The role of land surface processes and their feedbacks to the atmosphere is critical to an understanding of the climate system. We have undertaken process studies using the Regional Atmospheric Modeling System (RAMS) to understand the impacts of vegetation, soil moisture, and snow cover with a particular focus on the GAPP/GCIP region. In each of these studies, RAMS is used to explore the relative roles of land-surface and large-scale climate forcing on mesoscale processes and features.

RAMS has been dynamically coupled to ecological models to investigate the sensitivity of Great Plains climate to vegetation characteristics. Using the General Energy and Mass Transfer Model (GEMTM), Eastman et al. (2001) performed integrations with and without grazing. The grazing algorithm was employed to represent presettlement North American bison. Results indicate that grazing amplifies the diurnal temperature cycle and produces significant hydrological-cycle perturbations. Lu et al. (1999) and Lu (2001) coupled RAMS with the CENTURY ecological model to simulate two-way biosphere atmosphere feedbacks for the average year 1989, dry year 1988, and wet year 1993. They found that seasonal and interannual vegetation phenological variations strongly influence regional climate patterns through their control over land surface water and energy exchange.

Accounting for the effects of snow and its interaction with vegetation is also necessary for an accurate estimation of the surface energy and hydrologic budgets. Liston (2002) has developed a Subgrid-scale SNOW Distribution (SSNOWD) parameterization for use in regional and global atmospheric and hydrologic models. The snow-covered fraction is used to partition the surface energy fluxes over the snow-covered and snow-free portions of each grid cell. Liston (1999) showed that snow distribution is the first order influence on snow-cover depletion during snowmelt. In a study focused in the Great Plains, Strack et al. (2002) suggested that a mesoscale model must have accurate vegetation heights above the snow cover in order to reasonably simulate daytime low-level air temmperatures. The vegetation provides a moderating influence on temperature through longwave radiation emission and turbulent heat flux increases.

Current GAPP/GCIP work is investigating the physical processes of summer climate variability in the western and central U.S. In an observational study using the NCEP-NCAR reanalysis, Castro et al. (2001) showed the North American Monsoon has a time-evolving relationship to tropical and North Pacific SSTs which affects monsoon onset. Current RAMS simulations are exploring the effects of Pacific-SST associated summer teleconnection patterns on the evolution of monsoon hydrometeorological features. Later sensitivity experiments will explore the roles of soil moisture and snow cover. As part of this work, a new convection scheme (Kain-Fritsch), variable soil moisture, and variable soil types have been incorporated into RAMS. In a newly funded GAPP project, the earlier work of Eastman and Lu will be extended to explore the effects of modeling vegetation in a coupled seasonal weather prediction model. Specifically, an ensemble modeling system will be designed by coupoling GEMTM to both RAMS and the workstation ETA model. This system will be used to investigate warm season predictability under different summer weather regimes in the central U.S. The modeled surface energy budgets will be validated against data from the Oklahoma Mesonet and ARM-Cart Site. LDAS will be used to initialize the land state variables.