Environmental Sciences Department Seminar
"Land Surface Impacts on Convective Precipitation Development over the United States Great Plains"
Presented by Thomas W. Collow, Rutgers University
Thursday, August 22, 2013, 10:30 am
Conference Room, Bldg 815E
Hosted by: Wei Wu
A series of modeling experiments was conducted using the Weather Research and Forecasting Model to assess the sensitivity of mesoscale convective precipitation patterns to vegetation and soil moisture on a short time scale. For vegetation, runs were done over the Southern Great Plains of the United States using current vegetation cover, a uniform forest cover, a uniform barren land surface, and a pre-farming scenario in which cropland was changed to native grassland. The goal was to determine how vegetation impacts precipitation and whether pre-farming conditions would result in any meaningful alterations. Individual case studies were chosen to include days with both strong and weak synoptic forcing. Extreme changes in vegetation impacted precipitation, 2 m temperature, 2 m dewpoint, and the convective available potential energy (CAPE). Barren land decreases dewpoint, minimally affects temperature, and decreases CAPE. Forested land decreases temperature, increases dewpoint, and increases CAPE. Changes were more extreme for cases with little synoptic forcing but still substantial in all cases. Strong precipitation reductions occur with a barren land surface while some increases occur on a forested surface. Pre-farming conditions had little impact on the evolution of convective precipitation systems, showing that while vegetation cover is an important component in mesoscale precipitation, the switch from grassland to cropland was insignificant at this scale over this particular region. This means we found no evidence that "rain follows the plow." A similar procedure was followed for soil moisture in which initial model soil moisture at all levels was set to the porosity (very wet) and wilting point (very dry). The feasibility of using soil moisture data from the Soil Moisture Ocean Salinity (SMOS) Satellite was also analyzed. SMOS data were directly inserted into the WRF model and it was found that the changes were minimal compared to using orig