Study On Remote Sensing Information Model For Desertification Evaluation Of Keerqin Sandy Land

This study is supported by the program of National Science Foundation of China (30371192)It is one of the serious ecological problems arousing world attention to desertification that influences the development of human economy and stability of the society directly. This paper is mainly to study quantitative monitoring and evaluation for desertification by using Remote Sensing(RS) data. At first, the spatial registration of RS data and ground investigation dada are made by stationary differential GPS, then on the basis of analysing desertificated research current situation and trend synthetically, a rough evaluation index system of Keerqin sandy desertification was put forward, at the same time, the visual RS model for desertificated evaluation is established by the retrieval results, at last, according to the field data, the results of the evaluation are checked up and adjusted by GIS. This paper expectes that it can offer some new motheds and modes to monitor and evaluate the desertification quantitatively. Main research conclusions and innovative points are made from the paper:1. The research of the desertificated evaluation index is one of the key problems. Based on analysis the acton of all the factors choosed, three main influencing factors: geomorphology, soil and vegetation are selected.2. In the desertificated regions, the phenomena of mixed pixels is particularly serious. This paper uses ENVI to achieve the spectral mixed analysis and does the correlation analysis with NDVI and Tasseled Cap. The results prove that SMA is welcome. The occupation ratio of bare sandy land is one of the quantitative evaluated factors.3. The RS information model put forward in recent years is a main respect of the quantitative RS technology. This paper establishes the visual RS model for desertificated evaluation with the retrival results based on the reasonable index system. The precision of the evaluation is 86.7%, which proves that the mothed is feasible and scientific.4. The evaluated results enter GIS in raster, and get these conclusions:severe and extreme regions is 25.8 percent in 1999. but 23.02 percent in 2004;slight regions in 2004 are more than 263 km in 1999. It achieves to quantitatively monitor and evaluate the desertificated region.