Soil moisture estimation by using multipolarization SAR image Original Research Article

In the past studies, different soil moisture estimation models were developed for bare soil areas by using remotely-sensed data. However, there are few models that can be used to estimate soil moisture in vegetated areas. Water Cloud Model (WCM) model is a widely used soil moisture estimation model has been developed for vegetated areas. In this study, the WCM model was extended to take soil roughness parameter into consideration. The modeling and its accuracy assessment were done by using multi-polarization Airborne Synthetic Aperture Radar (AIRSAR) images and ground data collected during field Soil Moisture Experiments. It was shown that the backscatter coefficient of HV cross-polarization is more accurate than HH or VV co-polarization to be used in the proposed Improved Water Cloud Model (IWCM) model. Also, experiments showed that the co-polarization ratio σHH/σVV has high correlation with Vegetation Water Content (VWC). Therefore, in order to make the model independent from ground measurements, the ratio was used alternatively in the proposed IWCM model. The analysis on the IWCM model showed a meaningful improvement on soil moisture estimation accuracy.