new fv_prints.F90 from sunray3:/home/mirin/outgoing/.
- Added Art Mirin's "one-line-change" to resolution_parameters.xml
Added new line for resolution=".5x.625"
===============================================================
===============================================================
cam2_0_2_dev22
Originator: hender ( Tom Henderson)
Date: Fri Jun 20 15:50:11 MDT 2003
Model: CAM
Version: CAM2.0_2.dev22
One-line Summary: Rolled back dev21 changes (per CAM CRB).
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: none. I verified that the entire cam1 module is identical to
dev20. Testing will be done for dev23
Ran test-model.pl script: no
Tested on fv dynamics: no
Tested on eul dynamics: no
Tested on sld dynamics: no
Tested that restarts are bit-for-bit: no
Tested multiple constituents: no
Tested that different domain decompositions match bit-for-bit: no
Tested in adiabatic mode: no
Code reviewed by: myself
Restart files change: no
Changes answers:
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made: dev22 is identical to dev20
===============================================================
===============================================================
cam2_0_2_dev21
Originator: eaton ( Brian Eaton)
Date: Wed Jun 18 13:15:50 MDT 2003
Model: CAM
Version: CAM2.0_2.dev21
One-line Summary: dycore independent energy fixer, only active for FV
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, PC/lf95
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, boville
Restart files change: no
Changes answers: Yes. spectral dycores change to round-off. FV dycore has a physics change.
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
Changes made:
Add dycore independent energy fixer. Currently it is only applying an energy fix when run
with the FV dycore. For the spectral dycores it is diagnostic.
A round-off level change was introduced into the spectral dycores due to 2 changes:
1. The definition of dry static energy was changed to include the surface geopotential.
2. The conversion between dry and wet constituent mixing ratios was moved into the coupling layer,
and an extraneous conversion that previously happened at the end of tphysbc and at the beginning
of tphysac has been removed.
See http://www.cgd.ucar.edu/~cam/cam_checkins/ for details.
===============================================================
===============================================================
cam2_0_2_dev20
Originator: erik ( Erik Kluzek)
Date: Mon Jun 16 12:54:27 MDT 2003
Model: CAM
Version: CAM2.0_2.dev20
One-line Summary: Get rid of ccm366/lsm test in test-model.pl, clean more files in $MODEL_BLDDIR, send *.cprout files in test-model to $MODEL_EXEDIR
cam-bugs Requests resolved: 112, 118(partial)
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Linux(lahey)
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, hender
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB?no
Got rid of the ccm366/lsm test in test-model.pl.
Clean more files in the $MODEL_BLDDIR when using test-model.pl or run-model.pl
(fixes bug #112)
Send all *.cprout files to $MODEL_EXEDIR instead of $SCRIPT_DIR this prevents
overwrites of files by different machines. This is a partial fix to bug #118.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
http://www.cgd.ucar.edu/~cam/cam_checkins/
for most recent and proposed checkins
===============================================================
===============================================================
cam2_0_2_dev19
Originator: hender ( Tom Henderson)
Date: Fri Jun 13 10:28:19 MDT 2003
Model: CAM
Version: CAM2.0_2.dev19
One-line Summary: Merged "science" branch into "dev" branch.
cam-bugs Requests resolved: none
Requires change in build system: Yes (see science branch documentation)
Substantial timing or memory changes: Yes (see science branch documentation)
Requires change in run script: no
Tested to work coupled with CCSM: yes
Machines tested: IBM, SGI, Linux
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: many
Restart files change: yes
Changes answers: Yes (new-climate)
Changes to CLM land-model: Yes
Have you filled out the pre-check-in documentation as required by the CAM CRB? yes
http://www.cgd.ucar.edu/~cam/cam_checkins/checkin_process.shtml
Changes made:
See http://www.cgd.ucar.edu/~cam/cam_checkins/ for details of all changes
made on the science branch.
===============================================================
===============================================================
cam2_0_2_dev18
Originator: rosinski ( Jim Rosinski)
Date: Tue Apr 8 17:09:30 MDT 2003
Model: CAM
Version: CAM2.0_2.dev18
One-line Summary: Speedup mods. Particularly for large number of threads (e.g. typical SGI)
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: Yes. Speedup mods. See below for details.
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, PC
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Have you filled out the pre-check-in documentation as required by the CAM CRB? no but I will
Changes made:
o Speedup mods. Should see order 10% speedup on SGI, 3% on IBM. Even better than that for
more than 32 threads on SGI. Specifics were:
- fuse threaded loops to minimize threading overhead
- Don't dereference pointers in expensive history calculations.
- Thread some calculations called from physpkg that were previously single-threaded. This
likely is the reason for such a big speedup on SGI.
o Changed ifdef DEBUG to HDEBUG inside history.F90. Reason was model barfed somewhere in
physics when DEBUG ifdef was set.
o Fix max/min combined with flag_xyfill=.true. pathological history case (prv commit).
===============================================================
===============================================================
cam2_0_2_dev17
Originator: rosinski ( Jim Rosinski)
Date: Sat Apr 5 11:48:41 MST 2003
Model: CAM
Version: CAM2.0_2.dev17
One-line Summary: Added ISCCP cloud simulator
cam-bugs Requests resolved: none
Requires change in build system: yes (new namelist variables)
Substantial timing or memory changes: history speedup.
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, PC
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Changes made:
o Added ISCCP cloud simulator. Cost overhead is around 5%. You need to set
namelist variable doisccp = .true. to enable the simulator. Then
isccpdata = '/fs/cgd/csm/inputdata/atm/cam2/rad/isccp.tautab_invtau.nc'
to tell the simulator where the required boundary dataset is.
Had to set default doisccp = .false. because test-model will barf when cprnc
gets confused at the different values of the "levels" coordinate variable.
This needs to be fixed.
Output from the simulator is 6 fields. The key field is FISCCP1, dimensioned
lon x 49 x lat x time. The 49 is really 7 pressure levels by 7 optical
depths. Pressure and optical depth of each "level" can be determined from
dimension variable "isccp_prstau". Pressure is the number to the left of
the decimal point in mb. Optical depth is the number to the right of the
decimal point times 1000.
The simulator is only called during daytime (coszrs > 0). As a consequence,
routine outfld in history.F90 was modified to ignore input array locations
set to fillvalue. To enable this feature, a new argument to addfld named
flag_xyfill was added. Set it to .true. to enable outfld ignoring fillvalue
input. Currently this is only done for appropriate ISCCP fields. If this
feature is enabled, for multilevel fields the setting of fillvalue MUST be
invariant in the z-direction. The reason is that the accumulator (nacs) is
currently only dimensioned x-y. The history handler checks for this
consistency and barfs if it is not met. Changing the accumulator to be
x-y-z capable would not be too hard, but might have severe performance
ramifications.
===============================================================
cam2_0_2_dev16
Originator: tcraig ( Anthony Craig)
Date: Thu Apr 3 15:51:34 MST 2003
Model: CAM
Version: CAM2.0_2.dev16
One-line Summary: Updates for CCSM coupling with cpl6, CCSM coupling with FV, rain/snow in coupling
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Compaq
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Changes made:
ccsm_msg.F90
- Significant rewrite for cpl6, has no impact on stand-alone CAM
(Rob Jacob).
- Fixed bug in outfld call for CPLRAIN*, CPLSNOW*, CPLPRCER
(Mariana Vertenstein).
- Removed artificial repartitioning of snow and rain in fields
sent to coupler to be ready for CLM2.2.
- Fixed bug in read/write of precca, precla, precsca, precsla
(Mariana Vertenstein, Bryon Boville)
- Modified coupling for LR dycore, removed special mods for LR
dycore in sendgrid setting mask to zero at most pole latitudes.
This was required for CCSM coupled runs with the LR dycore.
spmdinit.F90
- Removed the MPI_INIT call for COUP_CSM mode, cpl6 now handles this.
- Removed use of shr_msg_comm_atm (cpl5). Added a call to
cpl_interface_init (cpl6) to return the MPI communicator group.
cam.F90
- Modified some code on COUP_CSM definitions, remove calls to
shr_msg_init, shr_msg_groups, and shr_msg_finalize (cpl5).
Added call to cpl_interface_finalize (cpl6).
phys_grid.F90
- Added two new routines, buff_to_chunk and chunk_to_buff (Rob Jacob).
These are used in ccsm_msg to rearrange data from a lon-lat buffer
into chunk structure.
===============================================================
===============================================================
cam2_0_2_dev15
Originator: jet ( John Truesdale)
Date: Wed Apr 2 08:27:51 MST 2003
Model: CAM
Version: CAM2.0_2.dev15
One-line Summary: Added capability to output history tape fields on a column by column basis
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
(one-line description of changes: Makefile, include files, directories, etc.)
(Detailed description below)
Substantial timing or memory changes: no or Yes (describe)
Requires change in run script: no
(one-line description of changes: data files, run-scripts, namelist etc.)
(Detailed description below)
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Compaq
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit) or Yes (same-to-roundoff, same-physics, new-physics, new-climate)
(Note: test-model.pl only does checks this -- if you run it with the "-compare" option)
Changes to CLM land-model: None (one-line description)
Changes made:
I've added code and a set of namelist variables that allow fields for
a particular history tape to be written on a column by column basis.
The new namelist variables are named fincl1lonlat ... fincl6lonlat and
are used in conjunction with the fincl1 ... fincl6 namelist variables.
Here is an example of how to use the new variables to define individual
columns or a lat/lon range of columns for output.
1. Define a history tape with a set of fields that you would like to see
output at specific columns
ex. fincl3 = 'LHFLX','SHFLX','PS','T','U'
2. Define the columns for that history tape using fincl[1..6]lonlat
Like the fincl[1..6] specifiers all of the column identifiers are strings.
Specify the longitude in degrees followed by a single character
(e)ast/(w)est modifier. Follow this with an underscore ('_') and then
specify the latitude in degrees followed by a single character
(n)orth/(s)outh modifier.
ex. fincl3lonlat = '30.5e_45.5n'
will output the model column closest to 30.5 degrees east longitude
and 45.5 degrees north latitude for all fields defined by history tape
3.
You can specify a range of longitudes (or latitudes) by separating
the min and max values with a colon (':') like so
ex. fincl3lonlat = '30e:40e_45n'
will output the closest model columns from 30 to 40 degrees east
and at 45 degrees north for all fields defined on history tape 3.
The latitudes/longitudes must be positive (real or integer).
longitude values range from [0w to 180w] and [0e to 180e]
latitude values range from [90s to 0s] and [0n to 90n]
3) Variables will be put on the history tape with the longitude and latitude
coded as part of the name.
Ex. Using these namelist variables
fincl3 = 'LHFLX','SHFLX','PS','T','U'
fincl3lonlat = '30.5e_45.5n','30e:40e_45n'
will produce the following fields on history tape 3.
LHFLX_30.5e_45.5n
LHFLX_30e_to_40e_45n
PS_30.5e_45.5n
PS_30e_to_40e_45n
SHFLX_30.5e_45.5n
SHFLX_30e_to_40e_45n
T_30.5e_45.5n
T_30e_to_40e_45n
U_30.5e_45.5n
U_30e_to_40e_45n
The netcdf attributes for each field will be set to include the
actual latitude and longitude. This allows tools like ferret
to correctly plot the correct part of the world grid.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_2_dev14
Originator: eaton ( Brian Eaton)
Date: Thu Mar 27 15:07:31 MST 2003
Model: CAM
Version: CAM2.0_2.dev14
One-line Summary: fix fv restart under lf95/debug and fix bug in pseudo-ozone test tracer
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, pc-linux-lf95
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Changes made:
>> physics/cam1/restart_physics.F90 - in subroutine read_restart_physics add
>> calls to get_ncols_p inside the chunk loops and only assign the correct
>> number of columns worth of restart data into srfflx_state2d and
>> surface_state2d member arrays.
>>
>> physics/cam1/test_tracers.F90 - fix bug in subroutine test_tracers_timestep_tend.
>> All levels of the pseudo-O3 were being zeroed rather than just the bottom level.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_2_dev13
Originator: hender ( Tom Henderson)
Date: Tue Mar 25 10:17:12 MST 2003
Model: CAM
Version: CAM2.0_2.dev13
One-line Summary: Fixes and upgrades to test scripts, change for omega in history output.
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Linux-lf95
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Erik Kluzek, Jim McCaa
Restart files change: no
Changes answers: no
Changes to CLM land-model: None
Changes made:
1) Fixes to test-model.pl so -compare works (Kluzek).
2) Shorten run length for SOM tests in test-model.pl (Kluzek).
3) Script improvements to allow Wei to easily test lf95, pgf90-pgcc, and
pgf90-gcc in his automated runs (Henderson).
4) Makefile changes to make pgf90 bfb restarts with different number of
processes work with the fv dycore (Henderson).
5) Makefile changes to add Fortran optimization override environment
variable F_OPTIMIZATION_OVERRIDE (Henderson).
6) test-model.pl upgrade to use F_OPTIMIZATION_OVERRIDE when pgf90 is
used for fv04-fv06 and fv10-fv13 tests (Henderson).
7) test-model.pl upgrade to allow multiple test-model.pl incarnations to
be launched simultaneously from ncar_batch.csh (i.e. via queuing
systems) (Henderson, Kluzek).
8) Modified ncar_batch.csh to pass the "-unique_id" option to
test-model.pl (Henderson).
9) Changed $COMPARE_DIR/cam1 to $COMPARE_DIR in ncar_batch.csh to help
folks who use non-standard directory naming (Henderson).
10) Upgraded CAM XML parser to latest version (Kluzek).
11) Modified history file output to replace Omega@600mb with Omega@500mb
(McCaa and Henderson).
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_2_dev12
Originator: erik ( Erik Kluzek)
Date: Thu Mar 20 15:48:06 MST 2003
Model: CAM
Version: CAM2.0_2.dev12
One-line Summary: Fix bug in test-model with -compare and update XML::Lite from 0.11 to 0.14
cam-bugs Requests resolved: 104
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Sun, Linux-PC
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, hender, eaton
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Changes made:
Update XML::Lite from 0.11 to 0.14. Creator of XML::Lite, gave us
a newer version as he said the old version had "a significant bug"
in it.
Fix bug in -compare option for test-model.pl (bug #104). Previously
worked ok if you specified the dynamics, or for eul dynamics. It would
fail when it changed dynamics.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_2_dev11
Originator: mirin ( Arthur Andrew Mirin)
Date: Thu Mar 13 17:21:11 MST 2003
Model: CAM
Version: CAM2.0_2.dev11
One-line Summary: Changes to transposes for 2D FV decomposition.
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no; however one can now make with 64-bit addressing on IBM.
Substantial timing or memory changes: no
Requires change in run script: no; however, there are some new input options.
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Compaq
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes; however, bug in script, so could not verify.
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Will Sawyer
Restart files change: no
Changes answers: no (bit-for-bit)
(Note: test-model.pl only does checks this -- if you run it with the "-compare" option)
Changes to CLM land-model: None
Changes made:
The main changes are in the area of transposes for the 2D FV decomposition.
a. Transposes now include ghost points, which obviates extra ghost point
communication following the transposes. This improves performance,
but with limited impact because so few of the transposes involve ghost
points.
b. Transposes now use the NASA/Goddard mod_comm package, which lies underneath
Pilgrim. This enables use of one-sided MPI-2 communications as well as
standard MPI. NASA/Goddard personnel report substantial gains on
the SGI when MPI-2 is used in place of standard MPI.
c. Arrays containing extra points in any of the dimensions still cannot
be handled and must be copied into (out of) arrays without extra
points. Instead, the associated copies are, to the extent possible,
overlapped with other messages. This improves performance.
d. The number of tracers to be transposed simultaneously is now specifiable.
Formerly, all tracers had to be transposed together. This facilitates
overlapping copying with communication (see above point). If there are
N tracers altogether, to be transposed in groups of M, one solves
N = Q * M + R, and does Q groups of M and 1 group of R. Preliminary
tests indicate that M=1 is good for coarse domain decompositions,
but that as the decomposition becomes finer, the latency of multiple
transpose calls outweighs the computation/communicaiton overlap
benefit. The default is M=6.
e. A new method for initializing the transposes has been implemented.
Unlike the old methodology, this new method works for ghost points.
The method scales fine spatially but very poorly with the number of
tracers to be simultaneously transposed. Until this is fixed,
one would best limit M (see above point) to 6 (the default).
f. We have been experimenting with MPI derived types, versus the current
method of copying data in and out of contiguous buffers. At present
we are getting mixed results. This option has been implemented for
MPI-1 only (not MPI-2). Further testing is underway.
The files affected are:
Directory ...../src/dynamics/fv, files cd_core.F90, dp_coupling.F90,
dynamics_vars.F90, dynpkg.F90, geopk.F90, inidat.F90,
p_d_adjust.F90, pmgrid.F90, spmd_dyn.F90, stepon.F90, trac2d.F90. These are virtually all transpose-related changes.
Directory ...../src/control, files cam.F90, spmdinit.F90.
The change to cam.F90 is to make sure TWOD_YZ is not set when SPMD
is unset. The change to spmdinit.F90 has to do with MPI
initialization on SGI machines when using MPI-2.
Directory ...../utils/pilgrim, files Makefile, Makefile.conf.AIX,
Makefile.conf.IRIX64, decompmodule.F90, ghostmodule.F90, memstuff.c,
mod_comm.F90, parutilitiesmodule.F90, puminterfaces.F90,
redistributemodule.F90.
Directory ...../utils/pilgrim/unit_testers, files Makefile,
decomptest.F90, ghosttest.F90, parpatterntest.F90,
parutilitiestest.F90, redistributetest.F90, unstructured.F90.
Directory ...../bld, files Makefile, configure_fv.pl, script.m4.
The change to Makefile is to provide option for 64-bit addressing
on IBM. To do so, switch actual and commented lines to set
ADDRESS := Q64. The other two files are NASA/Goddard-specific.
Directory ...../test/system, file llnl_frost.csh.
A full diff of changes is on goldhill in ~mirin/diffs/diff_archive_cam2_0_2_dev10.
Here is some information on new, as well as existing knobs:
Two parameters have been added to the mprun2d namelist (used for 2D FV
decomposition) - geopktrans and tracertrans. Geopktrans refers to the
geopotential computation/communication method in geopk.F90. A value
of 0 (default), which results in the setting of geopk16byte false,
means that transposes are computed before and after the calls to
geopk (from cd_core) in order to do the vertical indefinite integrals
in geopk. A value of 1 causes geopk16byte to be set to true,
meaning that instead of computing transposes, partial sums in each
vertical subdomain are calculated and then communicated in z as
needed; 16-byte arithmetic may be used to guarantee bit-for-bit agreement
across domain decomposition; the default (for purposes of speed) is
8-byte arithmethic. One can change to 16-byte arithmetic by editing
geopk.F90 to change DSIZE from 8 to 16 (switch actual and commented
statements). For performance purposes, the code should be used
as is with geopktrans equal to 1. The other parameter, tracertrans,
refers to the number of tracers simultaneously transposed (denoted by
M above); the default is 6 and in general need not be changed.
One can implement MPI derived types in the transposes by activating
USE_MPI_TYPES in models/utils/pilgrim/parutilitiesmodule.F90.
This is unsupported - do so at your own risk. This bypasses mod_comm
and works for only MPI-1. One can make use of MPI-2 in mod_comm by
activating MPI2 at the beginniing of mod_comm.F90. Use of this is
definitely at your own risk.
To run waccm configuration tests, please see instructions on goldhill
in ~mirin/waccm.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_2_dev10
Originator: hender ( Erik Kluzek, tagged by Tom Henderson)
Date: Wed Mar 12 13:46:38 MST 2003
Model: CAM
Version: CAM2.0_2.dev10
One-line Summary: Added SOM test to test-model.pl. Various bug fixes. Update to ESMF_0_0_2p8
cam-bugs Requests resolved: 101, 100, 99
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: yes, if running with new SOM
-som option in configure now works. Default will behave as before.
Substantial timing or memory changes: no
Requires change in run script: yes, if running with new SOM
Must use initial and SST datasets that are compatible with SOM. See
/ftp/pub/erik/cam2_0_2.scidac-atm.datasets.tar.gz
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Intel-Linux
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: Jim Rosinski, Erik Kluzek, Brian Eaton, Tom Henderson
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Changes to ESMF library: Update to ESMF_0_0_2p8, a pruned version of ESMF_0_0_2p5
Changes made:
Added SOM test to test-model.pl and fixed bugs:
Add a SOM test for all dycores and for the control library.
Add new datasets needed for running with SOM. Reset namelist
each test in test-model.pl to make test-model.pl more robust.
Add a no-fail option to cam_timing so that when using -nofail
if a simulation aborts test-model.pl won't stop. Fix reporting
problem of different history files with -nofail. Improve the
diagnostic output of test-model.pl especially for the -nofail
case. Also be more careful at cleaning up data, such that data files
are deleted only after needed comparisions are done.
New datasets are available under "datasets" on the cam developers
web page.
Put, support message about ESMF in Makefile.
U atm/cam/bld/DefaultCAMEXPNamelist.xml
U atm/cam/bld/Makefile
U atm/cam/doc/UsersGuide/test-model.html
U atm/cam/test/system/CAM_test.pm
U atm/cam/test/system/cam_timing.pm
U atm/cam/test/system/test-model.pl
Bug fixes to make Lahey compiler happy:
U atm/cam/src/ocnsice/som/ocn_srf.F90
U atm/cam/src/dynamics/fv/inidat.F90
U atm/cam/bld/configure
Changed these ESMF files to allow for removal of lots and lots of other
ESMF files:
Update to ESMF_0_0_2p8, so can prune out unused files and directories.
Also turn off dependency of OMP, PCL, and MPI libraries. This simplifies,
shortens, and makes the ESMF make more robust. It also removes half of the
amount of disk storage for ESMF than was used before, several of the unused
code directories were removed and only the minimal set needed for using
the ESMF time manager.
U utils/esmf/README
U utils/esmf/build/IRIX/ESMF_conf.h
U utils/esmf/build/IRIX64/ESMF_conf.h
U utils/esmf/build/alpha/ESMF_conf.h
U utils/esmf/build/linux_gnupgf90/ESMF_conf.h
U utils/esmf/build/linux_lf95/ESMF_conf.h
U utils/esmf/build/linux_pgi/ESMF_conf.h
U utils/esmf/build/rs6000_64/ESMF_conf.h
U utils/esmf/build/rs6000_64/base_variables
U utils/esmf/build/rs6000_sp/ESMF_conf.h
Ripped out lots and lots of ESMF files:
...
Upgraded to simplify automated testing on anchorage with pgf90-pgcc and
pgf90-gcc:
U atm/cam/test/system/ncar_batch.csh
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_2_dev9
Originator: rosinski ( Jim Rosinski)
Date: Wed Feb 26 17:59:48 MST 2003
Model: CAM
Version: CAM2.0_2.dev9
One-line Summary: Added SOM capability.
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit) except Linux-pgf90 is only roundoff
Changes to CLM land-model: None
Changes made:
o Added Slab Ocean Model (SOM) capability. To enable, change som/ to dom/
in Filepath, and add #define COUP_SOM in misc.h.
For prescribed-ice control runs that will create fluxes for SOM, to get
good results CNTL_FORSOM must also be #defined.
Climates have only been examined in EUL mode. SLD and FV do compile, but they
have not yet been tested.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_2_dev8
Originator: jet ( John Truesdale)
Date: Wed Feb 19 16:51:27 MST 2003
Model: CAM
Version: CAM2.0_2.dev8
One-line Summary: fix bug in srface diagnostic variables - fsnslnd fsnsocn etc
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
(one-line description of changes: Makefile, include files, directories, etc.)
(Detailed description below)
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Compaq
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
(Note: test-model.pl only does checks this -- if you run it with the "-compare" option)
Changes to CLM land-model: None (one-line description)
Changes made:
These mods were to fix a bug with the newly added surface diagnostics.
They were not giving bit for bit when restarting fv and sld runs. To
fix this I had to add some variables to the restart dataset
as well as put a n-1 timelevel of landfrac,ocnfrac,and icefrac into the
physics buffer. These n-1 timelevel values are used from the physics
buffer on restart.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_2_dev7
Originator: hender ( Tom Henderson)
Date: Mon Feb 10 09:49:09 MST 2003
Model: CAM
Version: CAM2.0_2.dev7
One-line Summary: Grab bag -- Makefile, script, and source code fixes for SGI, Linux (PGI and Lahey), IBM
cam-bugs Requests resolved: 92, 88
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
Lots of accumulated minor fixes for SGI, Linux, IBM. All bfb. Biggest change
is support for Linux-Lahey in source code and ncar_batch.csh. Brian Eaton,
Pat Worley, John Truesdale, Jim Rosinski, and Erik Kluzek contributed various
bits.
Substantial timing or memory changes: no
Requires change in run script: no
(one-line description of changes: data files, run-scripts, namelist etc.)
(Detailed description below)
Makefile, run-ibm.csh, ncar_batch.csh, ice_srf.F90, comsrfdiag.F90,
fv_prints.F90, test-model.pl (and files it uses)
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Linux-Lahey
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Brian Eaton (SGI bits in Makefile), Jim Rosinski
(Linux-Lahey bits in Makefile), John Truesdale (John made the fixes for
ice_srf.F90, comsrfdiag.F90, and fv_prints.F90), Erik Kluzek (Erik made the
fixes for test-model.pl)
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None (one-line description)
Changes made:
Note: This is item #018 on CAM Check-in List "Fix performance bug in Makefile
for SGI (formerly CAM task list #02.12.19.5)". It's scope has expanded
a bit.
A- Modified Makefile for SGI so LDFLAGS has options and libraries in the
correct order ( -mp -lmp -lmpi ) as described by Pat Worley.
B- Added Brian Eaton's fix for run-ibm.csh:
"On the IBM there are 3 environment variables that allow you to specify the
node/task configuration. The 2 that are set in the run-ibm.csh file imply
that the total number of mpi tasks (procs) is 2. It's likely that your job
has procs=1 due to the 3rd environment variable, MP_PROCS, being set
somewhere else, like your .cshrc or .profile file (perhaps by the SCD
supplied default version of these files). The fix I'd suggest is to add
the command "unsetenv MP_PROCS" to run-ibm.csh file in the POE Environment
settings section." -- Brian Eaton
C- Added Brian Eaton's fix for run-ibm.csh to ncar_batch.csh too.
D- Added support for the Lahey compiler on anchorage to ncar_batch.csh. PGI
becomes an option, Lahey the default.
E- Added John Truesdale's fixes for ice_srf.F90, comsrfdiag.F90, and
fv_prints.F90. This allows versions cam2_0_2_dev1 and later to work
with the Lahey compiler. Prior to these fixes, lf95 was complaining about
unintialized variables (etc.).
F- Erik Kluzek fixed test-model.pl to work around a problem on bluesky.
G- Changed pgf90 optimization from -fast to -O1 in Makefile to fix bfb problem
with test-model.pl test #fv06 (compare initial run to restart run with
different number of processors for fv dycore). All other tests were
passing bfb with -fast.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_2_dev6
Originator: pworley ( Patrick H Worley)
Date: Wed Feb 5 15:19:36 MST 2003
Model: CAM
Version: CAM2.0_2.dev6
One-line Summary: New 1D decomposition for Fourier and Spectral domains in SLD dycore, for improved performance
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: Yes, faster for large processor counts
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Compaq
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: No (bit-for-bit) on SGI and Compaq; Yes (same-to-roundoff) on IBM
Changes to CLM land-model: changed division by parameter to division by local copy in lnd_grid, to eliminate compiler complaint
Changes made:
Implemented a consistent wavenumber decomposition for Fourier
and Spectral domains in SLD dycore, decreasing MPI communication and
storage requirements. Wavenumber decomposition also modified,
to improve load balance and decrease amount of communication.
This modification is similar to that in cam2_0_1_dev3, except for
SLD instead of EUL.
Routines modified in dynamics/sld: comspe, courlim, dyn, dyndrv,
dynpkg, grcalc, hordif, quad, realloc4, realloc7, scan2, scandyn,
scanslt, spegrd, spmd_dyn, trunc, tstep, tstep1, vertnm.
Routines deleted in dynamics/sld: realloc3, realloc6 .
Also, changed division by parameter to division by a local
copy of the parameter in two places in lnd_grid, to eliminate
a compilation failure with some versions of the IBM compiler.
(Fix provided by Art Mirin.)
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_2_dev5
Originator: eaton ( Brian Eaton)
Date: Tue Jan 21 12:33:04 MST 2003
Model: CAM
Version: CAM2.0_2.dev5
One-line Summary: makefile/configure changes
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, pc-linux
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Changes made:
Add ability to specify additional flags to the C compiler and to the loader
(this ability already exists for the Fortran compiler). This provides
flexibility in porting to new platforms.
Change Makefile so that the Filepath, Srcfiles, and Depends files are
explicitly in the current directory. This fixes a bug which occurs when
make finds a Srcfiles file in the VPATH, and that Srcfiles is newer than
the Filepath (which results in Srcfiles not being created in the current
directory and the command to build the Depends file fails).
Add an option to configure to allow the configuration cache file to be
created in a different directory than the CAM build directory. The default
behavior of configure is unchanged.
Fix a broken configuration. The combination of omp on linux using pgf90
and gcc resulted in the timing library being built without threading
enabled. Removed the setting of the THREADED_PTHREADS and THREADED_OMP cpp
macros from utils/timing/gpt.h and set these in the Makefile. This also
allows them to be set only when SMP is true. They were previously set to
use threads whether the rest of the model was threaded or not.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_2_dev4
Originator: hender ( Tom Henderson)
Date: Fri Jan 17 11:23:15 MST 2003
Model: CAM
Version: CAM2.0_2.dev4
One-line Summary: ncar_batch.csh: added support for test-model.pl with -compare option
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no
Changes to CLM land-model: None
Changes made:
The new ncar_batch.csh can be used to do bitwise comparison with a previous
source version. The test-model.pl script is run internally with the -compare
option.
For NQS or PBS, the new usage is:
>> env SCRIPT_DIR=`pwd` COMPARE_DIR=$MYBASEDIR qsub ncar_batch.csh
For LoadLeveler, the new usage is:
env SCRIPT_DIR=`pwd` COMPARE_DIR=$MYBASEDIR llsubmit ncar_batch.csh
In both cases, directory $MYBASEDIR should contain the source code for the
previous version.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_2_dev3
Originator: jet ( John Truesdale)
Date: Mon Jan 13 08:53:33 MST 2003
Model: CAM
Version: CAM2.0_2.dev3
One-line Summary: Added diagnostics for surface budget calculation, moved sst/ocn interpolation to correct minor flux bug
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, Apache
Ran test-model.pl script: yes: Only on IBM
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: Yes (new-physics)
Changes to CLM land-model: None (one-line description)
Changes made:
Added diagnostics to allow for surface energy budget calculations. The
following diagnostics were added:
FSNSLND,FSNSOCN,FSNSICE,FLNSLND,FLNSOCN,FLNSICE,SHFLXLND,SHFLXOCN,SHFLXICE,
LHFLXLND,LHFLXOCN,LHFLXICE.
Also added the following diagnostics to provide ocean/ice forcing for som
model
FSNSOI,FLNSOI,SHFLXOI,LHFLXOI
Moved the interpolation of ocean and ice right before the surface routines
that use them. SSTINT and ICEINT were previously at the top of the time
loop before tphysbc in the routine advnce. I removed them from advnce and
placed them after tphysbc right before camocn and camice. This corrected
a minor bug in the models use of fluxes calculated by camocn and camice.
Previously new fluxes were computed by camocn and camice with a set of
land, ocean, and ice fractions. The time loop ended, advnce was called,
new surface fractions computed and then tphysbc used the fluxes calculated
the previously time around only using the new fractions instead of the
surface fractions used to compute the fluxes. tphysbc now uses the time
level of fractions appropriate for the fluxes that are being read in.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_2_dev2
Originator: rosinski ( Jim Rosinski)
Date: Thu Jan 9 17:16:38 MST 2003
Model: CAM
Version: CAM2.0_2.dev2
One-line Summary: Add evap bugfix to ice_srf.F90 per C. Bitz
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
(one-line description of changes: Makefile, include files, directories, etc.)
(Detailed description below)
Substantial timing or memory changes: no or Yes (describe)
Requires change in run script: no
(one-line description of changes: data files, run-scripts, namelist etc.)
(Detailed description below)
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Compaq
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit) or Yes (same-to-roundoff, same-physics, new-physics, new-climate)
(Note: test-model.pl only does checks this -- if you run it with the "-compare" option)
Changes to CLM land-model: None (one-line description)
Changes made:
o One-line chage adds sublimation over sea ice to evaporative flux in ice model.
Changes answers beyond roundoff. Analysis of effects of this change available at:
www.cgd.ucar.edu/cms/rosinski/analysis/20shrconstevap-cam20shrconst/sets.htm
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_2_dev1
Originator: hender ( Tom Henderson)
Date: Mon Jan 6 15:25:38 MST 2003
Model: CAM
Version: CAM2.0_2.dev1
One-line Summary: Minor improvements to batch execution of test-model and User's Guide updates
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: SGI
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself and Erik
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Changes made:
Description:
CAM_lab.pm Update linux defaults
CAM_test.pm Fixed bug so "FAIL" is printed when a test fails (instad of
"ran"). Only affected the -nofail option.
ncar_batch.csh Updated to chinook "ded_16" queue and added PBS stuff for
anchorage.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_1_dev17
Originator: erik ( Erik Kluzek)
Date: Fri Dec 20 15:42:18 MST 2002
Model: CAM
Version: CAM2.0_1.dev17
One-line Summary: Small script change for testing, Makefile change to run on Compaq
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Compaq, Sun, Linux-Lahey, Linux-PGI
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None (one-line description)
Changes made:
Point to new test-production.pl baseline.
Change Makefile to run on Compaq.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_1_dev16
Originator: mvertens ( Mariana Vertenstein)
Date: Thu Dec 19 14:31:58 MST 2002
Model: CAM
Version: CAM2.0_1.dev16
One-line Summary: updated cam documentation for cam2_0_2 release
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: None
Ran test-model.pl script: no
Tested on fv dynamics: no
Tested on eul dynamics: no
Tested on sld dynamics: no
Tested that restarts are bit-for-bit: no
Tested multiple constituents: no
Tested that different domain decompositions match bit-for-bit: no
Tested in adiabatic mode: no
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Changes made:
Updated cam documentation for cam2_0_2 release.
Clm documentation has already been updated.
Did not modify any code.
===============================================================
===============================================================
cam2_0_1_dev15
Originator: mvertens ( Mariana Vertenstein)
Date: Wed Dec 18 21:19:12 MST 2002
Model: CAM
Version: CAM2.0_1.dev15
One-line Summary: updated clm to clm2_deva_08
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: (see below)
tested that clm2_deva_08 worked with CCSM
Machines tested: IBM, SGI, Linux
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: Yes
Changes made:
1) Updated clm user's guide to clm2.1
2) Fixed restart problem for clm2.1 auxiliary history files
===============================================================
===============================================================
cam2_0_1_dev14
Originator: erik ( Erik Kluzek)
Date: Tue Dec 17 16:14:46 MST 2002
Model: CAM
Version: CAM2.0_1.dev14
One-line Summary: Fix script problems use new datasets, update to ccsm2_0_beta61
cam-bugs Requests resolved: 76, 79, 84, 85
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: yes (IBM)
Machines tested: IBM, SGI, Sun, Linux (PGI,Lahey)
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit) -- Except Z050 field changes!
Changes to CLM land-model: None
New datasets:
Add capability for 1x1.25 for FV. Use new datasets for FV 2x2.5.
Download the new cam2.0.2 set of SCIDAC datasets for use with
this version of the model.
Fix test-model.pl bug in comparision:
Fix comparision problem in test-model.pl. Also fix so that
doesn't use as much disk-space. And change so that does
a error-growth test with the control code.
Fix build-namelist bug:
Fix bug in writing out long lists.
Fix bug in Z050:
Z050 was 5 mbar surface rather than 50 mbar surface.
Fix PGI bug for open-MP processing:
Delete a specific OpenMP loop in dynamics/fv/inidat.F90 that
caused answers to change with number of processors.
Update to ccsm2_0_beta61:
Update models/bld and scripts to ccsm2_0_beta61.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_1_dev13
Originator: mvertens ( Mariana Vertenstein)
Date: Mon Dec 9 09:49:17 MST 2002
Model: CAM
Version: CAM2.0_1.dev13
One-line Summary: clm code updated to development tag clm2_deva_07 (clm2.1 code)
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no or Yes (describe)
Requires change in run script: no
New (converted) clm initial datasets must be used
see DefaultCLMEXPNamelist.xml
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Compaq, Sun, Linux
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: no
Code reviewed by: myself
Restart files change: Yes (for clm only)
Changes answers: Yes (same-to-roundoff, same-climate)
Changes to CLM land-model: YES (updated to clm2.1 data structures)
Changes made:
1) clm2.1 has replaced clm2.0 in the current code.
updated models/lnd/clm to clm development tag clm2_deva_07
The differences are roundoff.
Two climate simulations were done to verify this commit.
case: /OLESON/csm/eul_201dev10 - cam2_0_1_dev10 (clm2.0)
case: /OLESON/csm/eul_201dev10_06 - cam2_0_1_dev10 (clm2.1 - tag clm2_deva_06)
The two cases were been compared with the standard climatological SST simulation
using CAM2.0.dev24 (case eul20dev24_a) and show the same climate
Performance on blackforest is effectively the same.
2) initial clm datasets used with clm2.0 are no longer compatible with clm2.1
surface datasets used with clm2.0 will work with clm2.1
restart files used with clm2.0 are not compatible with clm2.1
a conversion tool has been created (see directory models/lnd/clm2/tools/convert_inic)
running gmake in this directory will generate the conversion tool executable, convert_inic
datasets needed by configure have already been converted and have been placed in the
directory /fs/cgd/csm/inputdata/lnd/clm2/inidata_2.1/cam
3) the xml file DefaultCLMEXPNamelist.xml in models/atm/cam/bld has been modified
to point to the converted clm2.1 initial datasets
4) the following tests were done using test-model.pl:
tests platform status
------------------------------------------------------------------
eul:1-6:64x128L26 ibm (blackforest) success
eul:1-6:64x128L26 sgi (chinookfe) success
eul:1-6:64x128L26 linux (apache) lf95, pgf90 success
eul:1-6:64x128L26 sun (flagstaf) success
eul:1-6:64x128L26 compaq (ornl-colt) cannot test due to know problem starting with
cam2_0_1_dev10 on compaq
eul:1-6:48x96L26 ibm (blackforest) success
eul:1-6:48x96L26 sgi (chinookfe) success
eul:1-6:48x96L26 linux (apache) pgf90 success
eul:1-6:48x96L26 sun (flagstaf) success
eul:1-6:48x96L26 compaq (ornl-colt) cannot test due to know problem starting with
cam2_0_1_dev10 on compaq
fv:1-6:2x2.5L26 ibm (blackforest) success
fv:1-6:2x2.5L26 sgi (chinookfe) success
fv:1-6:2x2.5L26 linux (apache) lf95 success (know problem with pgf90 on restart due to openmp)
fv:4-6:2x2.5L26 sun (flagstaf) success
fv:1-6:2x2.5L26 compaq (ornl-colt) cannot test due to know problem starting with
cam2_0_1_dev10 on compaq
sld:1-6 ibm (blackforest) success
sld:1-6 sgi (chinookfe) success
sld:1-6 linux (apache) lf95 success
sld:1-6 sun (flagstaf) success
sld:1-6 compaq (ornl-colt) cannot test due to know problem starting with
cam2_0_1_dev10 on compaq
sld:1-6:64x128L26 ibm (blackforest) success
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_1_dev12
Originator: eaton ( Brian Eaton)
Date: Tue Nov 26 17:48:40 MST 2002
Model: CAM
Version: CAM2.0_1.dev12
One-line Summary: new physics buffer and misc makefile changes
cam-bugs Requests resolved: none
Requires change in build system: yes
ccm366 physics now requires pnats=0 (done by configure)
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, pc-linux-lf95
Ran test-model.pl script: no
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: yes, one fewer timelevel saved for cld, tcwat, qcwat,
and lcwat fields.
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Changes made:
Implement new physics buffer (phys_buffer.F90) to manage fields that
persist across timesteps or that are communicated between different physics
packages within a time step. Eventually the buffer.F90 module will be
replaced. For now only the fields used by the stratiform cloud condensate
scheme are being managed by the physics buffer.
The new module in cldcond.F90 is the start of an implementation of the new
physics interface for the cloud condensate parameterization. It currently
just deals with the fields that use the physics buffer. These fields save
a dycore dependent number of timelevels in the buffer, and that number has
been reduced by one from the buffer.F90 implementation (i.e., eul saves 2
old timelevels while fv and sld just save one).
Removed the need to declare pnats=1 for a cloud water constituent which
wasn't used when running with ccm366 physics. The configure script now
sets pnats=0 in that case.
Modify Makefile, configure, and config_cache_defaults.xml to support using
netCDF f90 interface. The environment variable MOD_NETCDF should be set to
the directory that contains the netCDF module files. This capability is not yet
required as CAM isn't currently using the netCDF Fortran 90 interface.
Added -qmaxmem=-1 to the AIX fflags for optimization.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_1_dev11
Originator: eaton ( Brian Eaton)
Date: Thu Nov 21 19:12:10 MST 2002
Model: CAM
Version: CAM2.0_1.dev11
One-line Summary: constituent initialization mods
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, pc-linux-lf95
Ran test-model.pl script: no
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no, except for a specific non-production configuration described below
Changes to CLM land-model: None
Changes made:
Implemented mods to allow specifying for each constituent whether the
initial values are to be read from the initial file or set by an
initialization routine. The namelist variable readtrace now serves to set
the default value for all constituents except water vapor (its default
value is .true.). The default may be overridden by an optional argument in
the calls that register the constituents. The optional argument may be set
by namelist variables that are managed by the constituent's package, or may
be hardwired in the call. The ghg chemistry package uses the default value
determined by readtrace. The test_tracers package hardwires the initial
values to be set by an initialization routine. Thus it is no longer
necessary to set readtrace=.false. to run the test_tracer package in CAM.
The namelist variables nusr_adv and nusr_nad have been disabled.
In the FV dycore only, the initialization of constituents has been modified
to initialize one constituent at a time. This is to reduce the memory
requirements since we couldn't fit the waccm configuration (2x2.5, 66
levels, 50 constituents) on the ibm-winterhawk2 nodes.
The mods do not change answers in the production model. However the
following configuration has non-bfb changes: When the ghg chemistry is run
using initial values that are set by its initialization routine (rather
than being read from the initial file), the answers change due to changed
initial values. The ghg initial values are set with uniform distributions
rather than having a decay in the stratosphere as was previously done.
Byron believes this distribution will spin up to an equilibrium state more
quickly than the previous initial distribution.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_1_dev10
Originator: eaton ( Brian Eaton)
Date: Thu Nov 14 19:03:59 MST 2002
Model: CAM
Version: CAM2.0_1.dev10
One-line Summary: fix some uninitialized variables so lf95 debug mode works
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, pc-linux-lf95
Ran test-model.pl script: no
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself. Pat Worley provided the mods.
Restart files change: no
Changes answers: no
Changes to CLM land-model: None
Changes made:
These mods fix a couple of references to uninitialized variables that caused a
runtime error in executables built w/ lf95 and strict error checking.
The test-model.pl debug test (test 1) now works on all dycores with lf95.
Also fixed a couple of incorrect intent attributes.
Mods:
dynamics/{eul,sld}/linemsdyn.F90 - init vcour to 0.
vmax2d, vmax2dt, and vcour declared intent(inout) in
1) sld/realloc7.F90
2) sld/dyndrv.F90 (in both PVP and nonPVP branches)
3) sld/courlim.F90
cam1/radctl.F90 - fix outfld call for qrl
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_1_dev9
Originator: eaton ( Brian Eaton)
Date: Thu Nov 14 09:31:24 MST 2002
Model: CAM
Version: CAM2.0_1.dev9
One-line Summary: change AIX optimization from O3 to O2
cam-bugs Requests resolved: 75
Requires change in build system: no
Substantial timing or memory changes: didn't do performance tests
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM
Ran test-model.pl script: no
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no
Changes to CLM land-model: None
Changes made:
Makefile changes for AIX only:
O3 --> O2, and remove -qstrict which is only needed with -O3
Use -qsmp=omp:noopt in debug mode.
Surprisingly (at least to me) the answers are bfb when changing from "-O3 -qstrict" to "-O2".
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_1_dev8
Originator: forrest ( Forrest Hoffman)
Date: Wed Nov 13 12:37:22 MST 2002
Model: CAM
Version: CAM2.0_1.dev8
One-line Summary: Fix for new code introduced in cam2_0_1_dev7 that caused run-time failures on SGI and Linux
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM (cheetah), SGI (thanks to Mariana Vertenstein)
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Mariana Vertenstein
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: Fixes to lp_coupling
Changes made:
Replaced intent(out) with intent(inout) for srfflx to avoid passing
undefined elements back to the atmosphere physics since only land point
values are filled. This problem was exhibited under SGI IRIX and
Linux.
Added working scalars to the private clause of the OpenMP do parallel
statements to assure appropriate scope.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_1_dev7
Originator: forrest ( Forrest Hoffman)
Date: Fri Nov 8 15:46:07 MST 2002
Model: CAM
Version: CAM2.0.1
One-line Summary: introduced clump land model decomposition and implemented new alltoall comm between atm and lnd
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: Yes, about 7x speedup in cam<->clm communications (only)
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, Compaq
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Mariana Vertenstein, Pat Worley
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: New land model decomposition on clumps and alltoall comunication for CAM+CLM configuration
Changes made:
1. Added new subroutine get_chunk_coord_owner_p() to phys_grid for
both cam1 and ccm366 versions.
2. Added lnd_grid module containing the following:
public surface_grid_init ! initializes land surface decomposition
public get_nclumps ! returns the number of clumps defined
public get_clump_cell_id_coord ! returns clump/cell ids based on lon/lat
public get_clump_owner_id ! returns clump owner based on clump id
public get_clump_ncells_proc ! returns number of cells for process
public get_clump_ncells_id ! returns number of cells in clump
public get_clump_tpatch_proc ! returns number of patches for process
public get_clump_tpatch_id ! returns number of patches in clump
public get_clump_coord_id ! returns lon/lat coordinates based on id
public get_clump_patchinfo ! returns patch id and number of patches
public get_clump_patchwt ! returns patch weight based on patch num
surfage_grid_init() performs the land model domain decomposition into
clumps of grid cells containing patches. Clumps are private derived
data types which may be accessed through a suite of functions and
subroutines also contained in lnd_grid (see above).
3. Added lp_coupling module containing the following:
public lp_coupling_init ! initialize clump<-->chunk mapping
public lp_coupling_finalize ! destroy clump<-->chunk mapping
public alltoall_clump_to_chunk_init ! communicate fluxes from lnd to atm
public alltoall_clump_to_chunk ! communicate fluxes from lnd to atm
public alltoall_chunk_to_clump ! communicate fluxes from atm to lnd
lp_coupling_init() initializes the mapping between atmosphere
physics chunks and land model clumps. It also allocates buffers
for use by subroutines within the module which perform the alltoallv
communication/transposition of surface state and fluxes between the
atmosphere and the land. This coupling is used when CAM and CLM are
run together as a single executable (when COUP_CAM is defined).
4. Removed lnd_atmMod module which previously did processor mapping
for coupling.
5. Modified atm_lndMod module to use routines from lp_coupling for coupling.
Added TIMING_BARRIERS barriers for communications timing tests.
6. Modified clm_map subroutine to call surface_grid_init() (lnd_grid) for
domain decomposition and lp_coupling_init() (lp_coupling) to initialize
the atm<->lnd coupling.
7. Added npes=1 definition even for the case when SPMD is undefined so
that new domain decomposition works correctly when running SMP-only.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_1_dev6
Originator: pworley ( Patrick H Worley)
Date: Wed Nov 6 22:34:11 MST 2002
Model: CAM
Version: CAM2.0.1
One-line Summary: Fixed minor bugs introduced in cam2_0_1_dev3 that were found with Lahey and SGI compilers
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no or Yes (describe)
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Compaq
Ran test-model.pl script: yes
Tested on fv dynamics: no
Tested on eul dynamics: yes
Tested on sld dynamics: no
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Changes made:
All changes in dynamics/eul only:
1) Modified courlim and dyndrv to change intent of vmax2d,
vmax2dt, and vcour from in to inout.
2) Modified grcalc to set dalpn(2) to zero on processors
that do not calculate it. This eliminates a failure in
the SGI test-model DEBUG tests from using an uninitialized
variable. (dalpn(2) is used to calculate zurcor, which is
used only on the processor which also calculates dalpn(2). )
3) Moved an array initialization using array syntax
out of an OpenMP loop, as a
work around for a pgf90 compiler error.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_1_dev5
Originator: bshen ( Bo-Wen Shen)
Date: Wed Nov 6 09:22:18 MST 2002
Model: CAM
Version: CAM2.0.1
One-line Summary: Changed DAO scripts
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: Yes (in DAO scripts)
Tested to work coupled with CCSM: no
Machines tested: SGI
Ran test-model.pl script: no
Tested on fv dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: no
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None (one-line description)
Changes made:
Modified DAO scripts (configure_fv.pl and script.m4) to run
the model on DAO SGI machines.
1. Defined SMP in Make.macros when OpenMP is used
to run a hybrid mode on SGI.
2. Changed the way to call system_cmd, which is
written in C and does not accept a function name
(trim in this case) as an argument on DAO SGI.
3. Modified mswrite/msread which will generated at run time
4. Calculated total # of CPUs and passed the # to PBS
directives
===============================================================
===============================================================
cam2_0_1_dev4
Originator: eaton ( Brian Eaton)
Date: Wed Oct 30 07:46:47 MST 2002
Model: CAM
Version: CAM2.0.1
One-line Summary: implement ghg surface values module
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, pc-linux
Ran test-model.pl script: no
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Changes made:
This set of mods is in preparation for further mods to implement the new
physics interface design for initializing constituents. The current loop
to initialize constituents that are not read from the initial file (in
{eul,fv,sld}/inidat.F90) contains a call to ramp_ghg(). This function sets
surface values for the ghg constituents which are used by the radiation
codes as well as by the ghg chemistry code. The mods being committed
gather the code which is responsible for setting surface values of the ghg
constituents, and time interpolating them (ramping) from an annual dataset
if requested, into the single module in the file ghg_surfvals.F90.
The new module ghg_surfvals does the following:
. maintains the values co2vmr, co2mmr, n2ovmr, ch4vmr, f11vmr, and f12vmr
as public data
. maintains the namelist variables scenario_ghg and rampyear_ghg. The
meaning of these variables has not changed.
. provides ramping functionality (time interpolation) when scenario_ghg is
set to 'RAMPED'. The data used by the ramp is determined by an include
statement in the module. Either file ramp_ghg_bau.h or ramp_ghg_stab.h
may be included. These .h files contain data formated as module data.
When using ramped data the default is to interpolate to the current
time. This may be modified by setting the rampyear_ghg variable to a
valid year which is then used to produce values that will cycle within
that fixed year.
Files added:
physics/{cam1,ccm366}/ghg_surfvals.F90
Files removed:
physics/{cam1,ccm366}/ramp_ghg.F90
physics/ccm366/comvmr.F90
Files changed:
control/preset.F90
control/parse_namelist.F90
control/comctl.h
control/restart.F90
physics/cam1/constituents.F90
physics/cam1/{chemistry.F90,radae.F90,radini.F90,trcmix.F90}
physics/cam1/ramp.h
physics/cam1/{ramp_ghg_bau.h,ramp_ghg_stab.h}
physics/cam1/crdcon.h
physics/cam1/radini.F90
physics/cam1/{radcswmx.F90,trcpth.F90}
physics/cam1/advnce.F90
physics/ccm366/ramp_ghg.F90
physics/ccm366/{ramp_ghg_bau.F90,ramp_ghg_stab.F90}
physics/ccm366/constituents.F90
physics/ccm366/{radabs.F90,radems.F90,radinp.F90}
physics/ccm366/radini.F90
physics/ccm366/crdcon.h
physics/ccm366/{radcsw.F90,trcpth.F90}
physics/ccm366/trcmix.F90
physics/ccm366/chemistry.F90
physics/ccm366/physconst.F90
physics/ccm366/inidat.F90
physics/ccm366/advnce.F90
dynamics/{eul,fv,sld}/inital.F90
dynamics/{eul,fv,sld}/inidat.F90
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_1_dev3
Originator: pworley ( Patrick H Worley)
Date: Sun Oct 27 18:17:22 MST 2002
Model: CAM
Version: CAM2.0.1
One-line Summary: New 1D decomposition for Fourier and Spectral domains in EUL dycore, for improved performance
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: Yes, faster for large processor counts
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, Compaq
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: John Drake, Michael Ham
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None (one-line description)
Changes made:
Implemented a consistent wavenumber decomposition for Fourier
and Spectral domains in EUL dycore, decreasing MPI communication and
storage requirements. Wavenumber decomposition also modified,
to improve load balance and decrease amount of communication.
Routines modified: comspe, dp_coupling, dyn, dyndrv,
dynpkg, grcalc, hordif, linemsdyn, quad, realloc4, realloc7, scan2,
scandyn, scanslt, spegrd, spmd_dyn, trunc, and tstep. Also,
realloc3 and realloc6 no longer needed and deleted.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_1_dev2
Originator: pworley ( Patrick H Worley)
Date: Sun Oct 27 09:10:31 MST 2002
Model: CAM
Version: CAM2.0.1
One-line Summary: In-lined two temporaries in grcalc, so that new EUL dom. decomp. will be bit_for_bit on the IBM in next check-in
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, Compaq
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: no
Tested on eul dynamics: yes
Tested on sld dynamics: no
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself and Jim Rosinski
Restart files change: no
Changes answers: no on Compaq, yes on IBM (same-to-roundoff, same-physics)
Changes to CLM land-model: None
Changes made:
Eliminated tmpr and tmpi in grcalcs and grcalca in dynamics/eul/grcalc.F90:
tmpr = alps(ir)*alp(mr+n,irow)
tmpi = alps(ii)*alp(mr+n,irow)
in-lining the expressions in the two places where each
of them is used.This causes roundoff level changes to the solution
on the IBM, but not on the HP/Compaq. With this change, the restructuring
of the domain decomposition used in the spectral Eulerian (next check-in)
will be bit-for-bit. This check-in is used to isolate and document
the source of the roundoff change.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_1_dev1
Originator: sawyer ( William Barton Sawyer)
Date: Sun Oct 20 00:53:30 MDT 2002
Model: CAM
Version: CAM2.0.1
One-line Summary: Bug fix for tracer advection (from S.-J. Lin)
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: SGI
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit
Changes to CLM land-model: None
Changes made:
Retagged cam2_0_dev26 to confirm with new tagging scheme.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_dev26
Originator: sawyer ( William Barton Sawyer)
Date: Fri Oct 18 07:48:14 MDT 2002
Model: CAM
Version: CAM2.0.dev26
One-line Summary: Bug fix from FVGCM for multiple advected constituents (S.-J. Lin)
cam-bugs Requests resolved: unlisted
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: SGI
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Lin
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None (one-line description)
Changes made:
src/dynamics/fv/trac2d.F90: bug fix, dp2 copy to dp1
moved from above label 5000 to between labels 5000 and 6000
bld/script.m4: changes to avoid runtime errors in DAO runs
which resulted from bug fix 64 (shell cmd now returns error code).
Note: test-model.pl is successful but does not test the
multiple advected constituent case. This change will be
further tested by DAO runs.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_dev25
Originator: erik ( Erik Kluzek)
Date: Wed Oct 2 08:33:38 MDT 2002
Model: CAM
Version: CAM2.0.dev25
One-line Summary: Fix run-pc.csh script.
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: Linux-PGI
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None (one-line description)
Changes made:
Fix run-pc.csh so will run without a core-dump on apache. To do this
I changed it from pur-SPMD to pure-OpenMP mode.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_dev24
Originator: erik ( Erik Kluzek)
Date: Tue Sep 24 22:42:42 MDT 2002
Model: CAM
Version: CAM2.0.dev24
One-line Summary: Fix minor problems (tau units, SGI system call, FV array)
cam-bugs Requests resolved: 64,65,66
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: Yes! (SGI, IBM)
Machines tested: IBM, SGI, lf95, PGIF90, Solaris
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself,mvertens,boville (units)
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: Use "C" wrapper for SGI system call
Changes made:
Units change: (cam-bugs 66)
Surface stress was identified as N, consistently use N/m^2.
FV fix: (cam-bugs 65)
Make cut array allocatable and dimension acording to npes.
SGI System routine: (cam-bugs 64)
Use a "C" wrapper to system call on SGI so that the return
code can be queried.
INT long-name:
Change long-name on PRECCINT and PRECLINT to note that to
get intensity (average precipitation of precipitating events), you
divide by PRECCFRQ or PRECLFRQ respectively.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_dev23
Originator: eaton ( Brian Eaton)
Date: Fri Sep 20 15:41:12 MDT 2002
Model: CAM
Version: CAM2.0.dev23
One-line Summary: updates to User Guide
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: none
Ran test-model.pl script: no
Tested on fv dynamics: no
Tested on eul dynamics: no
Tested on sld dynamics: no
Tested that restarts are bit-for-bit: no
Tested multiple constituents: no
Tested that different domain decompositions match bit-for-bit: no
Tested in adiabatic mode: no
Code reviewed by: myself
Restart files change: no
Changes answers: no
Changes to CLM land-model: None
Changes made:
Made changes to the sections on configuring and building CAM. Added new
material on the run scripts. Edited details of config files and make
procedure in appendix B.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_dev22
Originator: rosinski ( Jim Rosinski)
Date: Fri Sep 20 14:50:14 MDT 2002
Model: CAM
Version: CAM2.0.dev22
One-line Summary: fix potential memory overwrite bug in radae.F90
cam-bugs Requests resolved: 63
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, PC
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Changes made:
o Fixed potential memory overwrite bug in radae.F90
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_dev21
Originator: erik ( Erik Kluzek)
Date: Mon Sep 16 23:41:41 MDT 2002
Model: CAM
Version: CAM2.0.dev21
One-line Summary: Add new output fields: PRECCINT, PRECCFRQ, PRECLINT, PRECLFRQ, and Z050
cam-bugs Requests resolved: 46, 61
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: Yes! (IBM)
Machines tested: IBM, SGI, Linux-PGI, Linux-Lahey, Solaris
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, mvertens
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Changes made:
New output fields and namelist items:
PRECCINT, PRECCFRQ, PRECLINT, and PRECLFRQ. Precipitation that
exceeds a given threshold (PRECC_THRESH = 0.1mm/hr and PRECL_THRESH=0.05mm/hr
by default, both can be set on the namelist).
Fix minor problem with FV with Lahey:
Update ccsm scripts to ccsm2_0_beta54:
Fix so Solaris would work:
Reduce optimization level.
Minor changes to test-production.pl:
Remove files after completion, new reference simulation.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_dev20
Originator: rosinski ( Jim Rosinski)
Date: Mon Sep 16 14:38:10 MDT 2002
Model: CAM
Version: CAM2.0.dev20
One-line Summary: Implement shared constants in ice code. NOTE: no control run for this tag yet exists.
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, PC
Ran test-model.pl script: not yet.
Tested on fv dynamics: no
Tested on eul dynamics: yes
Tested on sld dynamics: no
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: no
Tested that different domain decompositions match bit-for-bit: no
Tested in adiabatic mode:no
Code reviewed by: myself
Restart files change: no
Changes answers: Yes (new-climate)
Changes to CLM land-model: None (one-line description)
Changes made:
o Use shared constants in CSIM4 ice code.
o Have not yet run test-model, but will today. Wanted to get commit in
because folks are waiting for it.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_dev19
Originator: eaton ( Brian Eaton)
Date: Sat Sep 14 10:24:24 MDT 2002
Model: CAM
Version: CAM2.0.dev19
One-line Summary: implement the Makefile SMP option for linux-pgf90
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: pc-linux-pgf90
Ran test-model.pl script: no
Tested on fv dynamics: yes
Tested on eul dynamics: no
Tested on sld dynamics: no
Tested that restarts are bit-for-bit: no
Tested multiple constituents: no
Tested that different domain decompositions match bit-for-bit: no
Tested in adiabatic mode: no
Code reviewed by: myself
Restart files change: no
Changes answers: no
Changes to CLM land-model: None
Changes made:
Added SMP option to the pc-linux-pgf90 section of the Makefile. Set
the default value of SMP to true since that replicates previous Makefile
behavior (the -mp flags were hardwired into the pgf90 settings). But
note that the default in the configure script sets SMP false for pc-linux.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_dev18
Originator: olson ( Jerry Olson)
Date: Fri Sep 6 17:29:17 MDT 2002
Model: CAM
Version: CAM2.0.dev18
One-line Summary: bugfix in sld/scan2.F90
cam-bugs Requests resolved: none
Requires change in build system: no
(one-line description of changes: fixed treatment of non-advected tracers in sld/scan2.F90
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI
Ran test-model.pl script: no
Tested on fv dynamics: no
Tested on eul dynamics: no
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: no
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: no
Tested in adiabatic mode: no
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Changes made:
sld/scan2.F90 was copying cloud water from one time level to
another. Loop was generalized to copy *all* non-advected
tracers.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_dev17
Originator: sawyer ( William Barton Sawyer)
Date: Thu Sep 5 06:48:57 MDT 2002
Model: CAM
Version: CAM2.0.dev17
One-line Summary: Allowed 3 latitudes per process; improvements from fvgcm-1_3_71
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: SGI
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, S.-J. Lin
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None (one-line description)
Changes made:
In src/dynamics/fv:
spmd_dyn.F90 Test for >= 4 latitudes now
>= 3 latitudes per process; tested
72x46x26 with 12 processes and
144x91x26 with 24 processes. Works.
cd_core.F90 Integrated optimizations and restructuring
from fvgcm-1_3_71. Zero diff changes.
Note that some experimental changes
from FVGCM (HIGH_P, value of tau)
have not yet been integrated.
sw_core.F90 Now includes upol5, vpol5
upol5.F90 Removed
vpol5.F90 Removed
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_dev16
Originator: erik ( Erik Kluzek)
Date: Wed Aug 28 11:22:02 MDT 2002
Model: CAM
Version: CAM2.0.dev16
One-line Summary: Fix deadlock with scenario_so4, fix various bugs,
T85, T170 in scripts, update to clm2_3_dev32, ccsm2_0_beta53
cam-bugs Requests resolved: 37, 49, 51, 52
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: Yes! (IBM)
Machines tested: IBM, SGI, Compaq, Solaris, PGF90
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, mvertens
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: update to clm2_3_dev32
Changes made:
Build changes:
Fix bug in "configure -test", so that tests will be done
in cam_bld directory, and then return to pwd. Add ability to do
T85, and T170 resolutions. Change prun in CAM_lab.pm for running
in hybrid mode.
History restart fix:
History restart files weren't going to mass store -- now
they do.
Decomposition set in phys_grid_init rather than in call to it:
Previously there were four locations to determine the
chunking decomposition type.
Also remove the "is_lsm" and "is_phys366" methods. Always
assume that lsm will ONLY be run with phys366 and CLM will ONLY be
run with the new physics.
Update CLM to CLM2_3_dev32
This fixes a problem with getting clmi files at T170.
Fix deadlock problem with scenario_so4:
A parameter was only set on masterproc that needed to be
set on all nodes.
Update to ccsm2_0_beta53 and scripts_a020822
both scripts and models/bld directories. scripts_a020822 fixes
a minor problem in ccsm2_0_beta53 scripts.
Test scripts:
Add test-example.pl script to run the examples from the
users guide. Does limited checking to ensure things work as expected.
This is setup to run on blackforest.
Add ability to set the remote lab in the "-errgro" option
in test-model.pl. This allows for port validation for remote
machines. Also delete more files that test-model.pl produces when
running in the default "clean" configuration. This prevents having
lots of wasted disk space for test-model.pl tests.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_dev15
Originator: sawyer ( William Barton Sawyer)
Date: Wed Aug 7 07:58:09 MDT 2002
Model: CAM
Version: CAM2.0.dev15
One-line Summary: Completed merge of fvgcm-1_3_56 te_map, cleaned up remaining t3/tvm arrays
cam-bugs Requests resolved: none
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: SGI, Linux
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None (one-line description)
Changes made:
src/dynamics/fv:
te_map.F90: completed merge of fvgcm-1_3_56 te_map.F90
code is same except for 2D decomposition
additions
dynpkg.F90: removed tvm, tvmxy arguments (information is
in pt, ptxy)
dp_coupling.F90:Use pt/ptxy instead of t3/t3xy
stepon.F90: remove t3xy entirely, t3 from d_p_coupling
arguments
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_dev14
Originator: sawyer ( William Barton Sawyer)
Date: Tue Aug 6 09:01:22 MDT 2002
Model: CAM
Version: CAM2.0.dev14
One-line Summary: Patched FV communicator bug in coupled mode; T3 reset to phys_state%t in p_d_coupling; config. improvments
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: SGI
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None (one-line description)
Changes made:
models/utils/pilgrim:
parutilitiesmodule.F90, mod_comm.F90: extended initialization
routines to accept an optional communicator from which
the FV global communicator is determined.
models/atm/cam/bld
configure_fv.pl, fvtest.sh, script.m4: for DAO
code execution, renamed run script fvcam.j,
number of threads, processes renamed to DAO
convention AGCM_N_PROCESSES, AGCM_N_THREADS_PER_PROCESS
models/atm/cam/src/dynamics/fv
dp_coupling.F90 : t3 reset to phys_state%t in p_d_coupling
(suggestion from Brian).
stepon.F90 : modified arg list to p_d_coupling, removed
explicit transpose of T3XY->T3 for 2D decomp.
spmd_dyn.F90 : pass mpicom to ParInit.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_dev13
Originator: jet ( John Truesdale)
Date: Thu Aug 1 16:03:16 MDT 2002
Model: CAM
Version: CAM2.0.dev13
One-line Summary: Fix initialization problem with fractional land, doesn't affect climate, only for CAM not CCSM
cam-bugs Requests resolved: 43
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
(one-line description of changes: Makefile, include files, directories, etc.)
(Detailed description below)
Substantial timing or memory changes: no or Yes (describe)
Requires change in run script: no
(one-line description of changes: data files, run-scripts, namelist etc.)
(Detailed description below)
Tested to work coupled with CCSM: no
Machines tested: IBM
Ran test-model.pl script: no (does all of the following tests)
Tested on fv dynamics: no
Tested on eul dynamics: yes
Tested on sld dynamics: no
Tested that restarts are bit-for-bit: no
Tested multiple constituents: no
Tested that different domain decompositions match bit-for-bit: no
Tested in adiabatic mode: no
Code reviewed by: myself
Restart files change: no
Changes answers: Yes ( new-physics)
(Note: test-model.pl only does checks this -- if you run it with the "-compare" option)
Changes to CLM land-model: None (one-line description)
Changes made:
initext now uses the ts value from the land initialization routine instead
of the value returned from the update surface fluxes routine. I added an
array to save off these values and apply them after iceint is called.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_dev12
Originator: erik ( Erik Kluzek)
Date: Wed Jul 31 14:20:10 MDT 2002
Model: CAM
Version: CAM2.0.dev12
One-line Summary: Add 2D-decomp for FV to bld, Fix Solaris climate, add production test, add timing parser, fix various bugs
cam-bugs Requests resolved: 17, 20, 27, 30, 31, 33, 35, 36, 37, 39
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: yes
(makes use of the _OPENMP CPP token when SMP on, compiler options
must set this)
Substantial timing or memory changes: no
Requires change in run script: Yes
(dif4 now required on namelist if Eulerian and not T42)
Tested to work coupled with CCSM: Yes! (IBM)
Machines tested: IBM, SGI, Solaris, Linux
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, mvertens
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Changes made:
FV 2D parallel decomposition option added to bld scripts:
Requires a new namelist (mprun2d).
Also added a "-fv2d" option to test-model.pl to run fv tests
with 2D decompostion. Also fix the constraint in src/dynamics/fv
so that the npr_yz array from the namelist is checked properly.
Solaris:
Change Makefile so that model gives climate similar to IBM.
Makefile:
Add CPPDEF for CPP definitions used by all platforms. Also
include a new token "CAM" declared when compling code with CAM.
Remove "-d" on AIX, and turn on "-g, -qfullpath" in
production mode so that when the model core-dumps you can
query the core files to find where it died.
dif4:
Now require dif4 to be set on namelist for Eulerian dynamics not
at T42 resolution.
restart-pointer files:
Get rid of getfil for restart-pointer file and add an optional
argument to opnfil, so that "old" can be given for type so that
the model will die if a restart pointer file doesn't exist.
Fix some problems with using scenario_so4:
Fix write statements before "call endrun" so that they will be
written even if not on masterproc. Also broadcast sulfdata to all
nodes. There still seems to be other problems with this configuration.
namelist:
Fix namelist parsing so that repeat declarations can be used
(i.e. mfilt = 4*1). Also update the change method so that it parses
as a regular namelist which allows arrays to be entered and more
complex namelist entries.
test/system:
Add ccm366 test and SMP=FALSE test to test-model.pl (unless on
AIX, ensure that SMP test gives same answers as SPMD). Add "-fv2d"
option to test-model.pl to check FV 2D decompostion. Add some simple
performance checking to test-model.pl. Add test-production.pl
a script to run the production model from Sep to Feb/1 weekly
to ensure the model is running as expected.
check-timing.pl:
Add a script to do straight forward parsing of timing.* files to
look at performance. Simplist operation is as follows:
check-timing.pl directory1 directory2
Options also exist to load the log-files in, and casenames, as
well as a list of timers to parse in the file. To get help invoke
it with "-help" or without any arguments.
ESMF: Update to ESMF_0_0_2p5 (fixes problem with compiling on Sanitas)
This small update changes the name of "conf.h" to ESMF_conf.h
so that it won't find the wrong file in the search path. It also
removes the hard-wired use of /usr/local/include. Also previous
model versions didn't have all files updated to ESMF_0_0_2p4, and
this update fixes that problem.
FV now passes LF95 strict checking tests with test-model.pl:
Arrays in restart_physics:write_restart were going over pcols
rather than cols for FV. By having the arrays only go to ncols
this worked with LF95 and FV.
phys_grid.F90:
Change a difficult to understand OMP loop used to get the
number of threads to a more straight-forward OMP subroutine call.
CCSM:
Update to ccsm2_0_beta50. Add table for dif4 to atm.setup.csh.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_dev11
Originator: sawyer ( William Barton Sawyer)
Date: Tue Jul 30 09:22:20 MDT 2002
Model: CAM
Version: CAM2.0.dev11
One-line Summary: te_map.F90 aligned with latest FVGCM version; optimizations: overlapping communication/computation
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: SGI, Linux
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, W. Putman
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None (one-line description)
Changes made:
src/dynamics/fv/te_map.F90
o Removed CPP token OLDWAY
o Aligned with latest FVGCM version
o Restructured communication (particularly for 2D
decomposition) to overlap communication with computation
as much as possible. Slight performance gains conceivable.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_dev10
Originator: sawyer ( William Barton Sawyer)
Date: Mon Jul 15 10:20:32 MDT 2002
Model: CAM
Version: CAM2.0.dev10
One-line Summary: Merged latest work from DAO FVGCM version fvgcm-1_3_44
cam-bugs Requests resolved: none
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: SGI, Linux
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Lin, Putman, Shen
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None (one-line description)
Changes made:
atm/cam/src/dynamics/fv
sw_core.F90, tp_core.F90, upol5.F90, vpol5.F90 --
changes by Putman allow runs with only 3
latitudes per PE
trac2d.F90 -- cache optimizations by Lin for tracer advection
utils/pilgrim
mod_comm.F90 -- latest version from FVGCM
Makefile.conf.IRIX64 -- support recent change #include
unit_testers/Makefile rather than #include "file.h"
atm/cam/test/system
dao_batch.csh -- batch script to run test-model.pl on DAO
machines (currently tropic and dycore)
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_dev9
Originator: sawyer ( William Barton Sawyer)
Date: Thu Jul 4 14:12:43 MDT 2002
Model: CAM
Version: CAM2.0.dev9
One-line Summary: Bug fixes for 2D domain decomposition, LF95 strict error checking
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Changes made:
in src/dynamics/fv:
inidat.F90 : chunking patch suggested by Pat Worley
restart_dynamics.F90: bufres now allocated on all PEs
(for LF95 strict error checking)
sw_core.F90, tp_core.F90 : revision by S.-J. Lin
crx in tpcc now ghosted +/- ng_c latitudes
benergy.F90, cd_core.F90, trac2d.F90, uv3s_update.F90:
fix for 2D domain decomposition on IBM --
pole PEs now do not accidently send data
to lower or upper levels. Code still does
not work on all IBM 2D configurations but
it is getting closer.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_dev8
Originator: sawyer ( William Barton Sawyer)
Date: Mon Jun 24 02:12:01 MDT 2002
Model: CAM
Version: CAM2.0.dev8
One-line Summary: FV patches for Lahey strict error checking
Requires change in build system: no
ubstantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: SGI, Linux
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: no
Tested on sld dynamics: no
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None (one-line description)
Changes made:
src/dynamics/fv:
mapz_module.F90:Q2 now INOUT in map1_ppm, mapn_ppm,
because Q1 (IN) and Q2 (was OUT)
were being passed same array
sw_core.F90: crx now declared (im,jfirst-ng_d:jlast+ng_d)
passed to tpcc crx(1,jfirst-ng_d).
No error when ng_d == 0 (SMP-only)
tp_core.F90: In tpcc, dummy argument crx now declared
crx(im,jfirst-ng:jlast+ng). No error when
ng (== ng_d) == 0.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_dev7
Originator: erik ( Erik Kluzek)
Date: Thu Jun 20 23:48:55 MDT 2002
Model: CAM
Version: CAM2.0.dev7
One-line Summary: #include "" to <>, Fix Solaris make, Fix SLD for Lahey, change coupled restarts, fix sort, update csm_share/bld/scripts, fix clm restart
cam-bugs Requests resolved: 19, 18, 25, 26
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: yes! (IBM, restarts tested as well)
Machines tested: IBM, SGI (some testing on Solaris, and Linux/Lahey)
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, mvertens, olson (sld)
Restart files change: yes (but only if COUP_CSM set)
Changes answers: no (bit-for-bit)
Changes to CLM land-model: Fix restart pointer files for long casenames
Changes made:
SLD changes for Lahey strict error checking:
Small changes to q arrays sent to a few routines so that
dimension of arrays would pass strict error checking in Lahey
compiler.
Restarts:
Change restarts when running coupled (COUP_CSM on), so that
data not used for coupled runs not output/read. This also allows
model to use the restart files from the b20.007 simulation.
Sort routines:
Change sort routines in responce to bug found by Phil Rasch.
csm_share:
Update csm_share to share2.1.3.
Misc:
#include "" to #include <>
Initialize ctitle and let it be up to 256 characters.
Put "if (masterproc)" around write(6 statements for ramping.
test-model.pl, get rid of global data used for setting up tests.
Replace it with methods. This is a better more robust solution.
Add -qfullpath when debug on for IBM.
Update models/bld, and scripts directories to ccsm2_0_beta47.
Get coupled scripts to work with CAM directory structure, and
change so that restart can work.
CLM:
Don't limit length on string read in from restart pointer file,
as this causes problem when casenames are long.
Solaris:
Change Makefile so that model will work, get rid of special
F77 compilation and use new autopromotion flag in the new compiler.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_dev6
Originator: mvertens ( Mariana Vertenstein)
Date: Thu Jun 20 15:31:15 MDT 2002
Model: CAM
Version: CAM2.0.dev6
One-line Summary: put chunking into coupled model
Requires change in build system: no
cam-bugs Requests resolved: 24
(See http://www.cgd.ucar.edu/cgi-bin/wreq/req?list-5)
Substantial timing or memory changes: Yes
(significant performance improvement when chunking is turned on)
Requires change in run script: no
Tested to work coupled with CCSM: yes
Machines tested: IBM, SGI
Ran test-model.pl script: no (does all of the following tests)
Changes only affected coupled model model
Code reviewed by: myself, erik
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Changes made:
Put chunking into coupled model
Verified bfb with ccsm2_0 and that PCOLS=16 and PCOLS=128 was bfb the same
on the SGI and IBM in coupled mode
Verified that coupled model restarts and branches correctly
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_dev5
Originator: eaton ( Brian Eaton)
Date: Wed Jun 19 19:49:37 MDT 2002
Model: CAM
Version: CAM2.0.dev5
One-line Summary: build procedure mods, memory allocation mods
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM
Ran test-model.pl script: no
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Erik
Restart files change: no
Changes answers: no
Changes to CLM land-model: None
Changes made:
Remove ccm366/cldwat.F90 - not used
Add capability to bld/build-namelist to build LSM namelists
bld/configure
Add -[no]smp option
Add -pcols option
Makefile:
Add SMP macro to allow enabling/disabling openMP without having
to edit Makefile. Currently only implemented for AIX and SGI.
Get rid of the -macro_expand option to SGI f90 because
it's not needed when you use -cpp.
physics/cam1/phys_grid.F90 - move globalfield allocation from stack to heap
in write_field_from_chunk and read_chunk_from_field
dynamics/{eul,fv,sld}/stepon.F90 - move stack memory allocation of
3-D physics and dynamics data structures to heap
models/ice/csim4/ice_tstm.F - change first character of comment lines
starting with c to !
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_dev4
Originator: erik ( Erik Kluzek)
Date: Sat Jun 15 08:04:21 MDT 2002
Model: CAM
Version: CAM2.0.dev4
One-line Summary: Remove precision module, and use csm_share/shr_kind_mod
instead with renames to r8,r4,i8
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, mvertens, eaton
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: Yes, same change to precision as above.
Changes made:
Precision change:
Remove precision module and use csm_share shr_kind_mod module
to get precision types. Use the rename feature to get the names
that were used in the precision module.
Tools change:
Make same change to the tools: cprnc, definesurf, interpic, mkrgrid,
and cprlndnc. In this case keep a local copy of shr_kind_mod for
their use. Made sure each tool changed would build after the change.
Bug-Fix:
Added definition of CPP_INTEGER4 to pilgrim.h as was needed to
get fv to build.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_dev3
Originator: sawyer ( William Barton Sawyer)
Date: Thu Jun 13 07:39:36 MDT 2002
Model: CAM
Version: CAM2.0.dev3
One-line Summary: Solved SUN compilation problem; further merging of FVGCM into CAM; bug fix for SMP-only
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: SGI, SUN
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Putman, Mirin
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None (one-line description)
Changes made:
models/atm/cam/bld:
configure_fv.pl, script.m4, revised for cam2_0 data sets (Mirin)
models/atm/cam/src/dynamics/fv:
pft_module.F90 : merged from FVGCM (minor additions)
sw_core.F90 : merged from FVGCM
cd_core.F90 : bug fix in call of c_sw
modified to use pft_module and sw_core
models/atmlnd_share:
precision.F90 : added i4 needed by DAO (mod_comm)
models/utils/pilgrim:
mod_comm.F90 : fixed SUN compilation problem, upgrade (Putman) additions for STAND_ALONE mode (Sawyer)
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_dev2
Originator: sawyer ( William Barton Sawyer)
Date: Tue Jun 4 09:40:49 MDT 2002
Model: CAM
Version: CAM2.0.dev2
One-line Summary: Merge of FVGCM (DAO) software engineering improvements into CAM2
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM (test-model.pl) SGI (DAO A26, B26 cases)
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Bill Putman, others
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None (one-line description)
Changes made:
This merges cam1_8_dev6_brnch_fvgcm_merge with cam2_0_dev1.
models/utils/pilgrim:
Newest version of mod_comm (but does not yet solve
real -> real(r8) promotion problem on SUN).
Upgrade of other modules for use in CAM-DAS
restructuring and upgrades of unit testers.
models/atm/cam/src/dynamics/fv:
Software engineering improvements from FVGCM (DAO)
benergy.F90 cd_core.F90 d2a3dijk.F90 d2a3dikj.F90
dp_coupling.F90 dynpkg.F90 highp2.F90 hswf.F90
p_d_adjust.F90 te_map.F90 trac2d.F90 uv3s_update.F90
use new mod_comm interface.
mod_mpi.F90 removed.
mapz_module.F90 consolidates vertical interpolation routines:
kmppm.F90 map1_ppm.F90 map3_ppm.F90 ppm2m.F90 steepz.F90
inidat.F90 initcom.F90 pmgrid.F90 spmd_dyn.F90 trunc.F90
minor structural changes
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam2_0_dev1
Originator: erik ( Erik Kluzek)
Date: Mon Jun 3 14:10:23 MDT 2002
Model: CAM
Version: CAM2.0.dev1
One-line Summary: Fix test-model.pl problems, fix problems running Lahey
for Eul dy-core, add T31 namelist options
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM (some testing on Linux with Lahey--lf95)
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Changes made:
T31:
Add namelist options and datasets needed to run T31.
test-model.pl
Fix a couple problems with test-model.pl. Will now properly
run through the list of all dynamics. Start refactoring to a more
general way to run through the tests with object methods rather than
global data. Put quotes around run env variables, allow setting of
interactive option in CAM_namelist, make sure CAM objects have everything
properly set withe either setenv or setcfg.
Lahey fix:
A few changes to dynamics/eul so that qfcst arrays can be properly
dimensioned. Inside spegrd and tfilt_massfix copy qfcst from i1 to nlon+i1-1,
so that strange things with memory don't have to be done. This allows
tests 1 and 2 with test-model.pl to work for Eulerian dynamics. Test 13 fails.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_8_dev11
Originator: erik ( Erik Kluzek)
Date: Fri May 17 10:37:16 MDT 2002
Model: CAM
Version: CAM1.8.dev11
One-line Summary: License update, icefrac bug-fix, build fixes, dataset name changes, fix test-model.pl so that validation could be done
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: yes
(dataset names change)
Tested to work coupled with CCSM: yes! (IBM)
Machines tested: IBM, SGI, Linux-PC-PGI
(some testing on Compaq, Linux-Lahey, and Solaris)
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Changes made:
License update:
Update to CAM license as reviewed by Bill Collins and Catherine
Shea (NCAR lawyer).
icefrac bug-fix:
Fix problem introduced in cam1_8_dev10 on initialization of
icefrac.
Build fixes:
Various fixes to build scripts to work on machines at ORNL.
Name changes:
Dataset names change to be consistent with the following format:
_YYYY-MM-DD__L__.nc
resolution for spectral dycores is refered to as nlatxnlon, with
the spectral truncation following. So for example, Eulerian T42 is
refered to as: 64x128_T42, SLD T63 is 64x128_T63. Datasets that
don't have a specific spectral truncation associated with them
leave off the spectral truncation part (so initial datasets run
through definesurf or most land datasets).
Example:
cami_0000-09-01_64x128_L26_c020514.nc
test-model.pl
Fix test-model.pl so that release version of the model could be
validated.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_8_dev10
Originator: erik ( Erik Kluzek)
Date: Mon May 13 12:01:17 MDT 2002
Model: CAM
Version: CAM1.8.dev10
One-line Summary: Replace Numerical-Recipe code, Lahey fixes for multiple constituents, Update to ccsm2_0_beta45, update to clm2_3_dev30, some Solaris and Compaq fixes
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: yes
Machines tested: IBM, SGI, Linux-PGI, Linux-PGI/GCC
Some testing with Solaris and Compaq, but not everything works.
Ran test-model.pl script: yes (does all of the following tests)
(test-model.pl currently broken test-model ran with previous version)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: Update to clm2_3_dev30
Changes made:
Known problem:
This model core-dumps in the standard configuration due
to a problem with the initialization of icefrac.
Lahey fixes:
Do some fixes for strict error checking with Lahey compiler when
using multiple constituents.
Numerical Recipes sorting routines replaced:
Numerical Recipes routines CAN NOT be publically released, so
they were replaced by Tony Craig's implementation of the same alogrithm.
cprnc:
Fix cprnc so it would work on Solaris. Solution was a Kludge to
get it to work. More work should be done to find the underlying problem.
CCSM2_0_beta45 update:
Update models/bld and scripts to ccsm2_0_beta45.
CLM2_3_dev30 update:
Update CLM to clm2_3_dev30.
test-model.pl
With the new build stuff in place test-model.pl is currently
broken. It will be fixed as soon as possible (which may be after the
model release).
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_8_dev9
Originator: eaton ( Brian Eaton)
Date: Mon May 13 11:22:15 MDT 2002
Model: CAM
Version: CAM1.8.dev9
One-line Summary: merge newbld branch
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: none
Ran test-model.pl script: no - it's officially broken
Tested on fv dynamics: no
Tested on eul dynamics: no
Tested on sld dynamics: no
Tested that restarts are bit-for-bit: no
Tested multiple constituents: no
Tested that different domain decompositions match bit-for-bit: no
Tested in adiabatic mode: no
Code reviewed by: myself, Erik
Restart files change: no
Changes answers: no
Changes to CLM land-model: None (one-line description)
Changes made:
Added new build scripts and user docs.
atm/cam/bld/configure produces the Filepath, misc.h, params.h, preproc.h files,
and a Makefile.
atm/cam/bld/build-namelist produces a namelist file
test-model.pl is currently broken because Erik's mods to have it use CAM_namelist.pm
have not been completed.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_8_dev8
Originator: erik ( Erik Kluzek)
Date: Wed May 8 17:41:26 MDT 2002
Model: CAM
Version: CAM1.8.dev8
One-line Summary: Fix CCSM restart problem
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: yes! IBM
Machines tested: IBM
Ran test-model.pl script: no
Tested on fv dynamics: no
Tested on eul dynamics: yes
Tested on sld dynamics: no
Tested that restarts are bit-for-bit: yes (eul coupled only)
Tested multiple constituents: no
Tested that different domain decompositions match bit-for-bit: no
Tested in adiabatic mode: no
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None (one-line description)
Changes made:
Fix problem where restarts when running coupled were not bit-for-bit.
All changes are isolated to ccsm_msg.F90.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_8_dev7
Originator: jet ( John Truesdale)
Date: Wed May 8 16:22:05 MDT 2002
Model: CAM
Version: CAM1.8.dev7
One-line Summary: definesurf to extend ross ice shelf by default, clean up csim error messages
Requires change in build system: no
(one-line description of changes: Makefile, include files, directories, etc.)
(Detailed description below)
Substantial timing or memory changes: no or Yes (describe)
Requires change in run script: no
(one-line description of changes: data files, run-scripts, namelist etc.)
(Detailed description below)
Tested to work coupled with CCSM: no
Machines tested: IBM
Ran test-model.pl script: yes
Tested on fv dynamics: no
Tested on eul dynamics: yes
Tested on sld dynamics: no
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: no
Tested that different domain decompositions match bit-for-bit: no
Tested in adiabatic mode: no
Code reviewed by: myself
Restart files change: no
Changes answers: no
(Note: test-model.pl only does checks this -- if you run it with the "-compare" option)
Changes to CLM land-model: None (one-line description)
Changes made:
Fix to definesrf to make it extend ross ice shelf by default. Non-land
below S79 is set to land so that land will treat the area as glacier.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_8_dev6
Originator: sawyer ( William Barton Sawyer)
Date: Wed May 8 07:59:26 MDT 2002
Model: CAM
Version: CAM1.8.dev6
One-line Summary: FV patches to help pass lf95 strict error checking, Erik's fix of tphysbc.F90 to pass FV test 1
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: SGI, Linux
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, erik
Restart files change: no
Changes answers: no
Changes to CLM land-model: None
Changes made:
bld: configure_fv.pl and script.m4 upgraded for cam1_8
src/physics/cam1: tphysbc.F90 fix by Erik to pass FV test 1
src/dynamics/fv:
avgc.F90 : retired
cd_core.F90 : inlined avgc.F90
dp_coupling.F90 : intent(inout) u3s (it's updated during ghost)
dynamics_vars.F90: temporary arrays allocated for SPMD only
dynpkg.F90: 2D decomposition arrays now allocated consistently
pmgrid.F90: initialize 2D decomp variables for non-2D case
stepon.F90: 2D decomposition arrays now allocated consistently
utils/pilgrim:
mod_comm.F90: intent(inout) q in BufferUnPack*D
Note: it will pass lf95 strict error checking in SPMD mode on Linux
only if following routines are compiled with DEBUG FALSE:
fv_prints.F90, restart_dynamics.F90 and restart_physics.F90.
Strict error checking yields errors in those routines which
may be related to the chunking mechanism.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_8_dev5
Originator: erik ( Erik Kluzek)
Date: Tue May 7 13:03:41 MDT 2002
Model: CAM
Version: CAM1.8.dev5
One-line Summary: Fix problem with restarts when running coupled.
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: yes! IBM (full "K" test suite)
Machines tested: IBM
Ran test-model.pl script: no
Tested on fv dynamics: no
Tested on eul dynamics: yes
Tested on sld dynamics: no
Tested that restarts are bit-for-bit: No -- not when running coupled!
Tested multiple constituents: no
Tested that different domain decompositions match bit-for-bit: no
Tested in adiabatic mode: no
Code reviewed by: myself, mvertens
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None (one-line description)
Changes made:
Change ONLY ccsm_msg.F90 so that restarts would work. Ran the
full CCSM test GUI on the BIM (blackforest).
This version has a bug in it where restarts are not exact when
running coupled. This problem is fixed in cam1_8_dev8.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_8_dev4
Originator: rosinski ( Jim Rosinski)
Date: Thu May 2 15:12:01 MDT 2002
Model: CAM
Version: CAM1.8.dev4
One-line Summary: xlf90 bugfix for h_override. make clean deletes esmf. flint complaints. modified timing calls. Added lf95 to definesurf makefile.
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, PC
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None (one-line description)
Changes made:
o Bug in xlf90 caused h_override to behave improperly.
o "make clean" now deletes esmf/ directory as well as .o, etc. files
o Fixed some flint complaints.
o Added lf95 entry to definesurf makefile.
o Modified some of the calls to the timing library.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_8_dev3
Originator: erik ( Erik Kluzek)
Date: Wed May 1 22:25:25 MDT 2002
Model: CAM
Version: CAM1.8.dev3
One-line Summary: Output field changes, cam1 to cam2, FLAND to LANDFRAC,
fix vertinterp, longname for constituents
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: yes
(Restart pointer files are now cam2 instead of cam1, restart files have cam2 in name)
Tested to work coupled with CCSM: Yes! (SGI and IBM)
Machines tested: IBM, SGI, Linux-PGI
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, mvertens (and input from wcollins, pjr, and boville)
Restart files change: no
Changes answers: no (bit-for-bit) (although DTCOND different)
Changes made:
Output field changes:
DTCOND -- Changed to be defined by dry static energy instead of
temperature to be more consistent with current alogrithms.
This change is from Fabrizio Sassi.
New fields:
WSPEED -- Wind speed (default maximum)
FSDSC --- Clearsky downwelling solar flux at surface
PRECTMX - Maximum precipitation rate
(Add above to monthly files)
Put TREFMNAV and TREFMXAV on output monthly files for non-coupled case.
These changes are in ccsm-bugs as request 19.
Change instances of cam1 to cam2:
Change occurances of cam1 to cam2 in output file names and code comments.
FLAND on initial files is now LANDFRAC:
Have initial files that are produced use the name LANDFRAC instead
of FLAND to be consistent with history files. This also put attributes on
the field which didn't happen with the FLAND name. For backwards compatability
allow the names on the initial files to be either LANDFRAC or FLAND.
vertinterp:
Apply a fix to vertinterp as found by Art Mirin. This fixes a problem
if the interpolation pressure happens to exactly equal the pressure
at a data-point. Recorded in ccsm-bugs as request 29.
Constituents module now has longname data element:
Add longname to constiuents module and set it as an optional argument
in add_cnst. When add_cnst called if appropriate set longname to
something useful.
Coupled:
Change scripts to look at inputdata/atm/cam2 input directory.
Fix log-file output problem by adding a if( masterproc) on
the "stopping at end of day" print.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_8_dev2
Originator: erik ( Erik Kluzek)
Date: Sat Apr 27 09:39:39 MDT 2002
Model: CAM
Version: CAM1.8.dev2
One-line Summary: Update to cam1_8_dev1_brnchT_esmfup4 --
Update to ESMF_0_0_2p4, clm2_3_dev29, ccsm2_0_beta41, simple fixes,
nix scatter/gather aliasing for CCSM
Requires change in build system: Yes (ESMF build changes)
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: Yes! (SGI and IBM)
Machines tested: Some testing on all platforms:
IBM, SGI, Compaq, Solaris, Linux-PGI, Linux-Lahey, Linux-PGI/GCC
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, mvertens
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: Update to clm2_3_dev29
Changes to ESMF: Update to ESMF_0_0_2p4
Changes made:
ESMF update: Update to newer version of ESMF with name changes from
mf_ to esmf_. The newer ESMF simplifies the build by not needing
dependence on OMP or MPI, so nothing is referenced outside the
ESMF distribution.
CLM2 update: Update to clm2_3_dev29 which works with ESMF version above.
CCSM Update: Update models/bld and scripts directories to ccsm2_0_beta41
version of the scripts. Also put PCOLS define in CCSM scripts rather
than hardcoded in ppgrid.F90. Ran csm.csh test on babyblue and utefe.
Bit-for-bit with non-SPMD version and bit-for-bit with previous
model.
CCSM Scatter/Gather aliasing: Get rid of the aliasing (using the same array
for input and output) in the mpigather and mpiscatter calls in
ccsm_msg.F90. In the process create a new allocatable array allocated
on masterproc and an old array allocatable so that only dimensioned
from beglat:endlat. This reduces the total amount of memory needed
for the mpi scatters and gathers.
Solaris: Fix a couple subscript problems caught on solaris in history.F90.
Then as a particularly difficult problem was caught, turn subscript
checking off for solaris when DEBUG=TRUE.
phys_grid.F90: Get updates from Pat Worley and Art Mirin to generalize
phys_grid.F90 in cam1 for odd numbers for FV dynamics. Tried same
routine in ccm366 physics, but results in problems.
ORNL -Colt: Add settings needed to CAM_lab.pm to run on colt.ccs.ornl.gov.
Testing: Ran at least some test-model.pl tests on all platforms:
Linux-PGI (Apache, longs), Linux-Lahey (Apache, compile test on longs)
Linux-PGI/GCC (Apache, longs), SGI (utefe), IBM (babyblue, eagle@ornl),
Solaris (sanitas, test 1 works), Compaq (colt@ornl tests 2-5 with SPDM off)
There are problems with some of the standard tests in test-model.pl
on particular platforms. These problems will need to be cleanup in
before model release.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_8_dev1
Originator: erik ( Erik Kluzek)
Date: Mon Apr 22 13:11:18 MDT 2002
Model: CAM
Version: CAM1.8.dev1
One-line Summary: Move cam1_7_dev34_brnchT_fractional_8 branch to the development branch
Requires change in build system: Yes!
Filepath requires new directories
Substantial timing or memory changes: Yes (running with new ice-model is slower)
Requires change in run script: Yes
New set of SST and Initial condition files are required.
Tested to work coupled with CCSM: Yes! (IBM and SGI)
(bit-for-bit with ccsm2_0_beta38)
Machines tested: IBM, SGI, Compaq, Sun, Linux-Lahey, Linux-PGI
(Sun compiles but doesn't run, Compaq core-dumps on error-growth
test, Linux-PGI core-dumps on error-growth test with fv dynamics)
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, John Truesdale, Mariana Vertenstein, Jim Rosinski
Restart files change: yes!
Changes answers: new-physics
Changes to CLM land-model: Interface changes
Control cases:
/EATON/csm/fv2x-fland05 -- Finite-Volume with PCMDI SST's.
/JET/csm/sldnewicefrac01 -- SLD dynamics with PCMDI SST's.
/ROSINSKI/csm/newsstclim03 - Eulerian dynamics with Hurrell
Hadley Center OI / Reynolds blended SST data climatology.
/ROSINSKI/csm/newsstamip02 - Eulerian dynamics with Hurrell
blended monthly SST data.
Changes made:
Physics:
Fractional land and ocean: Old binary ORO flag is gone replaced
by the fractions: landfrac, icefrac, and ocnfrac. Also the old
flag on SST datasets is done away with and ice-cov is used for
sea-ice concentrations. comsrf.F90 was reworked as part of this change.
Also the offline tool "definesurf" was changed to work with the
new datasets. "interpic" continues to work with IC files as well.
New ice-model: A modified version of the CCSM Sea-Ice model
Community Sea-Ice Model (csim4) is included with the model
now in the "models/ice/csim4" directory. The old ice-model is
included in the new directory "models/atm/cam/src/ocnsice/dataicemodel".
CCM3.6.6 Physics:
The CCM3.6.6 physics is still supported although now it can ONLY
be run with Eulerian dynamics/LSM and the old data ice/sst model.
The old version of the data ice model codes are in the "physics/ccm366"
directory.
Namelist changes:
prognostic_icesnow -- If TRUE determine snow-depth over ice prognostically
otherwise use a climatology.
reset_csim_iceprops -- Reset the CSIM ice-model initial properties
(used on a new initial run if the model blows up).
Script / Makefile changes:
Filepath, datasets and new variable "ICEMODEL" added to deal with
the two ice-models (CSIM4 or CCMDATA). Lahey error-checking added
to Makefile. Initial datasets now require the land-fraction on
them in the variable "FLAND". When using the new ice model
SST datasets are required to have the variables "SST_cpl" (for SST), and "ice_cov" for sea-ice concentration.
"NEWBUILD=TRUE" option now the only way to use the standard makefile.
The old "makdep" dependency generator logic ripped out of the Makefile
and deleted from the "tools/makdep" directory.
Users guide updated to reflect changes. DatasetLog updated with
new datasets.
New datasets:
To obtain the new tar-ball of datasets:
ftp ftp.cgd.ucar.edu
cd pub/erik
bin
get cam1_8.scidac-atm.datasets.tar.gz
Error-growth:
Error-growth is much faster than previous versions of the model.
Error-growth is still high using the dataicemodel and active-land.
Using prognostic_icesnow=.false. improves error-growth, and using
standard T42L26 datasets provides reasonable error-growth. Also
running error-growth with aqua_planet on provides reasonable error-growth.
Coupled model:
landfrac, and ocnfrac are now binary values of 0.0 or 1.0 in
accordance to how ORO was previously assigned in the model. This
should eventually be changed to use the values the coupler send
explicitly.
test-model.pl:
Change restart tests to restart after 5 time steps.
Change error-growth full-physics test to use prognostic_icesnow=.false.
and aqua_planet=.true. and use date from dataset rather than Dec/31.
Delete the SOM test and add a new test to make sure using the
dataicemodel works.
Fix problem with recognizing old files.
interpsst gone:
The old tools/interpsst code removed as it will not work with the
new SST files. Also new tools for creating SST datasets are in
development. The new tools use the Hurrell SST datasets and also
modify data so that when a linear time-interpolation is applied
to the mid-monthly datapoints the observed monthly average is
reconstructed. It's unclear when the new tools will become part
of the distribution.
Changes to make most of the model Lahey compliant:
Small changes to large amounts of the code so that strict error-checking
with the Lahey compiler can be performed on most of the model. The
files: binary_io.F90, wrap_nf.F90, wrap_mpi.F90 have not been converted.
As part of this change the routines bouyan, and cldprp where moved
to inside the zm_conv.F90 file.
Bug-fixes:
Fix in phys_grid.F90 for SGI.
Fix in filenames.F90 for IBM.
Known problems:
IBM -- history.F90 still doesn't always compile when DEBUG=TRUE.
Compaq -- Model core-dumps with "floating invalid" when running
error-growth test on Compaq (Colt @ ORNL)
The model compiles but has trouble running on the Sun. Subscript overflow
problem when running compiler checking (Sanitas @ NCAR).
Linux-PGI -- fv: When running FV dynamics and Error-growth the model
aborts with an error
"DEALLOCATE: memory at 43d18c10 not allocated"
(Apache @ NCAR)
SOM -- New model does not work with the slab-ocean model.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_7_dev34
Originator: rosinski ( Jim Rosinski)
Date: Fri Mar 29 16:44:22 MST 2002
Model: CAM
Version: CAM1.7.dev34
One-line Summary: Bugfix to tphysbc.F90 to keep uninitialized data (in ptend) from being used.
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, PC
Ran test-model.pl script: yes (does all of the following tests)
NOTE: fv tests on IBM failed for both this commit and the comparison
previous library due to compiler bugs. But all ran OK on PC.
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit) presuming the short tests are sufficient to ensure no garbage
from stack being used
Changes to CLM land-model: None
Changes made:
o Contents of local variable ptend in tphysbc were being used uninitialized. No ill symptoms were
noticed in the dev branch, but the fractional branch was randomly seg faulting or encountering
a floating point error when restarting on 16 nodes on blackforest.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_7_dev33
Originator: pworley ( Patrick H Worley)
Date: Tue Mar 12 14:30:32 MST 2002
Model: CAM
Version: CAM1.7.dev33
One-line Summary: load-balanced physics option; PCOLS cpp token for setting pcols; SOM chunking bug fix
Requires change in build system: yes
PCOLS cpp token required to set pcols in ppgrid.F90. params.h.in.csh modified to do this automatically.
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM and Compaq at ORNL
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself and Marcia Branstetter
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Changes made:
1) Changed phys_grid_init call. Instead of specifying number of
columns per chunk, now specify number of chunks per thread.
This then sets the number of columns per chunk appropriately,
guaranteeing that the same number of chunks can be assigned
to each processor. Files affected:
control/restart.F90
control/wrap_mpi.F90
dynamics/eul/inital.F90
dynamics/sld/inital.F90
dynamics/fv/inital.F90
physics/cam1/phys_grid.F90
physics/ccm366/phys_grid.F90
2) Augmented physics domain decomposition options:
opt == -1: same as before (latitude-slice chunk, for compatibility
with ccm366 physics, lsm1, and CSM)
opt == 0: For a given MPI process, all chunks have the same number
of columns, and each thread associated with a process has
the same number of chunks. Columns are assigned to chunks
to improve load balance between chunks (compared to
the default ordering), but without requiring
interprocessor communication in dp_coupling.
opt == 1: All chunks have the same number of columns and each
thread has the same number of chunks, across all MPI
processes. Columns are assigned to chunks to balance
day/night, summer/winter "static" load imbalances.
Chunks are assigned to processes to minimize interprocessor
communication in dp_coupling, but it will occur
anyway unless the dynamics decomposition is compatbile.
Files affected:
dynamics/eul/dp_coupling.F90
dynamics/eul/dyn_grid.F90
dynamics/eul/inital.F90
dynamics/fv/dp_coupling.F90
dynamics/fv/dyn_grid.F90
dynamics/fv/inital.F90
dynamics/sld/dp_coupling.F90
dynamics/sld/dyn_grid.F90
dynamics/sld/inital.F90
physics/cam1/phys_grid.F90
physics/ccm366/phys_grid.F90
control/wrap_mpi.F90
control/restart.F90
3) Added PCOLS cpp token, used to set pcols parameter in ppgrid.F90 .
When running coupled, pcols (still) set to PLOND. params.h.in.csh
modified to set PCOLS default, taking care of LSM1 and CCM366 physics
special cases. (Code courtesy of Erik Kluzek.)
Files affected:
../bld/README
../bld/params.h.in.csh
physics/cam1/ppgrid.F90
physics/ccm366/ppgrid.F90
4) Fixed chunking bug in SOM. Files affected:
ocnsice/som/somoce.F90
ocnsice/som/srfsomi.F90
ocnsice/som/srfsomo.F90
ocnsice/som/srftsbi.F90
5) Miscellaneous clean-up (removing pchnk and pchnkd parameters,
renaming a phys_grid routine). Files affected:
physics/cam1/physgrid.F90
physics/cam1/ppgrid.F90
physics/cam1/cldwat.F90
physics/ccm366/physgrid.F90
physics/ccm366/ppgrid.F90
control/history.F90
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_7_dev32
Originator: erik ( Erik Kluzek)
Date: Thu Mar 7 23:00:41 MST 2002
Model: CAM
Version: CAM1.7.dev32
One-line Summary: Bug-fix to cam1_7_dev30 snowfall rates, also fix interpret_filename_spec
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: Yes!
Machines tested: IBM, SGI (some testing on Linux, SunOS as well)
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, mvertens, tcraig
Restart files change: no
Changes answers: Yes (fixes bug from cam1_7_dev30 where everything appears as snow)
Changes to CLM land-model: fix problem where numpatch was "use"'d twice
Changes made:
tphysbc.F90:
Fix bad merge from cam1_7_dev30 that caused all precipitation to fall
as snow.
filenames.F90:
Check if prev is present before using. Without this the SGI and Compaq
would seg-fault when running the model.
test-model.pl:
Fix problem in test 6. Tests skipped don't build-namelist.
run-model.pl:
Don't resubmit till year 1 after first simulation.
clm:
Fix double "use" of numpatch.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_7_dev31
Originator: mirin ( Arthur Andrew Mirin)
Date: Thu Mar 7 10:11:37 MST 2002
Model: CAM
Version: CAM1.7.dev31
One-line Summary: Completed full history capability for 2D FV decomposition
Known bug: All precipitation appears as snowfall!
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Changes made:
Added 32-bit accumulation buffers and initial history file capability for 2D FV decomposition. This involved a number of changes to history.F90, minor changes to stepon.F90, enhancements to restart_dynamics.F90, and enhancements (made by Will Sawyer) to the Pilgrim library (puminterfaces.F90, pilgrim.h, parutilitiesmodule.F90, redistributemodule.F90). Also made a minor unrelated improvement to trac2d.F90 (elimination of 2D temporary array).
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_7_dev30
Originator: erik ( Erik Kluzek)
Date: Wed Mar 6 08:58:37 MST 2002
Model: CAM
Version: CAM1.7.dev30
One-line Summary: Fix labels on monthly files, physics changes from mzh6,
fix CCSM logs, update to ccsm2_0_beta38
Known bug: All precipitation appears as snowfall!
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: Yes!
Machines tested: IBM, Linux (PGI)
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, mvertens
Restart files change: no
Changes answers: Yes! (same-physics -- less snowfall, reduced Cn over sea-ice)
Case names for changes: /CCSM/csm/b20.001
Changes made:
Bug introduced:
This version introduces a bug where all precipitation appears as snow.
This was due to incorrectly resolving a conflict in the merge from mzh5 to
mzh6.
Fix labels on monthly files:
Monthly files were labeled with the current time-stamp. Now they are labeled
with the previous month. Also don't output a monthly file, if nstep == 0,
as it will just be an instantaneous time-sample rather than an average.
Get code updates from ccm3_12_47_brnchT_mzh6:
Snowfall is reduced by requring average temperature to be -2C. Reduce cloud
condensation nuclei over sea-ice from 40 to 5. These code changes were tested
coupled, but not stand-alone yet. Both a new coupled and uncoupled simulation
will be run shortly.
Fix log-files when running coupled:
Only redirect masterproc output to atm.log file. Redirecting all output was
causing conflicts in the log file and causing most of the information
to be lost. Also add more checks on masterproc for output when running coupled.
Update CCSM scripts to ccsm2_0_beta38:
Update the "models/bld" and "scripts" directories to point to the "beta38"
version of CCSM2.0.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_7_dev29
Originator: erik ( Erik Kluzek)
Date: Fri Mar 1 20:53:28 MST 2002
Model: CAM
Version: CAM1.7.dev29
One-line Summary: Fix initial filenames and increase caseid from 16 to 32 characters
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM (eul,sld,fv), SGI (eul only)
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: Increase caseid from 16 to 32
Changes made:
Increase caseid from 16 to 32 characters.
Fix filenames of initial files (the .nc extension was dropped off).
Change test-model.pl. SOM test now run on all platforms (AIX was left
off before). $CASE is a little longer and more descriptive. Filewildcard
more explicit (solves a simple problem I ran into, where the old file
was found).
Ran run-model.pl with a 32-character caseid (both with CLM and LSM)
producing initial files to make sure worked with long caseid, and
initial files had proper name.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_7_dev28
Originator: erik ( Erik Kluzek)
Date: Fri Feb 15 17:56:56 MST 2002
Model: CAM
Version: CAM1.7.dev28
One-line Summary: Add new flexible filenames with CCSM names as defaults, update to CLM2.3.17
Requires change in build system: yes
(SHELL_MSS is no longer required)
Substantial timing or memory changes: no
Requires change in run script: no
(namelist options changed: rirt, nrmvn, nsmvn removed, irt => mss_irt,
nswrps => mss_wpass, add hfilename_spec, archive_dir option)
Tested to work coupled with CCSM: yes!
Machines tested: IBM, SGI (some testing on Linux PGI, Sun)
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, mvertens, eaton
Restart files change: yes!
Changes answers: no (bit-for-bit)
Changes to CLM land-model: Update to CLM2.3.17, get to work with changes
Changes made:
Flexible filenames:
New namelist option, hfilename_spec to specify filenames, newly
adopted CCSM standard filenames are the default. Filename specifiers set
the filenames, with the following interpretations:
%c = casid
%t = tape number
%y = current year
%m = month
%d = day
%s = seconds into current day
%% = % character
By default non-monthly files have a hfilename_spec of: "%c.cam1.h%t.%y-%m-%d-%s.nc"
Hence, for example, with a caseid="test", tape number 2, writing it's first data
for 12/1 UT 0:0 for year 1, will have the filename:
"test.cam1.h1.0001-12-01-00000.nc".
Namelist:
nsmvn, nrmvn, and rirt removed from namelist.
old irt renamed to mss_irt
old nswrps renamed to mss_wpass
Added hfilename_spec filename specifyer to history filenames.
Added archive_dir to give MSS archive directory
(now defaults to /USERNAME/caseid/csm/atm).
Filenames module:
New module to deal with filenames, and MSS options and interpret
filename specifiers (as above). Unit-test for this module was created.
Restarts:
history and abs-ems restart full filepaths added to master restart file.
Restart pointer file changed to include information on all files needed to
successfully do a restart. Restart files are disposed to MSS when closed
rather than in wrapup.
History:
History filenames are created and opened when data is first written to
them. Added methods: get_mtapes, get_hist_restart_filepath, get_hfilepath
to get private data in history module. Increase length of filenames to 256.
CCSM scripts:
models/bld updated to ccsm2_0_beta35, scripts updated to
scripts_a011228_brnchT_a020201. system_test script updated to changes in these
scripts. atm.setup.csh updated for new namelist changes.
test-model.pl:
New SOM dataset names added. Changed test-model.pl to use NEWBUILD
mechanism which should speedup build, changed so that error-growth tests
run without trace-gas, but with readtrace to exercise more code options. Set
nhtfrq(1) for restart tests so that all history files produce restart files.
Change Makefile so that explicit path to mkSrcfiles and mkDepends scripts
is given (using Rootdir).
CLM:
CLM updated to CLM2.3.17 (uses new CCSM filename convention). Update
to filenames module and namelist change.
Move files unused by CLM from atmlnd_share to control:
(Testing of files in atmlnd_share is insufficient to work with
CLM2. So we are working on seperating the two.)
ioFileMod.F90, gauaw_mod.F90, wrap_mpi.F90
ioFileMod.F90 changes so that comments, and mvn is not needed, and
SHELL_MSS option becomes standard.
MSS:
Remove MSS comments when mswrite done, and remove SHELL_MSS CPP token,
since always used.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_7_dev27
Originator: eaton ( Brian Eaton)
Date: Fri Feb 8 18:19:23 MST 2002
Model: CAM
Version: CAM1.7.dev27
One-line Summary: bugfix for CAM makefile
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, PC-Linux
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Erik
Restart files change: no
Changes answers: no
Changes to CLM land-model: None
Changes made:
The CAM build was failing when the new ESMF macro MF_BUILD was set to a
relative path. Added fix to CAM makefile to convert the relative pathname
to an absolute one.
Misc: modified bld/condense-path.pl to look for CAM.pm in $MODEL_CFGDIR.
This allows bld/configure.csh to be used in a directory other than bld.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_7_dev26
Originator: eaton ( Brian Eaton)
Date: Wed Feb 6 14:52:17 MST 2002
Model: CAM
Version: CAM1.7.dev26
One-line Summary: update esmf to MF_LLU_0_0_12p1 and add fixes for lf95
Requires change in build system: yes
CAM Makefile modified to set MF_BUILD macro used by esmf makefile.
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, PC-Linux
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Changes made:
David Neckels' summary of changes in MF_LLU_0_0_12p1:
> Version: MF_LLU_0.0.12
> One-line summary: Added relocatable build support.
> Changes made:
>
> -Added variable MF_BUILD, which allows the code to be built in a
> directory other than the default. To use, invoke:
> gmake BOPT=g MF_BUILD=/tmp/esmf_build
> This will build the library under "/tmp/esmf_build". The object files,
> the library and the mod files will all be created under the new
> directory.
>
> -Added an install target to makefile. The variables MF_LIB_INSTALL
> and MF_MOD_INSTALL determine where the libraries and mod files are
> copyed during the install: (e.g)
> gmake BOPT=g MF_BUILD=/tmp/esmf_build MF_LIB_INSTALL=/usr/lib \
> MF_MOD_INSTALL=/usr/mod install
> (Notice that once the MF_BUILD is used, it must be passed to the
> subsequent make commands).
The CAM Makefile was changed to set the MF_BUILD directory to $(MODEL_EXEDIR)/esmf.
The latest lf95 compiler (lf9561) required the following changes to the source code:
physics/cam1/{radae.F90,chemistry.F90} - removed save attribute in variable
declarations that were in the same scope as a save statement.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_7_dev25
Originator: eaton ( Brian Eaton)
Date: Tue Feb 5 12:04:52 MST 2002
Model: CAM
Version: CAM1.7.dev25
One-line Summary: remove prognostic sulfur cycle stubs
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, PC-Linux
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Changes made:
Only the interface to the prognostic sulfur cycle is being removed. The
following independent features have not been removed: 1) a specified
sulfate aerosol distribution may be read from a dataset, 2) a ramp function
may be applied to the specified sulfate distribution, and 3) the indirect
radiative effects of sulfate aerosols may be estimated.
Modify files:
control/history.F90
control/initext.F90
control/parse_namelist.F90
control/restart.F90
dynamics/eul/inidat.F90
dynamics/fv/inidat.F90
dynamics/sld/inidat.F90
physics/cam1/advnce.F90
physics/cam1/initindx.F90
physics/cam1/inti.F90
physics/cam1/moistconvection.F90 - left icwmr in arg list of cmfmca.
physics/cam1/radctl.F90
physics/cam1/tphysac.F90
physics/cam1/tphysbc.F90
physics/cam1/tracers.F90
physics/cam1/zm_conv.F90
physics/ccm366/advnce.F90
physics/ccm366/initindx.F90
physics/ccm366/tracers.F90
Remove files:
control/acbnd.F90
control/clddiag.F90
control/dmsbnd.F90
control/drydep.F90
control/massbgt.F90
control/scyc.F90 - move the indirect flag into comctl.h. Use this flag
directly rather than via the doindirect query method.
control/soxbnd.F90
control/sulbnd.F90
control/sulchem.F90
control/sulemis.F90
control/wetdep.F90
Misc:
Change variable count to count1 in history.F90(wshist) to fix a problem
with the pgf90 compiler. count is the name of an F90 intrinsic function
and apparently using it as a local variable was causing a problem.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_7_dev24
Originator: olson ( Jerry Olson)
Date: Thu Jan 31 17:06:44 MST 2002
Model: CAM
Version: CAM1.7.dev24
One-line Summary: modifications and bufixes to eulerian, sld, and physics
routines
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI
Ran test-model.pl script: yes (IBM only)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Restart files change: no
Changes answers: EUL: no (bit-for-bit; unless "DIVDAMPN" is used in namelist)
SLD: Yes (same-physics)
Changes to CLM land-model: None
Changes made:
bugfixes:
- divergence damper in Eulerian code
src/dynamics/eul/hordif.F90
- SLD energy fixer
src/physics/cam1/comsrf.F90
src/physics/cam1/physpkg.F90
src/physics/cam1/radctl.F90
src/physics/cam1/tphysbc.F90
other mods
- SLD vector-field advection routine
src/dynamics/sld/nunv1.F90
- src/advection/sld/engy_tdif.F90
- removed an option from idealized physics
src/physics/cam1/phys_idealized.F90
src/physics/cam1/tphysidl.F90
- modified "test_trace3" tracer
src/physics/cam1/tphysac.F90
===============================================================
===============================================================
cam1_7_dev23
Originator: sawyer ( William Barton Sawyer)
Date: Wed Jan 16 13:46:35 MST 2002
Model: CAM
Version: CAM1.7.dev23
One-line Summary: Upgraded fvCAM to newest mod_comm primitives
Requires change in build system: no
Substantial timing or memory changes: possibly faster
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Putman (mod_comm primitives)
Updated the DatasetLog (for new input data-files): not-applicable
Updated the documentation: not-applicable
Changes answers: no (bit-for-bit)
Name of tests cases on NCAR Mass-Store: (For changes that change answers)
Changes to CLM land-model: None (one-line description)
Changes made:
models/utils/pilgrim:
removed bufferpack/unpack 2d,3d,4d primitives (in mod_comm)
mod_comm.F90: added generic mp_send/recv primitives
models/atm/cam/src/dynamics/fv
cd_core.F90 d2a3dijk.F90 d2a3dikj.F90 dynpkg.F90
highp2.F90 hswf.F90 p_d_adjust.F90 te_map.F90 uv3s_update.F90
replaced parbegin/endtransfer primitives with mp_send/recv
primitives from mod_comm. The only tricky part about this
was determining the proper source and destination PE numbers:
in the 2D decomposition this depends on the ordering of
the mapping of the decomposition to the PEs (row-major or
column-major).
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_7_dev22
Originator: mirin ( Arthur Andrew Mirin)
Date: Tue Jan 15 14:49:36 MST 2002
Model: CAM
Version: CAM1.7.dev22
One-line Summary: Support history files for 2D FV decomposition
Requires change in build system: no
(one-line description of changes: Makefile, include files, directories, etc.)
(Detailed description below)
Substantial timing or memory changes: no
Requires change in run script: no
(one-line description of changes: data files, run-scripts, namelist etc.)
(Detailed description below)
Tested to work coupled with CCSM: no
Machines tested: IBM
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Sawyer
Updated the DatasetLog (for new input data-files): not-applicable
Updated the documentation:not-applicable
Changes answers: no (bit-for-bit)
(Note: test-model.pl only does checks this -- if you run it with the "-c" option)
Name of tests cases on NCAR Mass-Store: (For changes that change answers)
Changes to CLM land-model: None (one-line description)
Changes made:
Changed history.F90, uv3s_update.F90 and restart_dynamics.F90.
Also changed CAM_lab.pm and script.m4.
Main changes were to history.F90, in order to support 2D FV decomposition.
The field_info structure was generalized to include a vertical extent. Support
for dynamics variables having global vertical extent 1, plev and plevp was added;
however, only tested variables had global extent plev. New coding supports 64-bit
accumulation buffers. 32-bit accumulation buffers are still not supported for
2D FV but are supported for 1D FV decomposition.
Previous coding had bug involving non-preservation of subscript array
range for associated pointers. That bug affected the hbuf_accum routines and
hbuf_compute_avg. Bug has been fixed; inside those routines relevant arrays
have beginning subscript equal to 1.
uv3s_update.F90 needed minor fix to properly call the history routine.
restart_dynamics.F90 had a minor bug involving 2D FV decomposition.
script.m4 (DAO) was given 1-line change involving ccm to cam transition.
CAM_lab.pm had 1-character change involving working directory on LLNL frost.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_7_dev21
Originator: sawyer ( William Barton Sawyer)
Date: Mon Jan 7 07:38:50 MST 2002
Model: CAM
Version: CAM1.7.dev21
One-line Summary: Merge of current DAO fvGCM FV dycore into CAM baseline: ghosting of variables, mod_comm for communication
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Linux
Ran test-model.pl script: no (test-model.pl -l dao not currently working)
Tested on fv dynamics: yes
Tested on eul dynamics: no
Tested on sld dynamics: no
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: no
Code reviewed by: myself
Updated the DatasetLog (for new input data-files): no
Updated the documentation: not-applicable
Changes answers: no (bit-for-bit)
Name of tests cases on NCAR Mass-Store: n/a
Changes to CLM land-model: None (one-line description)
Changes made:
Phase 2 of fvGCM-CAM fv dycore merge: ghosted prognostic
variables U3S, V3S, Q3. Uses mod_comm for communication
(as well as pilgrim). More consistent use of "only" statement
in "use" lines. Upgrade of pilgrim (synchronized with DAO
Chemistry Transport Model development) and mod_comm (should
compile on Compaq now).
models/utils/pilgrim:
pilgrim.h, parutilitiesmodule.F90, ghostmodule.F90:
synchronized with DAO CTM development.
mod_comm.F90 : should now compile on Compaq (sorry,
not tested -- no access to a Compaq)
models/atm/cam/src/control:
history.F90: rather ugly code to support write of
ghosted U3S array. Well be improved in later version.
models/atm/cam/src/dynamics/fv
prognostics : ghosted variables U3S, V3S, Q3
cd_core.F90, trac2d.F90 : ghosting of U, V, Q3, replaced many
pilgrim calls with mod_comm.
benergy.F90, dryairm.F90, inidat.F90, restart_dynamics.F90
map1_ppm.F90, d2a3dijk.F90, d2a3dikj.F90, dynpkg.F90,
te_map.F90, fv_prints.F90, p_d_adjust.F90, dp_coupling.F90,
hswf.F90, uv3s_update.F90 : accommodated ghosted variables.
mod_mpi.F90 : retired (now uses mod_comm)
dynamics_vars.F90 : new temporary unghosted variables as
interim solution for 2D decomposition redistributions.
stepon.F90 : ghosted variable u3sxy
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_7_dev20
Originator: rosinski ( Jim Rosinski)
Date: Wed Jan 2 16:48:01 MST 2002
Model: CAM
Version: CAM1.7.dev20
One-line Summary: Mods to SOM: MLD renamed MLDANN on SST dataset., and contains
yearly values. SPMD SOM indexing fix. No resetting SICTHK,
SNOWH, TSSUB in SOMINI
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, PC
Ran test-model.pl script: no
Tested on fv dynamics: no
Tested on eul dynamics: yes
Tested on sld dynamics: no
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: no
Tested that different domain decompositions match bit-for-bit: no
Tested in adiabatic mode: no
Code reviewed by: myself
Updated the DatasetLog (for new input data-files): no
Updated the documentation: no
Changes answers: no (bit-for-bit) at least for non-SOM configurations
Changes to CLM land-model: None
Changes made:
NOTE: All changes are to slab ocean model (SOM) code.
o Changed name of mixed layer depths to MLDANN on SST boundary file
o Fixed indexing bug in SPMD mode (SOMINI, SOMINT)
o Removed code from SOMINI which reset SICTHK, SNOWH, TSSUB, ORO.
===============================================================
cam1_7_dev19
Originator: sawyer ( William Barton Sawyer)
Date: Thu Dec 27 07:47:01 MST 2001
Model: CAM
Version: CAM1.7.dev19
One-line Summary: FV core upgrade: Q3 indexing changed to (i,j,k,iq); no changes in order on history/restart files
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Art Mirin
Updated the DatasetLog (for new input data-files): not-applicable
Updated the documentation: not-applicable
Changes answers: no (bit-for-bit)
Name of tests cases on NCAR Mass-Store: (For changes that change answers)
Changes to CLM land-model: None (one-line description)
Changes made:
Changed indexing of prognostic var. Q3 from (i,k,iq,j) to
(i,j,k,iq) in FV dynamical core. This change affected
numerous files in FV,
cd_core.F90 fv_prints.F90 pmgrid.F90 stepon.F90
dp_coupling.F90 inidat.F90 prognostics.F90 sw_core.F90
dryairm.F90 mod_mpi.F90 restart_dynamics.F90 te_map.F90
dynpkg.F90 p_d_adjust.F90 spmd_dyn.F90 trac2d.F90
and one common file: src/control/history.F90.
The section in history is protected by an "if ( dycore_is('LR) ) then"
line (and is also embedded in an #ifdef STAGGERED CPP statement --
does any other dycore use STAGGERED?)
The order of Q3 on the initial/restart files has *not* changed.
For the time-being there is logic to read and write Q3 in the
original order. This might change (in restart_dynamics.F90) later on.
models/util/pilgrim/mod_comm.F90 has also been upgraded with
new routines; which are not yet used by the FV core. A code
review/revision of these changes by Putman/Shen is pending.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_7_dev18
Originator: eaton ( Brian Eaton)
Date: Tue Dec 18 12:40:35 MST 2001
Model: CAM
Version: CAM1.7.dev18
One-line Summary: changes for share constants
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Compaq, sun, pc/linux/pgf
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: Erik, myself, and Mariana
Updated the DatasetLog (for new input data-files): no
Updated the documentation: no
Changes answers: Yes, new constants sometimes agree to only 2 digits with old ones. This tag requires scientific validation.
Name of tests cases on NCAR Mass-Store: Erik will do production runs.
Changes to CLM land-model: Updated to clm2_3_dev10.
Changes made:
Update CLM to clm2_3_dev10.
The following changes were implemented by Erik and reviewed by me.
Use values of constants from models/csm_share/shr_const_mod.F90 for
consistency with other CCSM component models. The changes are mainly in
the physics/cam1/physconst.F90 module.
Also changed some hard-wired constants in individual parameterizations:
ocnsice/dom/srftsb.F90
cmair = cpair (was 1.00e3)
cmice = shr_const_cpice (was 2.07e3)
rhair = shr_const_rhodair (was 1.25)
rhice = shr_const_rhoice (was 9.20e2)
ocnsice/som/oceanconst.F90
tof = shr_const_tkfrz -1.9_r8 (was 271.26)
rhoocn = shr_const_rhosw (was 1.026e3) same
cpocn = shr_const_cpsw (now 3.996e3, was 3.930e3)
tsm = shr_const_tkfrz (was 273.16)
physics/cam1/cldwat.F90
rhonot = rhodair/1000. (was 1.275e-3)
t0 = tmelt (was 273.16)
replace hard-wired 273.16 with t0
physics/cam1/esinti.F90
pass tmelt as an arg
physics/cam1/gffgch.F90
replace hard-wired 273.16 with tmelt
physics/cam1/radae.F90
amd = mwdry (now 28.966, was 28.9644)
Made some interface changes to provide initialization routines for
setting constants in parameterizations.
Moved routines for water vapor pressure calcs into new wv_saturation
module.
Moved routines for shallow convection into moistconvection module.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_7_dev17
Originator: sawyer ( William Barton Sawyer)
Date: Tue Dec 18 07:25:41 MST 2001
Model: CAM
Version: CAM1.7.dev17
One-line Summary: Bug fix of mod_comm library for AIX and Linux
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, Linux
Ran test-model.pl script: no
Tested on fv dynamics: yes
Tested on eul dynamics: no
Tested on sld dynamics: no
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: no
Tested that different domain decompositions match bit-for-bit: no
Tested in adiabatic mode: no
Code reviewed by: myself, Shen
Updated the DatasetLog (for new input data-files): not-applicable
Updated the documentation: not-applicable
Changes answers: no (bit-for-bit)
Name of tests cases on NCAR Mass-Store: N/A
Changes to CLM land-model: None
Changes made:
models/utils/pilgrim: memstuff.c, mp_assign_to_cpu.c mod_comm.F90
Minor bug fixes to get mod_comm to compile and run on AIX and Linux
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_7_dev16
Originator: sawyer ( William Barton Sawyer)
Date: Mon Dec 17 03:19:29 MST 2001
Model: CAM
Version: CAM1.7.dev16
One-line Summary: FV dycore improvements: mod_comm comm. library; more support for 2D decomposition
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: SGI
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Putman, Lin
Updated the DatasetLog (for new input data-files): no
Updated the documentation: not-applicable
Changes answers: no (bit-for-bit)
Name of tests cases on NCAR Mass-Store: N/A
Changes to CLM land-model: None
Changes made:
model/utils/pilgrim:
mod_comm.F90: optimized comm. library (Lin, Putman, Shen)
memstuff.c: related utilities for MLP parallelism
mp_assign_to_cpu.c: CPU pinning utility (MLP and MPI modes)
model/atm/cam/bld
CAM_lab.pm: improvements for -l dao
configure_fv.pl, script.m4: Additional support for 2D decomp.
model/atm/cam/src/dynamics/fv
benergy.F90, cd_core.F90, dynpkg.F90, geopk.F90,
hswf.F90. prognostics.F90, restart_dynamics.F90,
spmd_dyn.F90, stepon.F90, te_map.F90:
Phase 1 of transition to mod_comm library,
mod_mpi replaced by mod_comm in cd_core.F90.
Variable PT ghosted throughout various routines.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_7_dev15
Originator: rosinski ( Jim Rosinski)
Date: Thu Dec 6 11:47:59 MST 2001
Model: CAM
Version: CAM1.7.dev15
One-line Summary: Add per-tape averaging flag. Delete inlining for SGI
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, PC
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Updated the DatasetLog (for new input data-files): no or Yes or not-applicable
Updated the documentation: no
Changes answers: no (bit-for-bit)
Name of tests cases on NCAR Mass-Store: (For changes that change answers)
Changes to CLM land-model: None
Changes made:
o Added namelist option avgflag_pertape. Usage ex: if set avgflag_pertape(2) = 'I',
all fields on history file 2 will by default be instantaneous. Can still
be overridden (for example) with fincl2 = 'T:A'
o Deleted inlining for SGI (a mod to bld/Makefile). Inlining slowed compilation
down by 60% and resulted in zero or negative speedup. Could also delete for
other machines, but Makefile mods, compilation time and execution time effects
of including inlining were not nearly as obnoxious as on SGI so left alone for now.
o Added addfld call for NSTEP. Writing this field to history file is often useful
for diagnosing mods to history.F90. By default this field is not written.
o Moved dycore print from parse_namelist.F90 to cam.F90
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_7_dev14
Originator: mvertens ( Mariana Vertenstein)
Date: Tue Dec 4 09:08:53 MST 2001
Model: CAM
Version: CAM1.7.dev14
One-line Summary: put in esmf calendar manager into clm2
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Compaq
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, erik
Updated the DatasetLog (for new input data-files): not-applicable
Updated the documentation: Yes (for clm2 only)
Changes answers: On utefe(sgi) and prospect (compaq) the result were
bfb when running test-model.pl with the -c option. On the ibm,
the changes were roundoff compared with cam1_7_dev13 when running with
the esmf libO version and bfb when running with the esmf libg version
(differences only appeared after 98 times steps and were due to
roundoff differences in the computation of the calendar day)
Changes to CLM land-model: Yes
1) Put in the esmf time manager into clm2
Brought cam code up to date with clm2_3_dev6.
Only clm2 code was changed.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_7_dev13
Originator: eaton ( Brian Eaton)
Date: Sun Dec 2 15:54:13 MST 2001
Model: CAM
Version: CAM1.7.dev13
One-line Summary: make history files conform to CF metadata conventions
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Compaq, sun, linux-pgf
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Updated the DatasetLog (for new input data-files): no
Updated the documentation: no
Changes answers: no (bit-for-bit)
Name of tests cases on NCAR Mass-Store:
Changes to CLM land-model: None
Changes made:
Changes are all in history.F90 except for some units changes in addfld
calls in gw_drag.F90. The modifications are:
.Fix units attributes.
.Fix convention attribute.
.Fix description of hybrid vertical coordinates.
.Write all coordinate variables as type double.
.Fix description of time averaged values.
.Add time interval boundaries.
.Remove _FillValue attribute for variables on a full grid.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_7_dev12
Originator: eaton ( Brian Eaton)
Date: Thu Nov 29 19:08:09 MST 2001
Model: CAM
Version: CAM1.7.dev12
One-line Summary: update ESMF lib to MF_LLU_0_0_11p3
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Compaq, sun, linux/pgf
Ran test-model.pl script: yes
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: David Neckels
Updated the DatasetLog (for new input data-files): no
Updated the documentation: no
Changes answers: no (bit-for-bit)
Name of tests cases on NCAR Mass-Store:
Changes to CLM land-model: None
Changes made:
Update the ESMF library from version MF_LLU_0_0_11p2 to MF_LLU_0_0_11p3.
Date increment incorrectly normalized itself when on the first month of the
year. The time management increment function did not completely construct
itself. These bugs were uncovered during unit testing of the ESMF library
by David Neckels.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_7_dev11
Originator: mirin ( Arthur Andrew Mirin)
Date: Wed Nov 28 23:02:45 MST 2001
Model: CAM
Version: CAM1.7.dev11
One-line Summary: Changes for multi-2D decomposition for FV dycore
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Sawyer
Updated the DatasetLog (for new input data-files): not-applicable
Updated the documentation: not-applicable
Changes answers: Yes (minor roundoff differences in FV dycore routine te_map.F90)
(Note: test-model.pl only does checks this -- if you run it with the "-c" option)
Name of tests cases on NCAR Mass-Store: (For changes that change answers)
Changes to CLM land-model: None (one-line description)
Changes made:
Changed 13 files in dynamics/fv; most changes were fairly minor.
Changed control/spmdinit.F90.
Changes 3 files in physics/cam1, to allow more general chunking. These changes were actually made by Pat Worley.
Changed 2 files in Pilgrim (minor).
Added batch capability for test-model.pl for LLNL IBM Frost.
Minor change to CAM_lab.pm for LLNL Frost.
Minor changes to DAO scripts configure_fv.pl, script.m4.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_7_dev10
Originator: mvertens ( Mariana Vertenstein)
Date: Mon Nov 26 08:32:45 MST 2001
Model: CAM
Version: CAM1.7.dev10
One-line Summary: changed mpi distribution of subgrid patches in clm2
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Compaq
Ran test-model.pl script: yes (does all of the following tests)
On Compaq only tested eul dynamics
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Updated the documentation: not-applicable
Changes answers: no (bit-for-bit)
Ran test-model.pl with -compare optio
Changes to CLM land-model only:
Brought cam code up to date with clm2_3_dev4
1) changed the distribution of subgrid patches among processors
so that patches for a given land point must all be on the same processor
2) changed all mpi gather/scatter calls to mpi gatherv/scatterv calls
and introduced subroutines to determine the asymmetric vector lengths
3) Put in CF1.0 conventions into clm history file
4) removed rtm_doavg from namelist - rtm averaging is now done
if rtm_nsteps is set to greater than 1 (this does not appply
to cam model unless cam is run with RTM - by default this is
not the case)
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_7_dev9
Originator: erik ( Erik Kluzek)
Date: Wed Nov 21 16:02:10 MST 2001
Model: CAM
Version: CAM1.7.dev9
One-line Summary: Get coupled model to work, update to scripts_a011121
and bld_b011121
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: Yes! (IBM only)
Machines tested: IBM
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Changes answers: no (bit-for-bit)
Changes to CLM land-model: Terminate string in clm_csmMod.F90 so would compile.
Changes made:
Make changes needed to get CAM to work with coupled with CCSM. Update models/bld
to bld_b011121, and scripts/test.a1 to scripts_a011121. Run csm.csh on babyblue
to make sure would work. Also ran on utefe, but lnd.setup.csh failed due to
a "getdata" problem, the compile of CAM was successful however. Update "csm.csh"
coupled version to ccsm2_0_beta26.
Change ccsm_msg.F90 to use time_manager to get nstep (from Lawrence Buja).
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_7_dev8
Originator: eaton ( Brian Eaton)
Date: Wed Nov 21 08:47:35 MST 2001
Model: CAM
Version: CAM1.7.dev8
One-line Summary: add perpetual calendar option to time manager
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Compaq, pc-linux, sun
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Updated the DatasetLog (for new input data-files): no
Updated the documentation: Yes
Changes answers: no (bit-for-bit)
Name of tests cases on NCAR Mass-Store:
Changes to CLM land-model: None
Changes made:
Add namelist variables to enable perpetual calendar:
logical perpetual_run - Set to .true. to specify that a perpetual run is
being done (default is .false.). If perpetual_ymd
is not set then read the perpetual date from the
initial file.
integer perpetual_ymd - Perpetual date specified as (year*1000 + month*100
+ day). This overrides value from initial file.
If aqua_planet=.true. then the values of perpetual_run and perpetual_ymd
are ignored. aqua_planet mode uses a perpetual date of 321.
Modify files:
control/time_manager.F90 - add perpetual_run, perpetual_ymd to public data.
add is_perpetual method to indicate that perpetual calendar is in use.
Modify internal logic so that perpetual calendar doesn't depend on
aqua_planet mode. Add an optional argument to specify an offset in the
get_curr_date method (needed by the CLM).
control/parse_namelist.F90 - add perpetual_run, perpetual_ymd.
Modify code in the following files to use the perpetual date when
interpolating boundary datasets if perpetual calendar is in use.
control/initext.F90
control/oznini.F90
control/oznint.F90
control/so4bnd.F90
dom/sst_data.F90
physics/advnce.F90
physics/chemistry.F90
physics/tphysac.F90
Misc:
control/initext.F90 - There was code in a #ifdef COUP_CSM section that was
nested inside a #if (!defined COUP_CSM) section. Remove this code which
wasn't being executed anyway.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_7_dev7
Originator: eaton ( Brian Eaton)
Date: Fri Nov 16 17:35:37 MST 2001
Model: CAM
Version: CAM1.7.dev7
One-line Summary: update ESMF library
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Compaq, pc, sun
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself and David Neckels
Updated the DatasetLog (for new input data-files): no
Updated the documentation: no
Changes answers: no (bit-for-bit)
Name of tests cases on NCAR Mass-Store:
Changes to CLM land-model: None
Changes made:
Update the ESMF library from version MF_LLU_0_0_11 to MF_LLU_0_0_11p2.
This includes bugfixes for the mfm_datedecrement and mfm_timemgrlaststep
functions.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_7_dev6
Originator: erik ( Erik Kluzek)
Date: Fri Nov 16 09:37:28 MST 2001
Model: CAM
Version: CAM1.7.dev6
One-line Summary: Update ESMF to MF_LLU_0_0_11, change test-model.pl interface, fix problem with run-model.pl
Requires change in build system: Yes
(Changed Makefile on sun to be compatable with new ESMF)
(Changed Makefile for Linux-PGF90 so sgexx.F doesn't get compiled with -fast)
Substantial timing or memory changes: no
Requires change in run script: Yes (run-model.pl scripts need to use
setup_directories( "support" )
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Compaq, Sun, Linux-PGI (NERSC-IBM, ORNL-IBM)
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Mariana Vertenstein
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Changes to ESMF: Update to MF_LLU_0_0_11
Changes made:
ESMF Update:
Get newer version of ESMF library. This fixes a problem with calculating
the ending date (which fixes a problem that the CLM had in writing out restart
files at the end of the run). It also fixes problems in building on several
platforms. The previous version did not build on the DAO SGI (tropic) for
example.
Makefile changes:
SUN: Use -dalign for DEBUG, and otherwise -fast to be compatable
with the new version of ESMF.
Linux-PGF90: Set to compile sgexx.F withotu -fast, so that SLD
simulations won't hang.
run-model.pl change:
Change so that CAM_run.pm sets up a directory to build makdep in
like test-model.pl does. This allows run-model.pl to use the same
source directory, but submit to different machines simultaneously.
The change requires the setup_directories( "support" ) method to
be called from run-model.pl type scripts.
test-model.pl changes:
Change interface from mixed short names like "-l" and "-nofail" to
consistent longer names: "-lab", "-compare", "-help" etcetera.
-l ==> -lab
-c ==> -compare
-h ==> -help
-t ==> Eliminated you now get the list of test from -help
-r ==> -resume
-dirty ==> -noclean
-e ==> -errgro
-s ==> -skip
For backwards compatability, -l, -c, -r, -e, and -s are still allowed, although
the longer names are prefered.
ornl: Change default CASE_DIR to a GPFS directory so that builds will work.
llnl: Change default SPMG_CMND depending on whether machine has prun or dmpirun.
Will now automatically detect which exists and use the one that works.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_7_dev5
Originator: eaton ( Brian Eaton)
Date: Tue Nov 13 10:19:40 MST 2001
Model: CAM
Version: CAM1.7.dev5
One-line Summary: makefile changes for ESMF library build
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Compaq, PC/linux-pgf
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself and Erik
Updated the DatasetLog (for new input data-files): no
Updated the documentation: no
Changes answers: no (bit-for-bit)
Name of tests cases on NCAR Mass-Store:
Changes to CLM land-model: None
Changes made:
Modified the CAM Makefile to prevent unnecessary rebuilding of the ESMF
library.
The ESMF Makefile has been improved (by David Neckels) to include support for
more Fortran compilers under Linux, and to allow simultaneous builds of the
library from different platforms with access to the same source on a shared
filesystem. There is also a README file for the ESMF build in
top_dir/models/utils/esmf/README.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_7_dev4
Originator: erik ( Erik Kluzek)
Date: Fri Nov 9 11:44:51 MST 2001
Model: CAM
Version: CAM1.7.dev4
One-line Summary: Fix problem with run-model.pl, add condense-path utility,
update ESMF to MF_LLU_0_0_10p1
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI, Compaq (partial tests on Linux and Sun)
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Known problems: SOM test fails on AIX, LSM/phys366 test fails on Linux
SLD test 3 (in test-model.pl) hangs on Linux-PGf90
Code reviewed by: myself, Brian Eaton
Changes answers: no (bit-for-bit)
Changes to CLM land-model: None
Changes to ESMF: Update to MF_LLU_0_0_10p1
Changes made:
Fix path problems with run-model.pl
Add a condense-path.pl utility to shorten pathnames with "/.." in them.
This is important since the longer directory names of CAM and the multiple
levels cause line length problems when building.
test-model.pl: Turn SOM test on for all platforms but AIX
Let the SOM test happen for all platforms except the IBM where it fails.
Once this is fixed we should re-enable it.
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
cam1_7_dev3
Originator: mvertens ( Mariana Vertenstein)
Date: Wed Nov 7 13:00:31 MST 2001
Model: CAM
Version: CAM1.7.dev3
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: no
Tested to work coupled with CCSM: no
Machines tested: IBM
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: no
Tested on eul dynamics: yes
Tested on sld dynamics: no
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself
Updated the DatasetLog (for new input data-files): not-applicable
Updated the documentation: not-applicable
Changes answers: no (bit-for-bit)
Simulations:
/ERIK/csm/eul17dev3_a ---- 5 year simulation with
PCMDI climatological SST's
Changes made:
1) put in restart consistency checks between cam and clm2
i.e ensure that both models restart with same time step
===============================================================
===============================================================
cam1_7_dev2
Originator: mvertens ( Mariana Vertenstein)
Date: Wed Nov 7 10:16:12 MST 2001
Model: CAM
Version: CAM1.7.dev2
One-line Summary: Put in three clm2 bug fixes that change answers
Requires change in build system: no
Substantial timing or memory changes: no
Requires change in run script: Yes
CLM2 namelist must have the following changes:
1) fpftcon = $LM_DATDIR/pftdata/pft-physiology-vegdyn-cleanup-ratio
2) remove mksrf_firr from namelist
Tested to work coupled with CCSM: no
Machines tested: IBM, SGI
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Erik Kluzek
Updated the DatasetLog (for new input data-files): not-applicable
Updated the documentation: not-applicable
Changes answers: Yes (new-physics - bug fixes)
Changes to CLM land-model: YES
Changes made (CLM2 code):
Put in changes from clm2_2_12, clm2_2_13, clm_2_14
clm2_2_12:
1) Update to clm2 changes from CCM3.12.62, from Brian Eaton
"Use public data and methods from time_manager mod for initialization."
clm2_2_13:
1) cleaned up pft physiology file and related code
removed anything related to biogeochemistry when DGVM is off (not defined)
2) used ratio of roughness length and displacement height to canopy top height
instead of actual values for roughness lenght and displacement height
3) fixed openmp bug in RtmMod.F90
4) fixed branch bug in AccumulMod.F90
5) fixed DGVM bug in Fire.F90
clm2_2_14:
The clm2 code in this tag is clm2_2_14 with Erik's cam1_7_1 mods
where ccm->cam in various files min the clm2/main directory
1) (Biogeophysics_Lake.F90) - Fixed bug in lake model where latent heat of
fusion was being used (incorrect) instead of latent heat of sublimation
(correct).
Also Fixed two "problems" noticed in the b202.20.0 simulations:
2) (CombineSnowLayers.F90) - Global water and ice contents in the
top soil layer were increasing with time. Eliminated ponding in
wetlands by assigning snow melt to runoff.
3) (SurfaceRunoff.F90) - Eliminated formation and growth of basal ice layer
beneath snow cover by assigning snow melt reaching impermeable ground
surface to surface runoff.
The above three fixes were tested by examining output from several single
point and 21-year standalone simulations. No trends in surface soil
water and ice were noted. Lake bug fix has small effects on surface
fluxes from grid cells with lakes that have snow cover.
4) Put in volr in initial dataset
5) Put in restart id tag on restart file for consistency -
if restart file is modified, the parameter rest_id in restFileMod.F90
must be updated.
6) Set maximum number of auxillary history files to 2 to be
consistent with parameter settings.
===============================================================
===============================================================
cam1_7_dev1
Originator: erik ( Erik Kluzek)
Date: Wed Nov 7 09:11:22 MST 2001
Model: CAM
Version: CAM1.7.dev1
One-line Summary: Move from CCM to CAM, mask error-growth problems by not
letting CLM see snowfall, Change build
Requires change in build system: Yes!
(Change COUP_CCM to COUP_CAM for CLM, Makefile uses Rootdir file for
ESMF build, ESMF build is done from within Makefile, directory
structure different)
Substantial timing or memory changes: no
Requires change in run script: Yes!
(namelist is now camexp instead of ccmexp, input data files directory
changed, Env variable for datasets is now $CSMDATA)
Tested to work coupled with CCSM: no (need to update CCSM build for ESMF)
Machines tested: IBM, SGI, Compaq, Sun, Linux-PGF90
(SOM test fails on IBM, LSM/PHYS366 test fails on Linux)
Ran test-model.pl script: yes (does all of the following tests)
Tested on fv dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Code reviewed by: myself, Mariana Vertenstein, Brian Eaton (som changes)
Updated the DatasetLog (for new input data-files): no (but need to)
Did create 26 level T21 datasets for testing, and got test-datasets
at T42L18 for SOM
(new datasets at ftp.cgd.ucar; cd pub/erik;
get cam1_7.scidac-atm.datasets.tar.gz)
Updated the documentation: Yes (first pass at changes from CCM to CAM)
Changes answers: no (bit-for-bit)
Changes to CLM land-model: Yes (COUP_CCM to COUP_CAM, nix references to ccm)
Changes made:
Change references from CCM to CAM.
datafiles: ccmr and ccmi files are now camr and cami
filenames: ccm3, conv_ccm, ccmoce, are now cam, conv_cam, camoce
lnd filenames: ccm_lndMod, lnd_ccmMod, are now atm_lndMod, and lnd_atmMod
CPP Tokens: COUP_CCM is now COUP_CAM
Directory structure: New directory structure compatable with the "cam1"
cvs module rather than the "ccm" cvs module.
Other references in code and comments to CCM were changed to CAM.
New build:
Build changed to create a Rootdir file with the top level directory
of "cam1". This is used by the Makefile to figure out the location of the ESMF
directory. The Makefile then builds the ESMF library directly from the
Makefile. This new build system comes from David Neckels via Brian Eaton.
It is a much easier more robust way of building the model. The env variable
DYNAMICS for Lin-Rood should now be set to "fv" in agreement with the new directory name. For the SUN take out the "-fast" option to be compatible with the
ESMF library build (put this back in when ESMF build is updated).
Run scripts:
Run scripts use "CSMDATA" instead of MODEL_DATDIR. AM_DATDIR, and LM_DATDIR
env variables removed. New directory structure for cam with subdirectories
for: inic, ggas, rad, hrtopo, ozone, scyc, and sst for initial conditions,
greenhouse gas, radiation, high-resolution topography, ozone, sulfur-cycle
and sea surface temperatures respectively. The environment variable OCEAN
was added to refer to the ocean model to use "dom" or "som". The env variable
PHYSICS now either should be "cam" or "ccm366" to refer to the type of physics
used.
Error-growth problem:
Mask the current error-growth problem to some extent by only letting
CLM see rain and not snowfall. Error-growth for FV still seems to be large.
Remove Sulfur-cycle codes:
Remove the sulfur-cycle codes (they are still accessible on the
cam_brnch_scyc branch and with the cam1_scyc module). In their place the
dummy stubs from dscyc were put into the control directory. Eventually, we
will rip these out of the main development and only leave them on the
sulfur-cycle branch.
Changes to test-model.pl:
Change vertical resolution of the sld and eul tests to 26, in agreement
with the vertical resolution used for production simulations (the difference
in vertical resolution can have an impact on how changes to physics are
perceived). Remove "-clean" option and make cleaning old directories out the
default, add the option "-dirty" to not clean files before the build.
Users Guide:
Users Guide now part of the distribution, first pass on moving from
CCM3.12 to CAM was completed. Namelist documentation is up to date, directory
structure is outlined. Further updates should update documentation at the
same time that code is changed.
SOM:
SOM now part of the distribution, SOM was updated to use the new
timing routines. "test-model.pl" has a test for SOM that is currently commented
out since it doesn't run on the IBM (does complete on other platforms however).
See: http://www.cgd.ucar.edu/cgi-bin/cms/ncar_only/view_change_file.cgi?ChangeLog
for latest version of this file on the web.
===============================================================
===============================================================
ccm3_12
Originator: erik ( Erik Kluzek)
Date: Wed Jun 13 11:35:28 MDT 2001
Model: CCM
Version: CCM3.12
One-line Summary: New H2O LW abs/ems, geo-potential, precip evap, cloud-water changes, no oro-drag over ocean
Requires change in build system: no (but, Makefile updated for in-lining, see below)
Substantial timing or memory changes: no
Requires change in run script: Yes (New abs_ems_dataset, see below)
Tested to work coupled with CSM: Yes (IBM and SGI)
Machines tested: IBM, SGI, Compaq, Sun, Linux (see below)
Ran test-model.pl script: yes (does all of the following tests)
Tested on lr dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: yes
Updated the DatasetLog (for new input data-files): Yes
Updated the documentation: no (but need to)
Changes answers: Yes (new-physics)
Name of tests cases on NCAR Mass-Store:
/JET/csm/zm31139evap06 ---- Climatological SST's T42L26, Eulerian dynamics
/JET/csm/amipevap06 ------- AMIP2 SST's T42L26, Eulerian dynamics
/JET/csm/amipevap06_3h ---- Daily output same as above.
/CCSM/csm/b202.02 --------- Coupled simulations, T42L26, Eulerian dynamics
Changes made:
IMPORTANT: Problems on SUN, and Linux, problems on batch submission to:
blue, eagle, gseaborg (see below). Need new datasets (see below).
----------------------------------------------------------------------------
Physical changes:
Precipitation evaporation: J.J. Hack.
Cloud water changes for mid-atm.: P.J. Rasch
LW water-vapor abs/ems changes: W. Collins
no orographic drag over ocean: B.B. Boville
fix to geo-potential: B.B. Boville
Land-model:
Tested with both LSM1 and CLM2 land-models (see below). CLM2
was updated to CLM2.1.20.
Interface changes:
Makefile changed so that in-lining is done.
radems/radabs became a module.
LINUX CPP token changed to Linux (so you can use uname -s)
Block data removed (except for phys366).
Check that ORO is valid from IC file.
Testing:
Code was tested both on many NCAR platforms and several off-site
platforms as well. test-model.pl was primarily used to validate
that the model started/restarted for all dynamics as well as
both the "physics" and "phys366" physics directories and "lsm1"
and "clm2" land models. It also verified that restarts give exact
answers to the history files (both average and instant values),
and that changing the decomposition gives the same answers and
that SPMD and non-SPMD gives the same answers. It also checks
that the code can run with bounds checking on (on Sun and Linux),
and with the compiler initializing variables to indefinite.
CCSM tests run ccm/test/csm.csh to ensure that the model works
with the CCSM coupler. This version of the model works with
ccsm2_0_beta11.
Machine lsm1-physics CCSM clm2-physics clm2-phys366 lsm1-phys366
babyblue(aix) X X X X X
blackforest(aix) X X
gseaborg(aix) X
eagle(aix) X
blue(aix) X
longs(PC) X X X
prospect(osf) X X X X
compass(osf) X
utefe(sgi) X X X X
tropic(sgi) X X X
sanitas(sun) X X X
Error growth:
The error-growth is faster than the previous model, especially
SLD but, it was deemed acceptable. HOWEVER,
THE ERROR-GROWTH ON THE SUN IN COMPARISON TO THE IBM IS
UNACCEPTABLE. Until this is fixed the Sun should not be
used for simulations.
Script changes:
test-model.pl was made more robust. Tests
to compare SPMD non-SPMD were added and to check both avg
and instant history files for restart tests, and new
input options added:
set the env variables:
LANDMODEL --- to define whether to use LSM1 or CLM2
PHYSICS ----- to define whether to use "phys366" or "physics".
skip option (-s) now allows an end-point and allows resolution
to be specified. So
-s eul:3-5:T42L26
Will run tests 3-5 with Eulerian dynamics at T42 and 26 levels.
Error-growth option. To do a port validation you can validate
against a given trusted machine. To do this use something like:
-e "blackforest.ucar.edu(aix)"
This assumes you already ran test-model.pl on blackforest (and
left everything in the standard locations). Then the "errg*.cprout"
files will contain a comparison to the above machine.
New datasets:
Bundle of datasets needed to run the model and use "test-model.pl".
ftp.cgd.ucar.edu; cd pub/erik; bin;
get ccm3.12.acpi-atm.datasets.tar.gz
abs_ems_factors_fastvx.052001.nc --- Required for use with
the new LW code.
clm datasets ---- In order to run with clm2.
fv.4x5L18 -------- 4x5 degree resolution dataset with 18 levels
for LR dynamics from Sharon Nebuda.
Known-Bugs:
Linux: S.-J. had trouble with 10x15 on Linux with LR and the
Portland group compiler. I had trouble with Linux at
2x2.5 with LR.
Sun: Error-growth was bad. Ran out of memory at some resolutions,
especially when doing SPMD, or when using CLM2.
Batch: Had trouble with batch submission on the IBM-SP machines:
blue, gseaborg, and eagle. Interactive, didn't have the
same trouble, so I'm unclear why this is the case.
test-model.pl: When using "-s all:" option you can't give an
end-point or it will only run for the first dynamics.
Changes since CCM3.11:
Many enhancements, bug-fixes, and progress on 2D-decompostion to LR
dynamics. Update to F90 and initial Protex pass for LR.
Replace Rayleigh friction with 1st order upwind transport.
Revised mapping algorithm near the surface for winds.
slaveio namelist item removed. Namelist variable "IDEALDYN"
changed to "IDEAL_PHYS"
Support for Lahey and Fujitsu compilers on Linux. Work on chunking.
Columns given by phys_grid methods. timing library added.
Ability for bounds-checking added. FORTFFT cpp parameter deleted
use USEFFTLIB if library available. SPMD now doesn't require that
the number of MPI tasks evenly divide into the number of latitudes.
History modularized. Progress on physics/dynamics split. Restart
dyanmics and physics split out better. Sulfur-cycle codes fixed.
SLD energy fixer added.
phys366 and clm2 directories added. Generic interface to both
land-models.
Options added to scripts: PHYSICS, LANDMODEL, tests to run.
NO_SWITCH option for LLNL Compaq. "-clean", "-nofail" and "-t"
options added. Fix interpic for multiple times if available in file.
===============================================================
===============================================================
ccm3_11
Originator: erik ( Erik Kluzek)
Date: Thu Dec 21 16:22:20 MST 2000
Model: CCM
Version: CCM3.11
One-line Summary: New LW absorption,new vertical diffusion,SLD energy fixer,
scripts,Lin/Rood fixes,split dynamics
Requires change in build system: yes, new build system, use 26 level model
Tested on lr dynamics: yes
Tested on eul dynamics: yes
Tested on sld dynamics: yes
Tested to work coupled with CSM: no
Tested that restarts are bit-for-bit: yes
Tested multiple constituents: yes
Tested that different domain decompositions match bit-for-bit: yes
Tested in adiabatic mode: no (currently broken)
Substantial timing or memory changes: no
Machines tested: IBM, SGI, Sun, PC
Requires change in run script: yes
(New namelist parameter: absems_data)
(New namelist parameter for Lin-Rood: nsplit)
(New namelist parameter: empty_htapes)
(New namelist parameter: aqua-planet)
(New build/run environment provided)
(Need to use new 26 level datasets)
Updated the DatasetLog (for new input data-files): Yes
Updated the documentation: no (will need to though)
Changes answers: Yes (improved physics)
Name of tests cases on NCAR Mass-Store: See CCM3.10.60 below
Changes made:
Physics changes from CCM3.10:
New LW absorption of water vapor as per Dr. William Collins.
New vertical diffusion on dry static energy as per Dr. Byron Boville.
Energy fixer for SLD from Dr. David Williamson and Jerry Olson.
Many bug-fixes to Lin/Rood dynamics.
Many routines migrated to modules.
New build run environment provided:
A completly new set of scripts provided to configure, build, and
run the model. Also a test script provided for testing before
committing code changes.
A script provided to help with the process of interpolating and
creating new initial condition datasets.
Code changes:
Work to split dynamics and physics better.
Initial work on 2D decomposition.
Open MP works on PC.
More platforms supported with LR.
Add version, revision_id to history files.
Fix floating point trapping on Compaq.
Various bug-fixes.
===============================================================
===============================================================
ccm3_10
Originator: erik ( Erik Kluzek)
Date: Tue Jun 20 08:35:26 MDT 2000
Model: CCM
Version: CCM3.10
One-line Summary: Update code to F90 and coding standard,
optimize radcsw, fix bugs, new LR core,
Finish reduced grid tools
Requires change in build system: no
Tested on lr dynamics: Yes
Tested on eul dynamics: Yes
Tested on sld dynamics: Yes
Tested to work coupled with CSM: Yes
Requires change in run script: Yes (requires I/C files with LANDM)
Updated the DatasetLog (for new input data-files): Yes
Updated the documentation: Yes
Changes answers: Yes (different climate)
Name of tests cases on NCAR Mass-Store: /JET/csm/ccm3911 (full grid Eulerian)
/ROSINSKI/csm/ccm3911_r1up (reduced grid Eul)
Changes made:
Double tagged as CCM3.9.28.
Update code to F90 free-form and coding standard.
New optimized version of radcsw. Fix bug with CSM with flxave on. Fix bug
with first time-step on Eulerian. Get offline tools to work with reduced
grid and L/R dynamics. Change so LANDM is calculated in offline tool rather
than inside model. Update L/R to new version, add "precision" changes. Fix
Linux Makefile for F90 files. Fix bug with SLD and hortal smoothing and hordif.
Fix problems with SLD on Sun, IBM and Compaq. Fix PHIS filter bug.
===============================================================
===============================================================
ccm3_9
Originator: erik ( Erik Kluzek)
Date: Thu Apr 27 15:04:53 MDT 2000
Model: CCM
Version: CCM3.9
One-line Summary: New cloud overlap, reduced grid option, addition of Lin-Rood dynamics.
Requires change in build system: Yes
Requires change in run script: Yes
(New options for reduced grid, or Lin-Rood)
Updated the DatasetLog (for new input data-files): not-applicable
Updated the CCM development schedule: no
Updated the documentation: Yes
Changes answers: Yes (new-climate)
Name of tests cases on NCAR Mass-Store: /JET/csm/bill30e
Changes made:
Added new cloud overlap assumptions from Bill Collins (random overlap when gap
between clouds in column, maximum overlap when clouds exist in both adjoining
layers).
Added the reduced grid option from Dave Williamson.
Added Lin-Rood dynamics as a new dynamics configuration.
Output cleanup. Code cleanup. Two default output streams monthly and daily
files.
Support IBM and Compaq. Will compile/run on Cray, but radiation changes
(which are highly scalar) make it run very slow. Cray will no longer be
supported.
Atmosphere determines partitioning of rain/snow.
SPMD mode now works. Support hybrid SPMD/OMP memory running on IBM, Compaq.
===============================================================
===============================================================
ccm3_8
Originator: rosinski ( ROSINSKI JIM 1397 ML CGD)
Date: Fri Aug 20 11:24:07 MDT 1999
Model: CCM
Version: CCM3.8
One-line Summary: Merged netcdfhtape branch to trunk
Requires change in build system: Yes. makdep now required. New Makefile
Requires change in run script: Yes. Namelist changes.
Run through comtest.spectral.csh and comtest.sld.csh: Yes
Run through csm.test.nqs: no
Updated the DatasetLog: no
Changes answers: no (bit-for-bit)
Changes made:
o Model now writes netcdf history tapes.
o Requires f90 compiler since modules and allocatable arrays are
used. Code now should be fully f90-compliant. No more Cray
pointers or user-defined Fortran interfaces to malloc.
o Cloud water is a non-advected tracer as opposed to advecting it
and throwing away the results of the advection (mvertens).
o Model at least runs on IBM. Have not done error growth analysis.
o Moved time filter and mass fixer to scan2. This became possible
with the time-split physics.
o Precision now specified in code rather than relying on vendor flag
(e.g. -r8). This change has not yet been put in the library code
(e.g. ecmfft), so -r8 flag (or equivalent) is still needed for
these files.
o Build procedure now REQUIRES use of the makdep routine provided
in tools/makdep. The reason is that no standard UNIX tool I
know of can provide the full dependency analysis needed.
o Eliminated out-of-core option. History, radiation, and main
model buffers are now modules. Integer pointers to buffer
locations are gone, as are /comgrd/ and points.F
o Namelist changes:
old namelist new namelist
------------ ------------
&CCMEXP &CCMEXP
nlfilt ! write extra time sample------>
! to 1st volume
stfnum ! starting file number--------->
incorrad ! abs/ems in core-------------->
incorbuf ! model buffer in core--------->
incorhst ! history buffer in core------->
ndens = [1,2, or 4] ! packing density---> ndens = [1 or 2] ! double or float
primary = 'T', 'Q',...------------------> fincl1 = 'T', 'Q',...
auxf = '1', 'T', 'Q',...,------------> fincl2 = 'T', 'Q',...
'2', 'V', 'U',...-------------> fincl3 = 'V', 'U',...
exclude = 'PS','PHIS',...---------------> fexcl1 = 'PS','PHIS',...
--------------------------> fexcl2 = 'CMFMCA','SOLIN',...
--------------------------> xzy = [T or F] ! non-coords order
/ /
===============================================================
===============================================================
ccm3_7
Originator: erik ( Erik Kluzek)
Date: Fri May 28 13:38:40 MDT 1999
Model: CCM
Version: CCM3.7
One-line Summary: Prognostic cloud-water and sulfur cycle
Requires change in build system: Yes
(new directory scyc)
Requires change in run script: Yes
(need to use readtrace=.F. and cldw_adv=.T.)
Run through comtest.spectral.csh and comtest.sld.csh: Yes
Run through csm.test.nqs: Yes
Updated the DatasetLog (for new input data-files): Yes
Changes answers: yes
Name of tests cases on NCAR Mass-Store: /JET/csm/prgcld01
Changes made:
Prognostic cloud-water and sulfur cycle put on the CCM development trunk.
Changes are huge, in essense this puts ccm3_6_20_brnchT_scycRad1_13 on the
trunk. Standard mode to run is now with 30 levels and with the NOAA Ozone
dataset. So use: SEP1.T42L30_diffPHIS.0599.nc and noaao3.1990.21999.nc.
Using these datasets readtrace must be set to FALSE and normal operation
is with cldw_adv TRUE. To build you must put the "src/scyc" directory in
your path (optionally you can checkout the dscyc module which has dummy
stub routines for the sulfur codes). The default operation is without any
of the sulfur cycle stuff turned on.
Caveat's: MODEL DOES NOT WORK WITH SPMD MODE!
CSM was tested to ensure it works, but there were
changes that haven't had a long simulation done to verify they
are working correctly.
SLD mode was tested to ensure it works, but long simulations
were not performed.
Model tested on SGI, Cray, and Sun. But, the most extensive testing
was on the SGI (ute).
SOM not tested. Only dom mode was tested.
A change was put into the LSM for arbitrary initialization. The
change is bit-for-bit when not using arbitrary initialization.
But, new spin-up datasets have not been created nor long-simulations
with the change (the change makes points southward of 60S land ice and
initializes them with a snow depth of 1m).
Current version does not work with the 44 mid-atmosphere model.
(a fix is on the way soon).
Sulfur code has it's own version of the solar zenith angle. This
routine is incompatible with the version in csm_share. This means
the model should not be used for Paleo simulations, with sulfur cycle
turned on.
(a fix is on the way soon).
===============================================================