> 
> Hi Ben,
>    Sam and I were trying to figure out how to run the NO and CO2 cooling
> rates for the new CIRRIS runs.  You may recall you did this for us back in
> November but the CO2 densities were used in the NO cooling algorithm.
> Could you run them again, or tell us what we need to do to run them?
> 

John:

This is a bit confusing, so I'm going to review the situation.
Maybe we should also talk on the phone.

As you know, there are two main processor codes:

1. tgcmproc. This code does purely post-model processing from the model 
   histories, and so produces only fields from the histories, or derived
   fields calculated from history fields. One class of post-model derived
   fields are emissions, including E6300, E5577, EO200, EOH83, ECO215u,
   and ENO53u. The error reported earlier was for the ENO53u (5.3 micron
   NO emission). CO2 densities were being passed in instead of NO densities
   in the ENO53u calculation.

2. modelproc. This code is a "concurrent-model" processor. It executes
   the model for a single time step to obtain fields not saved on the
   histories, such as cooling and heating rates, including CO2 and NO
   cooling.

Tgcmproc does not calculate heating/cooling rates, and modelproc does
not calculate emission fields.  If I recall correctly, you guys have 
and can run tgcmproc, but not modelproc. Is this correct? This is why 
I sent you the notice about the error, so you could redo any tgcmproc 
runs that calculated ENO53u.

Whenever we calculate heating/cooling rates, we are using modelproc,
not tgcmproc, so any NO and CO2 cooling rates I have sent you were
*not* affected by the tgcmproc error.

However, I have in my notes that Ray and I did change the way in
which modelproc obtains CO2 cooling from the model. It is now saved
from RADCOOL rather than from DT. I have in my notes that I made
this change, and sent you "new" CO2 and NO cooling on 12/5, with 
separate files for degK/day and ergs/sec/gm. Did you receive those?? 
I did not compare them with the data sent to you on or about 11/9, 
but they may be different. Let me know if I need to redo these runs
of modelproc.

I'm sorry if this is too confusing. Maybe I should also send you
a script to run modelproc. It may be easier to understand if you
have both codes to work with.

--Ben