GEOS-Chem Newsletter February 2009

From Geos-chem
Jump to: navigation, search

Dear GEOS-Chem Users,

Here is the February 2009 edition of the GEOS-Chem newsletter. As always, thanks for your continued support of GEOS-Chem!

Bob Yantosca, Philippe Le Sager, and Claire Carouge
geos-chem-support@as.harvard.edu

Requests for help

Just a reminder: all GEOS-Chem questions should be sent to the email address geos-chem-support@as.harvard.edu. This will ensure that Bob, Philippe, and Claire will all receive your email question at the same time. This will allow us to serve you better. If one of us is unavailable then the others will be able to respond to you.

GEOS-Chem update

Emissions release v8-01-04

Our emissions release, GEOS-Chem v8-01-04 is in the final testing phase. We are running a couple of 1-year benchmarks in order to assess how the most up-to-date emissions inventories impact the full-chemistry simulation.

Because the menu of emissions options is more complex than in prior versions, we are also in the process of documenting everything up in a clear and consistent manner.

Bug with Intel Fortran Compiler ("IFORT") 11

A couple of users have reported problems when migrating GEOS-Chem from the Intel "IFORT" compiler version 10 to version 11. Please visit this page to view the complete discussion thread.

In the interim, our recommendation is for GEOS-Chem users to stay with IFORT 10, as we know that this compiler produces code that optimizes well on multi-core chipsets. We have not yet had a chance to examine this problem more thoroughly.

ESMF update

Max Suarez from GSFC visited us in mid February. While Max was present, we held a telecon with some other members of the GSFC team to discuss the best way in which GEOS-Chem could be incorporated into the GEOS-5 GCM at NASA. Based on their recommendations, we are moving forward to provide them with a column-ized version of the chemistry solver of GEOS-Chem.

Bob and Philippe also have recently created an ESMF-ized version of the GEOS-Chem emissions module. This module will also be an important part of the future GEOS-Chem model.

GEOS-5 met field reprocessing

Status

As described in the January 2009 newsletter, we are reprocessing the entire archive of GEOS-5 met field data. As of February 17, 2009 the following fields have been completed:

  • 2008: entire year
  • 2004: July - December

for each of the following geographic domains:

  • China nested 0.5 x 0.667
  • Global 2 x 2.5
  • Global 4 x 5

The files have been TAR'd and Gzip'd for your convienience (as this dramatically reduces the size of the data). In each directory there are files 01.tar.gz, 02.tar.gz .. 12.tar.gz, each of which contains a single month of met field data. GEOS-Chem users outside Harvard may obtain these data files via anonymous FTP:

 ftp ftp.as.harvard.edu
 cd pub/geos-chem/data/GEOS_0.5x0.666_CH.d/GEOS_5/2008
 mget *.tar.gz
 cd pub/geos-chem/data/GEOS_2x2.5.d/GEOS_5/2008
 mget *.tar.gz
 cd pub/geos-chem/data/GEOS_4x5.d/GEOS_5/2008
 mget *.tar.gz
 cd pub/geos-chem/data/GEOS_0.5x0.666_CH.d/GEOS_5/2004
 mget *.tar.gz
 cd pub/geos-chem/data/GEOS_2x2.5.d/GEOS_5/2004
 mget *.tar.gz
 cd pub/geos-chem/data/GEOS_4x5.d/GEOS_5/2004
 mget *.tar.gz

However, to simplify the FTP process you might want to use a 3rd party FTP software program with a GUI interface, such as Mozilla FireFTP.

At present, the throughput rate is approximately 28 days of model data per day. Therefore it will still be a couple of months before the entire GEOS-5 dataset has been reprocessed.

Required code modifications

To use the new, reprocessed GEOS-5 met data in GEOS-Chem, you must first install GEOS-Chem v8-01-02 (or higher) and make sure the following switch in define.h is uncommented.

   !----- FOR GEOS-5 MET FIELDS ONLY -----
   ! NOTE: If you are using GEOS-5 met fields that were reprocessed to 
   ! correctly regrid the in-cloud optical depth and cloud fraction fields, 
   ! then be sure to uncomment the following line of code.  This will cause 
   ! FAST-J to interpret the optical depth correctly.  Leaving this option
   ! commented will cause a "quick fix" (i.e. multiplying the optical depth
   ! by the cloud fracton) to be applied, which should be a good enough fix
   ! in the meantime. (bmy, hyl, 10/24/08)
   #define IN_CLOUD_OD  'IN_CLOUD_OD'

Also, if you are using v8-01-02 or v8-01-03, you must also insert a minor bug fix to manually set the variable N_A6_FIELDS to 21. Otherwise the fields will not be read in correctly. This has to be done in routine OPEN_A6_FIELDS (located in source code file a6_read_mod.f, directly after the variable N_A6_FIELDS is read in:

         ! The last 2 digits of the ident string
         ! is the # of fields contained in the file
         READ( IDENT(7:8), '(i2.2)' ) N_A6_FIELDS

#if   defined( GEOS_5 ) && defined( IN_CLOUD_OD ) 
         !%%% KLUDGE: set N_A6_FIELDS=21 when using the reprocessed
         !%%% GEOS-5 met.   This accounts for CMFMC (which doesn't seem
         !%%% to get counted) as well as for MOISTQ, which is an extra
         !%%% derived field.  (jmao, bmy, 2/12/09)
         N_A6_FIELDS = 21
#endif

However, it is probably just easier to wait for the release of v8-01-04, which is slated for release shortly. The above bug for N_A6_FIELDS is fixed in v8-01-04.

Please also see our new wiki page that details the GEOS-5 met field regridding in more depth.

GEOS-Chem Meeting: April 7-10, 2009

The 4th Annual GEOS-Chem Scientific & Users' Meeting is rapidly approaching. Daniel has provided a first draft of the meeting agenda (click here for PDF format). Please direct your questions about the agenda to Daniel Jacob (djacob@fas.harvard.edu).

If you have not registered for the meeting but would like to attend, then please contact Brenda Mathieu (blm@as.harvard.edu) with your name and affiliation as you would like it to appear on your badge. There is no registration fee.

--Bob Y. 12:55, 17 February 2009 (EST)