Abstract
MICRO-SWEAT, a soil–vegetation–atmosphere transfer scheme (SWEAT) coupled with a microwave emission model, was used to predict the microwave brightness temperatures (TB) measured at El Reno, Oklahoma, during the Southern Great Plains 1997 (SGP97) field experiment. Comparison with soil-moisture time series measured at four intensively monitored sites revealed the need for a substantially greater soil-saturated hydraulic conductivity than that estimated from soil maps. After revision of the hydraulic conductivity, the modeled and measured time series of soil moisture and surface energy fluxes showed excellent agreement with observations at these 4 sites and with the measurements of the surface soil moisture and TB at the remaining 11 measurement sites. A two-dimensional array of calibrated MICRO-SWEAT models was implemented at 200-m resolution for the El Reno area. There were noticeable differences between the spatial distributions of modeled and measured TB. These differences likely result from imperfect knowledge of the spatial distributions of soil properties, precipitation, and the estimated optical depth of the vegetation used in MICRO-SWEAT. A statistical measure of the usefulness of assimilating the observed soil moisture was explored by assuming the estimation of the optical depth provided the main source of error in the relationship between soil moisture and microwave brightness temperature. Analyses indicated that there is merit in assimilating TB observations for significant portions of the modeled domain, but it is suggested that this would be enhanced if the optical depth of the vegetation were also directly remotely sensed, as proposed in the Soil Moisture and Ocean Salinity (SMOS) mission.
Corresponding author address: Dr. Eleanor Burke, Dept. of Hydrology and Water Resources, University of Arizona, Tucson, AZ 86721. Email: eleanor@hwr.arizona.edu
