Surface soil moisture retrieval using a regression method applied on polarimetric SAR images

Soil moisture plays a critical role in many hydrological processes including infiltration, evaporation, and runoff. Satellite-based passive microwave sensors are sensitive to soil moisture content, (Cashiona et al., 2005), but the spatial resolution of these sensors is much lower than visible/infrared (VIS/IR) satellite data. However, soil moisture estimates from VIS/IR sensors usually require surface micro-meteorological and atmospheric information that is not routinely available. To overcome these problems, radar remote sensing technology has been widely used for soil moisture retrieving because of its capability to operate in all weather conditions (Xiao et al., 2005). This study is concerned with retrieving percentage of soil moisture content using polarimetric PALSAR data in Iran. Scattering matrix decomposition has been used to find optimum features that have higher correlation with soil moisture content. Therefore, a regression based method has been used to estimate the soil moisture content based on these features. According to the validation, as expected, the retrieved soil moistures were consistent with those of field measurements. It can be concluded in this study that proposed method is appropriate in mapping soil moisture content as a suitable alternative to sparsely distributed meteorological stations measurements.

کلیدواژه ها:

Soil moisture ، SAR ، scattering matrix decomposition ، feature extraction ، regression method