Difference between revisions of "Dicarbonyls simulation"

From Geos-chem
Jump to: navigation, search
Line 24: Line 24:
  
 
--[[User:Ccarouge|Ccarouge]] 15:51, 17 February 2010 (EST)
 
--[[User:Ccarouge|Ccarouge]] 15:51, 17 February 2010 (EST)
 +
 +
== Known issues ==
 +
 +
=== Input file issues ===
 +
 +
'''''[mailto:dbm@umn.edu Dylan Millet] wrote:'''''
 +
 +
:I found some problems with the standard v8-2-1 code and specific run directories for running dicarbonyl simulations, which lead to mysterious and cryptic run death. globchem.dat has an extra line under the HNO3 emission and before the NO2 drydep (line 2170 needs to be deleted). I also ran into other problems which I traced to read-in errors with jv_spec.dat (...).
 +
 +
:--[[User:Dbm|Dbm]] 00:00, 12 October 2009 (EDT)
 +
 +
As explained above, Dylan Millet had some problems with the input files for dicarbonyl simulation. There were some problems in the format of globchem.dat and jv_spec.dat for dicarbonyls. The files had been updated. If you want to use the dicarbonyl simulation, please redownload the run directories from the ftp:
 +
 +
  ftp ftp.as.harvard.edu
 +
  cp pub/geos-chem/public_releases
 +
  get GEOS-Chem.v8-02-01-public-dicarb.rundir.4x5.tar.gz
 +
 +
or
 +
 +
  get GEOS-Chem.v8-02-01-public-dicarb.rundir.2x25.tar.gz
 +
 +
--[[User:Ccarouge|Ccarouge]] 12:54, 16 October 2009 (EDT)
 +
 +
=== Out-of-bounds error in SOAG_LIGGIO_DIFF ===
 +
 +
There was an out-of-bounds error in SOAG_LIGGIO_DIFF.  Here is the solution:
 +
 +
'''''[mailto:ccarouge@seas.harvard.edu Claire Carouge] wrote:'''''
 +
 +
:In fact, in SOAG_LIGGIO_DIFF (carbon_mod.f), there is a loop over LLTROP. It's the maximum number of levels for the tropopause. But in the loop we use JLOP indexes which are 0 outside the actual tropopause.
 +
 +
:So the fix is to add the line:
 +
            ! JLOP equal 0 if we're not in the tropopause. (ccc,10/16/09)
 +
            IF ( JLOOP == 0 ) CYCLE
 +
 +
:after line 1224:
 +
            ! Get 1-D index
 +
            JLOOP  = JLOP( I, J, L )
 +
 +
:There is the same kind of loop in SOAM_LIGGIO_DIFF just after. You should add the same fix.
 +
 +
--[[User:Ccarouge|Ccarouge]] 12:54, 16 October 2009 (EDT)

Revision as of 22:00, 19 February 2010

Overview

As an option in GEOS-Chem, you can run with more tracers to study the dicarbonyls chemistry.

NOTE: The files needed to use the KPP solver were not generated for the dicarbonyls chemistry. You can not run the dicarbonyls chemistry with the KPP solver. You have to use SMVGEAR.

Input files

For this simulation, you need different input files for the chemistry:

You will also need a restart file with additional tracers. You can create your own (see GAMAP manual) or download one from the ftp:

 ftp ftp.as.harvard.edu
 cd pub/geos-chem/public_releases/
 get GEOS-Chem.v8-02-01-public-dicarb.rundir.4x5.tar.gz

or

 get GEOS-Chem.v8-02-01-public-dicarb.rundir.2x25.tar.gz

You'll also need a specific restart file for the SOA tracers. You can find one in the same archive file on the ftp.

References

  1. Fu, T.-M., D.J. Jacob, and C.L. Heald, Aqueous-phase reactive uptake of dicarbonyls as a source of organic aerosol over eastern North America, Atmos. Environ., 43, 1,814-1,822, 2009.PDF

--Ccarouge 15:51, 17 February 2010 (EST)

Known issues

Input file issues

Dylan Millet wrote:

I found some problems with the standard v8-2-1 code and specific run directories for running dicarbonyl simulations, which lead to mysterious and cryptic run death. globchem.dat has an extra line under the HNO3 emission and before the NO2 drydep (line 2170 needs to be deleted). I also ran into other problems which I traced to read-in errors with jv_spec.dat (...).
--Dbm 00:00, 12 October 2009 (EDT)

As explained above, Dylan Millet had some problems with the input files for dicarbonyl simulation. There were some problems in the format of globchem.dat and jv_spec.dat for dicarbonyls. The files had been updated. If you want to use the dicarbonyl simulation, please redownload the run directories from the ftp:

  ftp ftp.as.harvard.edu
  cp pub/geos-chem/public_releases
  get GEOS-Chem.v8-02-01-public-dicarb.rundir.4x5.tar.gz

or

  get GEOS-Chem.v8-02-01-public-dicarb.rundir.2x25.tar.gz

--Ccarouge 12:54, 16 October 2009 (EDT)

Out-of-bounds error in SOAG_LIGGIO_DIFF

There was an out-of-bounds error in SOAG_LIGGIO_DIFF. Here is the solution:

Claire Carouge wrote:

In fact, in SOAG_LIGGIO_DIFF (carbon_mod.f), there is a loop over LLTROP. It's the maximum number of levels for the tropopause. But in the loop we use JLOP indexes which are 0 outside the actual tropopause.
So the fix is to add the line:
           ! JLOP equal 0 if we're not in the tropopause. (ccc,10/16/09)
           IF ( JLOOP == 0 ) CYCLE
after line 1224:
           ! Get 1-D index
           JLOOP   = JLOP( I, J, L )
There is the same kind of loop in SOAM_LIGGIO_DIFF just after. You should add the same fix.

--Ccarouge 12:54, 16 October 2009 (EDT)