Difference between revisions of "MERRA implementation details"

Revision as of 19:20, 12 September 2013

On this page we list the necessary modifications to GEOS-Chem for MERRA compatibility.

Overview

The GMAO MERRA data product is a 30-year reanalysis done with the GEOS-5.2.0 assimilation system. Therefore, within GEOS-Chem we can (for the most part) treat MERRA in the same way as we do for the GEOS-5 meteorology.

However, in order to reduce the volume of data produced (and for other technical concerns), GMAO chose to reduce the vertical and/or horizontal resolution of some important MERRA data fields. Also, some MERRA data fields have different units than in the GEOS-5 product. Because of this, some additional special handling may be required within GEOS-Chem.

Key similarities to GEOS-5

The following aspects of MERRA are identical to the GEOS-5 operational data product.

1. The MERRA high-resolution grid (0.5° x 0.666°) is identical to that of GEOS-5.
2. The MERRA vertical grid (72 hybrid levels) is identical to that of GEOS-5.
3. By and large, the same data fields that were in the GEOS-5 product are also contained in MERRA.

Key differences from GEOS-5

The following aspects of MERRA differ from the GEOS-5 operational data product:

1. The MERRA temporal resolution differs from GEOS-5 as follows:
   • Surface data: hourly averages (MERRA) vs. 3-hour averages (GEOS-5)
   • 3-D data: 3-hour averages (MERRA) vs. 6-hour averages (GEOS-5)
2. Differences in individual MERRA met fields:
   • Most of the relevant MERRA 3-D data for GEOS-Chem are only available at 1° x 1.25° horizontal resolution.
   • OMEGA, PV, RH are 6-hr instantaneous data in MERRA but are 6-hr averaged data in GEOS-5
   • SNOMAS (snow mass) in MERRA was missing coverage over Antarctica and Greenland (this has been corrected for).
   • Many of the precipitation fields (e.g. DQIDTMST, DQVDTMST, DQLDTMST, etc.) are saved on a 1.25° x 1.25° x 42 pressure levels instead of 72 hybrid levels. P -> hybrid interpolation is necessary.
   • The MERRA fields FRLAKE, FRLAND, FRLANDIC, FROCEAN, PHIS are time-invariant.

Required coding changes in GEOS-Chem

As noted in the section above, the similarities between GEOS-5 and MERRA (especially of the vertical grids) means that in most cases we can just treat MERRA in the same way as for GEOS-5. In many areas of the code, the coding changes are straightforward and simply involve an extension of C-preprocessor statements such as:

#if defined( GEOS_5 )

to

#if defined( GEOS_5 ) || defined( MERRA ) 

In other areas of GEOS-Chem the changes are more involved. Below we provide a description of the areas of GEOS-Chem that were modified for MERRA:

Headers directory

The following modifications for MERRA were made in the Headers/ directory:

GeosUtil directory

The following modifications for MERRA were made in the GeosUtil/ directory:

GeosCore directory

The following modifications for MERRA were made in the GeosCore/ directory:

GeosTomas directory

The following modifications for MERRA were made in the GeosTomas/ directory:

--Bob Y. 11:07, 23 September 2010 (EDT)

Resolved issues

Fractional sea ice

In order to preserve some information about sea ice leads during the regridding process, we have binned the fractional sea ice coverage into deciles (0-10%, 10-20% ... 90-100% coverage) and saved those as separate fields in the MERRA data archive for GEOS-Chem. Please see this wiki post for more information.

SNOMAS

The SNOMAS (snow depth) field in MERRA differs from that in GEOS-5. It appears that in GEOS-5 a default snow mass value was applied to grid boxes that had > 90% land ice coverage. This was not done in MERRA. Bob Yantosca has implemented an algorithm to adjust the MERRA SNOMAS field to be backwards-compatible with GEOS-5.

Dust emissions

The dust emissions in GEOS-Chem rely on "tunable parameters" to scale the dust emissions to a specified total. This is because the dust emissions are very highly dependent on the surface winds, and a small difference in winds (e.g. from GEOS-5 to MERRA) can cause a huge difference in dust emission flux.

Duncan Fairlie supplied us with updated tuning parameters for dust. In routine DST_MBL (in dust_dead_mod.f), we have added the following code:

     !----------------------------------------------------------------
     ! Based on results from MERRA 4x5 for years 2004-2005:
     !
     !   (GEOS-5 - MERRA)/GEOS-5 * 100  is 26.9% in each size bin.
     !
     ! We need to scale to the parameter FLX_MSS_FDG_FCT to make the 
     ! dust emissions consistent.  Consequently, to bring MERRA 4x5 
     ! dust emissions up to GEOS-5 levels, we need to DIVIDE the 
     ! FLX_MSS_FDG_FCT used for GEOS-5 by (1. - 0.269) = 0.731.
     !
     !    -- Duncan Fairlie (t.d.fairlie@nasa.gov), 29 Oct 2010
     !----------------------------------------------------------------
     REAL*8,  PARAMETER     :: FLX_MSS_FDG_FCT = 7.0d-4 / 0.731d0

--Bob Y. 16:23, 13 April 2011 (EDT)

Lightning

Lee Murray has provided us with updated OTD/LIS redistribution factors for the MERRA 4x5 met fields. These were implemented into GEOS-Chem v9-01-01, along with some important updates to the lightning NOx parameterization.

--Bob Y. 16:23, 13 April 2011 (EDT)

Wet deposition and scavenging

Helen Amos and Bess Corbitt have implemented an updated algorithm for wet scavenging that takes advantage of the new precipitation fields from the MERRA data archive. This was included in GEOS-Chem v9-01-01.

--Bob Y. 16:23, 13 April 2011 (EDT)

Remove negative values from MERRA QI, QL fields

GeosCore/merra_a3_mod.F: Set negatives in QI, QL to zero, to avoid anomalous behavior elsewhere. These are mixing ratios and should not really be negative. Small negatives may have entered through the regridding process. Added this code to routine READ_A3 in GeosCore/merra_a3_mod.F:

     ! Reset negatives to zero in QI, QL fields (bmy, 9/5/13)
     WHERE( QI < 0d0 ) QI = 0d0
     WHERE( QL < 0d0 ) QL = 0d0

This error condition was caught by the GEOS-Chem unit tester when compiling with option FPE=yes.

--Bob Y. 15:20, 12 September 2013 (EDT)

Outstanding issues

Acetone ocean source

NOTE: This fix was rendered obsolete in 2011 when we added the Emily Fischer's updates for the ocean source and sink for acetone (cf. Emily Fischer) into GEOS-Chem v9-01-03b.

As a first cut, the ocean source of acetone for MERRA (computed in routine OCEAN_SOURCE_ACET in acetone_mod.f) is computed in the same way as for GEOS-5. The following #if block was based on the GEOS-5 code:

#elif defined( MERRA )

      !%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
      !%%% NOTE: Treat MERRA in the same way as GEOS-5, although    %%%
      !%%% we may want to rescale this in the future (bmy, 8/13/10) %%%
      !%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

#if defined( GRID4x5 )
      REAL*8, PARAMETER :: SCALE_FACTOR = 0.9551d0   ! MERRA 4x5
#elif defined( GRID2x25 )
      REAL*8, PARAMETER :: SCALE_FACTOR = 0.2212d0   ! MERRA 2x25
#endif

At some point we may need to rescale this. However this cannot be done until we have sufficient MERRA met fields processed and regridded.

--Bob Y. 16:24, 13 April 2011 (EDT)