Hi Liying, Astrid, and Wenbin:

We have now merged Astrid and Liying's mods, and made several 
coding changes toward a pre-release of tiegcm1.8. I have saved 
this code as tiegcm1-8dev1. Could we please test this model 
until I get back on Thursday next week. 

This is not yet well-test code, so I'm sure we will find things 
to fix right away. Maybe we can assemble mods to this development 
code on Thursday, and make a pre-release of tiegcm1.8 the following 
week of Apr 18.

I have put tar files of tiegcm1-8dev1 in the following locations:

HAO: /toshi/ftp/pub/tgcm/dev/tiegcm1-8dev1.tar.Z
     (http://download.hao.ucar.edu/pub/tgcm/dev)

SCD: /hao/tgcm/dev/tiegcm1-8dev1.tar.Z.

I have also cvs tagged this code, so if you're feeling
adventurous, you should be able to check it out at hao
by first setting "setenv CVSROOT $TGCMROOT/cvs", then:

cvs checkout -r tiegcm1-8dev1 tiegcm 

This should create a tiegcm/ subdir, but I think it does not include 
subdirs scripts/ or doc/. Copying one of the above tar files is 
the best bet right now.

A basic procedure to make a short default run is as follows, from 
a /ptmp directory on bluesky in this case, altho the code should 
build on bs, lightning, and hao Linux serial boxes (fields.F subs 
were split out, enabling compilation on callisto, crumb, etc):

bs> cp /hao/tgcm/dev/tiegcm1-8dev1.tar.Z . # copy tar file
bs> uncompress tiegcm1-8dev1.tar.Z         # uncompress (gunzip at hao linux)
bs> tar xf tiegcm1-8dev1.tar               # extract (will write tiegcm/ subdir)
bs> cp $TGCMROOT/scripts/tiegcm-ibm.job .  # copy job file
bs> Edit .job file as follows:

 set model       = tiegcm
 set wrkdir      = .
 set tgcmroot    = .

bs> tiegcm-ibm.job >&! tiegcm-ibm.out &    # make short interactive run

After this, you can mv tiegcm_default.inp to tiegcm.inp, and fix it
for your own runs, or copy your own input file from elsewhere. Don't
forget to "set input = ..." in the .job file.

I did not have time to test setting periodic points in dt.F, as Wenbin
has suggested, but I promise to do this when I get back. If you want
them, open up the call to mp_periodic_f3d near the end of sub dt.

Also, Ray said it was ok to postpone Argon and Helium for tiegcm1.9
or tiegcm2.0.

The change log is tiegcm/src/README, but also see tiegcm/doc for more
extensive documentation (we will of course continue to update files in doc/).

Good luck!

--Ben