HR: 0830h
AN: A31E-0088    [PDF]
TI: Initialization of the Purdue Mesoscale Model with the Land Data Assimilation System
AU: * MIn, K
EM: min@purdue.edu
AF: Earth and Atmospheric Science Purdue University, 550 Stadium Mall Dr, West Lafayette, IN 47907 United States
AU: Sun, W
EM: wysun@purdue.edu
AF: Earth and Atmospheric Science Purdue University, 550 Stadium Mall Dr, West Lafayette, IN 47907 United States
AU: Bosilovich, M G
EM: mikeb@dao.gsfc.nasa.gov
AF: Global Modeling and Assimilation Office NASA Goddard Space Flight Center, Code 900.3, Greenbelt, MD 20771 United States
AU: Chern, J
EM: jchern@dao.gsfc.nasa.gov
AF: Global Modeling and Assimilation Office NASA Goddard Space Flight Center, Code 900.3, Greenbelt, MD 20771 United States
AB: Analysis data derived from the Finite Volume Data Assimilation System (FVDAS, NASA Global Modeling and Assimilation Office) and Land Data Assimilation System (LDAS, NASA Hydrological Sciences Branch) have been successfully implemented in a meso-scale model and are tested to provide initial conditions and lateral boundary forcings. Initialization of the Purdue Mesoscale Model­_s precipitation forecast with FVDAS and LDAS high-resolution land surface and soil moisture data are compared with ECMWF data. Initial conditions of the land surface provided by FVDAS/LDAS show significant differences in both soil moisture and ground temperature compared to ECMWF, which results in a much different prediction of atmospheric state and weather variables. The simulation result shows that significant changes to the forecasted weather system are due to such differences, especially for the atmospheric state over the land surface. Comparison of precipitation, moisture budgets, and surface energy shows that, not only the intensity and location of precipitation over the Midwest is in better agreement with FVDAS/LDAS run, but also the temperature forecast is in better agreement when compared to ECMWF data. However, the precipitation over the Rockies is too large due to the cumulus parameterization scheme used in the model. The RMS errors and biases of FVDAS/LDAS are smaller than the control run and show statistical significance supporting the conclusion that use of LDAS improves the precipitation and temperature forecast for the case of Midwestern flood which occurred during May 21~23, 1998. The implementation of FVDAS/LDAS data to a mesoscale model is the first case experiment to date, and more case studies will be conducted in the near future.
DE: 3300 METEOROLOGY AND ATMOSPHERIC DYNAMICS
DE: 3322 Land/atmosphere interactions
DE: 3329 Mesoscale meteorology
DE: 3337 Numerical modeling and data assimilation
SC: Atmospheric Sciences [A]
MN: 2003 Fall Meeting