Soil moisture estimation using special satellite microwave/imager satellite data over a grassland region

Abstract

The special sensor microwave/imager (SSM/I) is an instrument that has been a component of several satellite platforms since 1987. Although not designed for soil moisture sensing, it is possible based on theory to extract soil moisture information under some conditions. The limiting feature of the SSM/I for soil moisture-related studies is that the frequencies are quite high and are significantly affected by vegetation. However, other features of the data, such as the frequency of measurements, are very good for observing time-varying hydrologic variables such as soil moisture. There have been no quantitative evaluations of the SSM/I using observed soil moisture data. In this study, data collected in two large-scale experiments conducted over the Little Washita watershed, in Oklahoma were available for evaluating the capabilities of SSM/I data for soil moisture mapping. Physically based models were used to relate the satellite data to the ground observations. The results indicated that for this grass-dominated subhumid area a soil moisture-emissivity relationship with an error of estimate of 5.3% could be developed that incorporated the range of temperature and vegetation conditions encountered. An approach to adapting this approach for other vegetation regimes is still needed for wider application.