Soil Moisture Inversion with Sparse Vegetation Coverage Areas and Analysis of Spatial Characteristics Based on RADARSAT-2

Abstract

Backscatter coefficient of radar imagery is mainly affected by the surface soil moisture and roughness. This paper was based on the RADARSAT-2 polarimetric SAR image data, field measured soil moisture data, surface roughness, soil particle composition and soil bulk density data. The soil moisture from SAR was retrieved by using Oh2004 semi empirical model inversion method, Oh2004 semi empirical model LUT method, AIEM theory model which is based on the effective correlation length and AIEM theory model on account of the effective RMS height. After soil moisture retrieval of the study area sampling points at 0-6cm depth, this study analyzed the corresponding to the four kinds of methods of the optimal filtering window size, made a comparative analysis of the four methods and determined the optimum inversion model of the study area soil moisture microwave remote sensing at 0-6cm depth. This article used the best model for retrieving soil moisture in depth of 0-6cm, made the distribution map of soil moisture at 0-6cm depth, and analyzed spatial distribution characteristics.