Experiment and analysis of retrieving land surface parameters using polarization radar images in Genhe area of China

For supporting the applications of GF-3 and GEOSAR satellite projects, which have been approved by Chinese government, a multi surface parameter inversion algorithm using alternative Radarsat-2 polarimetric images is tested and improved in this study. Result shows that, though the algorithm can retrieve vegetation height and water content with relatively good accuracies, soil moisture and surface roughness inversed results are poor when the above two vegetation parameters retrieved values as inputs to inverse two soil parameters. Analyzing the reason, scattering model could amplify the errors caused by the two retrieved vegetation parameters, which are inputted as “true value” in the model. The conclusion of this study is that, surface parameters can not be inversed step by step as the idea of engineering, scientists should focus on certain parameter(s), and take other parameters as disturbance factors to eliminate or suppress.