Toward an Operational Bare Soil Moisture Mapping Using TerraSAR-X Data Acquired Over Agricultural Areas

摘要: TerraSAR-X data are processed for an “operational” mapping of bare soils moisture in agricultural areas. Empirical relationships between signal and soil were established validated over different North European study sites. The results show that the mean error on estimation is less than 4% regardless configuration (incidence angle, polarization) surface characteristics (soil roughness, composition). Furthermore, potential (signal, texture features) to discriminate from other land cover classes watershed was evaluated. backscattered can be easily differentiated signals when neighboring plots covered by fully developed crops. This observed conditions. When early growth crops, a image acquired under wet conditions useful discriminating soils. Bare masks calculated object-oriented classifications mono-configuration data. overall accuracies higher 84% validation based object pixel. method applied images generate maps. sensors provide monitoring spatial variations at within-plot scale. methods this paper used operational applications agriculture, hydrology.