Craig:

A few years ago, we added the capability to write vis5d files
from tgcmproc (sendv5d), and we used vis5d up at the ML vis
lab to make 3-d volume animations of tgcm results. They have
fast machines up there, as well as a large screen, 3-d glasses,
etc., etc.

Last month Tim Scheitlin (scd vis lab) wrote me the following:

!-----------------------------------------------------------------------
From scheitln@cyclone.scd.ucar.EDU Mon Jul 17 10:48 MDT 2000
From: scheitln@cyclone.scd.ucar.EDU (Tim Scheitlin)
Subject: Geomagnetic Storm Visualization
To: roble@ucar.edu, foster@ucar.edu
Date: Mon, 17 Jul 2000 10:48:49 -0600 (MDT)
Cc: don@ucar.edu, scheitln@ucar.edu
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit


Ray and Ben,

Back in '97 we worked together on a TIEGCM animation featuring a
visualization of Electron Density and Neutral Wind variables of an
October '95 Geomagnetic Storm:

 http://www.scd.ucar.edu/vg/scheitln/visualizations/magneticstorm/LNE2.mpg

As I am sure you know, this year and next are expected to be peak
periods for solar storm activity, and consequently, we are likely to
receive more visitor and media interest on this topic.

If it would be of interest and value to you, I would like to generate a
new Geomagnetic Storm Visualization with current data or data
representing a historical solar maximum.  If so, do you have any data
that might be of scientific and public interest when visualized as we
did for the '95 Geomagnetic Storm?

Tim Scheitlin
scheitln@ucar.edu
x1829
!-----------------------------------------------------------------------


Take a look at the animation at the URL Tim mentions above. It is
Ne (or TEC) diffs from a storm that Gang Lu was simulating. This is
a classic satellite view projection, but we did other projections 
as well.

I talked w/ Ray about this and he got rather excited about trying
v5d out on some of his time-gcm full year runs ("equinox transition"
runs). If you are interested in this, please try making some sample
vis5d files from tgcmproc, using the below histories from Ray's runs.
Also take a look at the cgm file 
/ouray/e/roble/eqtran4/eqtran4n.vincent.d1-360 
these are results from the year 4 eqtran run, which has lots of 
"realistic" forcing, and variability. This could be shown alone, 
or as differences from the base case without forcing.

I don't think v5d files have been made from dataproc, so probably
test it on the Cray first. I even installed a version of vis5d
on vishnu for validation of v5d files, see /vishnu/e/foster/vis5d
(see /vishnu/e/foster/vis5d/vis5d-4.3/REAMDE for overview).
This worked for testing files, but is too slow on the Sparc stations
for animation.
 
Also, I talked w/ Mark Rast recently, and he (and probably Peter Fox)
are considering pushing for better vis display capabilities here
at hao from software running at the ML. This way we could run
vis5d on their fast machines up there, but display on our local
workstations...

As you know, I'll be gone on vacation 8/31-9/14, or so, i.e.,
leaving Thursday. Maybe we could make a couple v5d files and
send them to Tim by then...maybe we can talk about this on Monday...

--Ben

BTW, the new tgcmproc executable seems to be working fine...maybe
when I get back we can try compiling tgcmproc on a local Sun...




Year 4 equinox transition timegcm tgcm22mt runs:

With f10.7, auroral forcing, and NMC forcing:
/ROBLE/KIB22MT/PEQN401 -> PEQN490
/ouray/e/roble/eqtran4/eqtran4n.vincent.d1-360

No f10.7, aurora, or NMC forcing (base case):
(snoe forcing after day 180)
/ROBLE/KIB22MT/PEQT401-490 

Also other runs documented in ~roble/timegcm/tgcm22/eqtran/eqtran.log

See msls /ROBLE/KIB22MT for complete listing of primary
and secondary history files.

Tim Scheitlin of scd vis lab: scheitln@ucar.edu, x1829