Remote Sensing Information Extraction Of Soil Factors In Land Degradation Assessment

At present, the land degradation research has become the hot spot which the domestic and foreign soil science, the agronomy, the environmental science and the global change research pay attention to. The remote sensing is one of the indispensable ways to gain the foundational data of factors in the land degradation synthesis assessment. It is very important in academic research and the reality applications to get along with the newest trend of the domestic and foreign land degradation research promptly, and study the technology of extracting surveillance factors by remote sensing in land degradation assessment on the basis of the fact in our country, creatively. This research is precisely to study the methods of extracting information of the factors about soil by remote sensing, based on the study area of the arid environment. The practical and applicable results are expected.The main results and the innovation of the dissertation are as following:I, The test of soil classification based on remote sensing in the study area of Ebinur Lake in Xinjiang provinceTo establishment and confirm a method, we take the area of Ebinur Lake as study area which stands for a typical arid environment. With the field-derived data and supplement data of soil genesis taxonomy, we analyze the relations between soil and landscape characteristics of remote sensing information, and establish soil classification system suitable for remote sensing in the study area. Then, the classification model has been tested and the effect has been analyzed and appraised.In the test, the unit of soil remote sensing classification refers to the subclass of soil genesis taxonomy. According to J-M distances between soil subclasses based on characteristics mainly derived from remote sensing information, some subclasses are adjusted to adapt to remote sensing classification. There are 8 in 17 subclasses have been merged or deleted, and 9 new classes have been gained at last. The merged subclasses mainly include some ones with the same landscapes of farmland, covered by the crop.