LDAS Logo
Meeting Minutes




meeting.12.21.98

Notes From the Fifth LDAS Meeting: December 21st, 9:30am, NASA GSFC


Decisions on LDAS issues:

The next LDAS meeting will take place at NCEP at 9:30 on February 8th, 1999.

VIC will run at NCEP and at Princeton.

A GRIB information session will take place after the next LDAS meeting at NCEP. During this meeting, Curtis Marshall will cover the mechanics of reading/writing GRIB using the GRIB libraries.

During Phase 1 of the LDAS project:

--All groups will use the vegetation data sets that are appropriate for their respective models.

--All groups will use the same land/sea mask.

--No forcing will be provided over water points.

--Each group will only be required to model land points.

--Output will only be required for land points.

--All groups will write out hourly model output of all state variables.

During Phase 2 of the LDAS project:

--Vegetation classes will be unified such that all groups will use a unified vegetation data set within their models.

1-km greenness data and vegetation fraction data will be obtained from Dan Tarpley.

The greenness seasonal cycle will be rendered onto the LDAS grid. (OH)

Greenness data will be placed on the LDAS web site. (NASA)

OH will locate and obtain 1km IGBP global soil data.

OH will provide 1km soils data on LDAS grid from Miller's data (11 layers of data down to 1.52 meters) and will try to extrapolate down to 2 meters.

A primary database of soil hydrological properties will be constructed by OH and will contain:

    1. Soil Texture
    2. Soil Depth

A secondary database of soil hydrological properties will be constructed by OH and will contain:

    1. Ksat
    2. Porosity
    3. % Rock
    4. Wilting Point
    5. B Parameter
    6. Field Capacity
    7. Diffusivity

At the next LDAS meeting, OH will present a table containing a list of soil hydrological properties and sources of the hydrological data.

NCEP will provide real-time data, but no retrospective data.

NCEP will develop a precipitation algorithm that blends daily rainfall gauge data with national stage 2 NEXRAD data and EDAS data. The NEXRAD data will be used to provide hourly weights for the 24-hour gauge data, while the EDAS data may be used to fill in missing data.

A proposed set of common output variables for all models has been created. These variables will be written out on an hourly basis into 1 file, with daily totals arising from post processing activities:

  1. State Variables
  2. --Volumetric Soil Moisture

    --Volumetric Soil Ice

    --Soil Temperature (K)

    --Snow Water Equivalent (mm)

    --Snow Depth (mm)

    **Snow density can be derived from snow depth and water equivalent

    --Canopy Water (mm)

    --Skin Temperature [consistent with upward longwave values] (K)

  3. Energy Fluxes (Hourly Mean Values)
  4. --Downward Solar Radiation (W/m2)

    --Upward Solar Radiation (W/m2)

    --Downward Longwave Radiation (W/m2)

    --Upward Longwave Radiation (W/m2)

    --Latent Heat Flux (W/m2)

    --Sensible Heat Flux (W/m2)

    --Ground Heat Flux (W/m2)

    --Snow Phase Change Heat Flux (W/m2)

  5. Vegetation State
  6. --Canopy Conductance (m/s)

    --Soil Resistance (s/m)

    --Net Effective Resistance (s/m)

    --Aerodynamic Resistance (s/m)

  7. Water Fluxes
  8. --Precipitation (mm)

    --Snowmelt (mm)

    --Potential Evaporation (mm)

    --Actual Evaporation (mm)

    --Direct Evaporation (mm)

    --Transpiration (mm)

    --Canopy Evaporation (mm)

    --Surface Runoff (mm)

    --Base-flow (mm)

 

A list of output data variables will be posted on LDAS web site (NASA).

Format specifications of output data will be posted on LDAS web site (NASA).

 

Suggestions:

UMD vegetation may be the best choice, as it appears that vegetation parameter data is available for this data set.

3 soil data sets may be used: STATSGO soils, IGBP soils, and a blend of both where IGBP data is used to fill in missing STATSGO data.

LDAS groups may purchase hard drives for use at NCEP for storing LDAS data.

Satellite data could be used to fill in missing precipitation data.

Both raw and merged precipitation data sets should be made available to LDAS participants. These data sets may include EDAS, 24-hour, NEXRAD and merged precipitation data.

A diagram of the structure of the file that contains the common output variables should be constructed.

 

Updates:

Configuration of NCEP SGI is going well.

Fortran 90 and C compilers loaded onto NCEP SGI.

W3 library loaded onto machine.

Unified land/sea masks (based on the IGBP data) for the IGBP, SiB, BATS and UMD data sets have been placed on the LDAS web site. In order to create this mask, it was necessary to change 9 pixels in the BATS and SiB data, and approximately 500 pixels in the UMD data.

No progress made on obtaining IGBP soils data.

A one-day snapshot is available from NCEP in GRIB format.

--Snapshot now includes GOES 1/2 degree radiation data that has been gridded to 1/8th degree.





This site was constructed and is maintained by Brian A. Cosgrove: Brian.Cosgrove@gsfc.nasa.gov