A Surface Moisture Inversion Technique Using Multi-polarization And Multi-angle Radar Images

Spatially distributed soil moisture and its change information are important for hydrology, climatology and ecology. Also, soil moisture information is a very important index for drought-forecast and agricultural yield estimation. Originally commence with optical remote sense data, researchers have taken advance in retrieving soil moisture from remote sensed data, in spite of limits of these data. With extension of active microwave remote sense applications, these data are fetching up the optical sensed data in recent years. Some new methods are developed. However, some crucial problems are to be resolved in retrieving soil moisture from SAR data.The satellites that can currently meet the spatial resolution and coverage required for watershed management are active microwave sensors. Practically, C-Band SAR fits to soil moisture retrieval. In this thesis, we use ENVISAT-ASAR for this aim. And our test area locates in Hetian Xinjiang, which is between Kalakulun Moutain and Takylamakan Deserts. The ASAR system can provide resolutions from 10-100m over a swath width of 50-100km, thus meeting most spatial requirements for watershed-scale application. We demonstrate a method using the multi-polarization and multi-angle C-band ENVISAT ASAR data to estimate the relative soil moisture. Then the inversion algorithm was applied to the VV and VH polarization measurements of ENVISAT ASAR. The results can show that the estimated values of soil moisture for bare and short-vegetated fields are compared favorably with in situ data.