Using SIR-C SAR and Cloude's decomposition for the determination of soil moisture in vegetated areas

Decomposition of SIR-C polarimetric SAR backscattering coefficients into odd (direct return, single-bounce) and even (dihedral, double-bounce) reflection mechanisms was performed on images of selected study areas of the Chickasha, Oklahoma hydrological supersite. The sensitivity of the direct and dihedral reflection strengths to field-measured volumetric soil moisture was evaluated, and the results show that the strength of the direct reflection component changes at a different rate than the dihedral component in response to soil moisture changes, thus the ratio of the two reflection mechanisms can be used to estimate soil moisture. This ratio changes at a constant rate with soil moisture for several different landcover types and bare soil. Soil moisture estimation using this method may be possible in areas with moderate to heavy vegetation, areas where previous soil moisture algorithms produced poor results.