Meeting Minutes
Notes From the Third LDAS meeting: October 21st, 11am, NASA GSFC
LDAS Decisions and Discussion Summary:
NOTE: Please see the new LDAS Central Web Site at
http://ldas.gsfc.nasa.gov/ - many of the new LDAS datasets and
configuration issues will be detailed on this site. Please forward any
changes or new data sets for this site to Brian Cosgrove
(bac@ldas.gsfc.nasa.gov) or myself (Paul.Houser@gsfc.nasa.gov). This
site will eventually be moved to the NCEP server and be given an NCEP
address once some security/premission issues are resolved.
1) Decisions on Practical LDAS Issues
--Output fields from models involved in LDAS project should be written in GRIB format.
--Binning of elevation data to capture snow and orographic effects will be examined. (Princeton)
--Mean areal water equivalent, percent area covered by snow and coefficient of variation will each be
computed. This information will be used to derive the mean water equivalent and coefficient of variation
of snow covered areas.
--Data sets (vegetation, soils, etc...) will be merged through use of a GIS system.
(Princeton/Washington)
--IGBP soils data set will be used. (GSFC and Office of Hydrology will pursue data)
--Some variables will need to be corrected for elevation.
**AWIPS 212 elevation data set will be interpolated to 1/8th degree. The difference between this
field and the GTOPO 1/8th degree elevation field will then be computed. This field will then be used with
the standard published U.S. lapse rate to arrive at a temperature field corrected for elevation.
**Humidity will be corrected in a manner such that the relative humidity will remain constant (and
specific humidity will change) as temperatures are changed.
**Pressure must also be corrected.
--Radiation values will remain uncorrected for elevation. Zenith angle correction will be used for
temporal interpolation
--The same cutoff temperature should be used by all models to determine the form of precipitation
(rain/snow).
--Dan Tarpley's snow cover data will be made available for use by models at 1/8th degree resolution.
(NCEP)
--Tom Carol's snow water equivalent data will be made available for use by models at 1/8th degree
resolution. (NCEP).
--The one-day and one-month NCEP snapshots will not include any elevation or zenith angle corrections.
The snapshots will consist of EDAS fields at an hourly time interval and 1/8th degree spatial resolution.
The main purpose of these snapshots is to allow LDAS groups to work out the process of reading GRIB data
into their respective models. NCEP will provide standardized routines for reading and writing GRIB files.
--EDAS data or H. Van den dool's soil moisture climatology will be used to initialize the model from the
cold-start condition.
--First soil layer in all LDAS models will be 5cm, and the group is leaning toward using a total of 3 soil
layers.
NCEP will provide standardized routines for reading and writing GRIB files.
--IGBP vegetation characteristics will be posted on the LDAS web page. (GSFC)
--IGBP soils data will be located. (GSFC)
2) Suggestions:
--All models should use same soil and vegetation classes.
--Seasonality can be treated differently by each model.
--Sacramento model may be used.
3) Updates:
--Configuration of NCEP SGI going slowly.
--Fortran 90 compiler loaded onto NCEP SGI.
--W3 library will be tested and loaded onto machine.
This site was constructed and is maintained by Brian A. Cosgrove: Brian.Cosgrove@gsfc.nasa.gov