Combination of passive and active microwave data for soil moisture estimates

Various remote sensing techniques have been evaluated and proven to be a valuable source of information for different hydrological applications. For example, with the actual Earth observation satellites, we can observe the entire river basin in rather than sparse points and provide unique information about properties of the surface or shallow layers of the Earth. Furthermore, the actual remote sensing sensors offer the potential of measuring new hydrologic variables not generally possible with traditional techniques such as soil moisture, snow status, land cover parameters etc. Previous researches in microwave remote sensing technology indicate that surface soil moisture can be inferred with remote sensing systems operating in the microwave region of the electromagnetic spectrum. The ability to estimate soil moisture in the upper surface layer by microwave remote sensing (active and passive) has been demonstrated under a variety of the topographic and land-cover conditions. The primary intent of this project is to produce a spatial estimation of soil moisture from active microwave data with sufficient spatial and temporal resolution using neural networks. The derived soil moisture was analyzed in conjunction with vegetation data to understand the effect of land cover on the soil moisture variation. This paper describes the first steps in evaluating the performance of the neural network classification and presents some of the early results.