Overview of GMAO met data products

From Geos-chem

Jump to: navigation, search

On this page we shall list the features of the various meteorological data products produced by NASA's Global Modeling and Assimilation Office. These products are used to drive GEOS-Chem.

Contents

GMAO met field comparison chart

  GEOS-1 GEOS-STRAT GEOS-3 GEOS-4 GEOS-5 MERRA GEOS-5.7.2
Notes on GEOS vertical grids 20 sigma 26 sigma 48 sigma 55 hybrid 72 hybrid 72 hybrid or
42 pressure 		72 hybrid
Native horizontal grid 2° x 2.5° 2° x 2.5° 1° x 1° 1° x 1.25° 0.5° x 0.666° 0.5° x 0.666°
1.25° x 1.25°
1° x 1.25° 		0.25° x 0.3125°
Native file format Binary Binary HDF4-EOS HDF4-EOS HDF4-EOS HDF4-EOS netCDF4/HDF5
Regridded to 2° x 2.5°
4 x 5 		2° x 2.5°
4° x 5° 		1° x 1° nested
2 x 2.5°
4° x 5° 		2° x 2.5°
4° x 5° 		0.5° x 0.666° nested
2° x 2.5°
4° x 5° 		4° x 5° TBD
Temporal coverage 1985 - 1995 1996 - 1998 2000 - 2002 1985 - 2007 2004 - present 1979 - 2010 2011 -

NOTES:

  1. GEOS-1 and GEOS-STRAT are now obsolete and no longer supported by GEOS-Chem
  2. GEOS-3 is still supported but we recommend users to migrate to GEOS-4 or GEOS-5.
  3. MERRA data will be made available at 4 x 5 resolution in the near future. MERRA is a long-term reanalysis done with version GEOS-5.2.0.
  4. The new GEOS-5.7.2 data product is slated to be released in Summer 2011. This data set will have an increased horizontal resolution (1/4° lat x 5/16° lon).
  5. Please also see our Notes regarding GEOS vertical grids wiki page for more information about how the sigma and hybrid grids are defined.

GEOS-5 and MERRA

GEOS-5 is the current GMAO data product. It is the first ultra-high-resolution data product that GMAO has produced. MERRA is a multi-year reanalysis done with the GEOS-5 GCM (version 5.2.0).

Summary

Assimilation GCM used: NCAR fvCCM GCM + NASA convection modules
Native horizontal grid:
Native vertical grid: 72 hybrid levels; model top at 0.01 hPa
Native file format: HDF4-EOS
Temporal coverage:
  • GEOS-5 operational data: 2004-present; some 2003
  • MERRA reanalysis data: 1979-2010
  • GEOS-5.7.1 operational data: 2011-
We regridded to:

GEOS-5 operational data:

MERRA reanalysis data:

  • 4° x 5° with 72 hybrid layers
Strengths:
  • High resolution will facilitate comparison to satellites
  • Convection, optical depth better than in GEOS–4
Weaknesses:
  • Large data volume: 1 day of “raw” data files = 3 GB!

NOTES:

  1. The GEOS-5 and MERRA data are kept at 72 vertical levels, and are only regridded to coarser horizontal resolution. However, GEOS-Chem has the capability to lump levels on the fly from 72 to 47 vertical levels. This is done for computational expediency.
  2. In the GEOS-Chem data archive:
  3. The new GEOS-5.7.2 data product was officially released in 2011. This data set has increased horizontal resolution (0.25° x 0.3125°), but the same # of vertical levels (72) as the older GEOS-5 and MERRA products.
  4. Please also see this document which details the evolution of the GEOS-5 system from GEOS-5.2.0/MERRA to GEOS-5.7.2 (aka "Fortuna").

References

  1. List of GEOS-5 met fields used by GEOS-Chem
  2. GEOS-5 issues encountered w/r/t GEOS-Chem
  3. Reprocessing of GEOS-5 met data for GEOS-Chem (completed Jul 2009)
  4. GEOS-5 products page at GMAO
  5. GEOS-5 File Specification Document (version 5.1.0)
  6. GEOS-5 File Specification Document (version 5.2.0)
  7. MERRA on the GEOS-Chem wiki
  8. MERRA grid structure
  9. MERRA file naming convention
  10. MERRA mainstream and spinup data
  11. List of MERRA met fields used by GEOS-Chem
  12. MERRA File Specification Document
  13. MERRA data holdings at GSFC
  14. GEOS-5.7.2
  15. List of GEOS-5.7.2 met fields for use with GEOS-Chem

--Bob Y. 16:33, 2 August 2011 (EDT)

GEOS-4

GEOS-4 was produced at higher vertical resolution but slightly coarser horizontal resolution than GEOS-3. There are approximately 20 years of GEOS-4 data available to be used with GEOS-Chem. GEOS-4 has been used extensively by the GEOS-Chem community in various applications.

Please see our GEOS-4 wiki page for more information.

Summary

Assimilation GCM used: NCAR fvCCM GCM
Native horizontal grid: 1° x 1.25°
Native vertical grid: 55 hybrid levels; model top at 0.01 hPa
Native file format: HDF4-EOS
Temporal coverage: 1985-2006; some 2007
We regridded to: 4° x 5° at 55 sigma layers
Strengths:
  • Long temporal horizon – 20 years of consistent met data
  • U and V winds are 6-hr averages, not instantaneous
  • Much better STE than any prior GEOS met fields
Weaknesses:
  • Convection variables totally different than GEOS–3
  • GMAO was unhappy w/ the convection results
  • Cloud OD’s are too low; J-values, chemistry are “too hot”

NOTES:

  1. The GEOS-4 data are kept at 55 vertical levels, and are only regridded to coarser horizontal resolution. However, GEOS-Chem has the capability to lump levels on the fly from 72 to 30 vertical levels. This is done for computational expediency.

References

  1. Documents which describe the validation of GEOS-4 met fields in GEOS-Chem
  2. GEOS-4 references
  3. List of GEOS-4 met fields
  4. Availability of GEOS-4 met fields

--Bob Y. 12:40, 29 July 2010 (EDT)

GEOS-3

Removal of support for GEOS-3

This update was tested in the 1-month benchmark simulation v9-01-03h and approved on 09 Mar 2012.

Bob Yantosca wrote:

As you may know, the GEOS-3 met (for data years 2000-2001) is now several versions behind the current GMAO operational product. If there is no longer a significant demand to keep GEOS-3 met, then I would propose removing the GEOS-3 specific source code (denoted by “#if defined( GEOS_3 )” statements). This will help to clean up the source code, which in turn will facilitate the work that Mike Long and I are doing on the Grid-Independent GEOS-Chem project.

Support for GEOS-3 meteorology will be removed in GEOS-Chem v9-01-03.

--Melissa Payer 12:18, 2 March 2012 (EST)

Summary

GEOS-3 was an improvement over GEOS-1 and GEOS-STRAT. However, because GEOS-3 was a pure-sigma model, there were problems in the stratosphere caused by excess noise in the winds. The GEOS-3 product was extensively used in GEOS-Chem for TRACE-P mission analysis.

Assimilation GCM used: NASA/GSFC GEOS–3 GCM
Native horizontal grid: 1° x 1°
Native vertical grid: 48 sigma levels; model top at 0.01 hPa
Native file format: HDF4-EOS
Temporal coverage: 2000-2002
We regridded to: 1° x 1° nested grids, 2° x 2.5°, 4° x 5° at 48 sigma layers
Strengths:
  • High resolution made 1° x 1deg; nested sims possible
  • Higher model top
Weaknesses:
  • P to sigma vertical interpolation had to be done
  • STE still not realistic
  • U, V winds were very noisy at the top of the atmosphere

NOTES:

  1. The GEOS-3 data are kept at 48 vertical levels, and are only regridded to coarser horizontal resolution. However, GEOS-Chem has the capability to lump levels on the fly from 48 to 30 vertical levels. This is done for computational expediency.

--Bob Y. 12:41, 29 July 2010 (EDT)

GEOS-1 and GEOS-STRAT

GEOS-1 was the first GMAO met field data product compatible with GEOS-Chem. GEOS-STRAT was a continuation of GEOS-1, but with a finer vertical grid. Both products have been superceded by newer versions of GMAO met products and are no longer supported in GEOS-Chem.

GEOS-1 features

Assimilation GCM used: NASA/GSFC GEOS1 GCM
Native horizontal grid: 2° x 2.5°
Native vertical grid: 20 pure-sigma layers, model top at 10 hPa
Native file format: NASA PHOENIX format (binary Fortran unformatted)
Temporal coverage: 1985-1995
We regridded to: 4 x 5; 20 pure-sigma layers
Strengths:
  • Long temporal coverage (10 years)
Weaknesses:
  • Very low model top (10 hPa)
  • Strat-trop exchange was too fast by 2-3X

GEOS-STRAT features

Assimilation GCM used: NASA/GSFC GEOS1 GCM
Native horizontal grid: 2° x 2.5°
Native vertical grid: 46 pure-sigma layers, model top at 0.1 hPa
Native file format: NASA PHOENIX format (binary Fortran unformatted)
Temporal coverage: 1996-1997 (some 1998)
We regridded to: 4 x 5; 26 pure-sigma layers
Strengths:
  • Higher model top, better stratospheric representation
Weaknesses:
  • Vertical grid did not correspond to GEOS–1
  • STE still too fast by 2-3X
  • Surface data had a 6-hr temporal resolution, we had to interpolate to a 3-hr temporal resolution ourselves

--Bob Y. 16:43, 27 July 2010 (EDT)

Retrieved from "http://wiki.seas.harvard.edu/geos-chem/index.php/Overview_of_GMAO_met_data_products"
Personal tools