Comprehensive Application Of Remote Sensing Technology In Southeast Yunnan Rock Desertification

Southwest China karst areas restricted by regional special geological environmental, and karst rocky desertification as the performance characteristics of ecological environment degradation phenomenon gradually intensified. Karst rocky desertification has become the third major ecological problems in China, following the desertification in Northwest Territories and the soil erosion in Loess Plateau. By using the advantages of remote sensing, it can macroscopically, quickly, exactly and economically access to the rocky desertification status and distribution and through multi-temporal remote sensing data dynamic monitoring can be obtained the temporal dynamics evolution of karst rocky desertification, and afford the scientific accordance in following karst rocky desertification comprehensive managements.This paper takes1990,2000, and2007three-phase TM/ETM+image data as remote sensing original data, and takes ASTER GDEM data as the terrain sources data, obtains the lithology distribution data by vectoring the regional geological maps. Establishes the karst rocky desertification classification index system in study area by analyzing the current research on rocky desertification classification indicators, respectively using decision tree method, comprehensive analysis method and quantitative synthesis method remote sensing monitoring the rocky desertification status, extracts the karst rocky desertification classification information in study areas, and quantitative synthesis method is best after accuracy evaluation. Then using the difference method dynamic monitoring rocky desertification in study area, through statistical analysis results obtains the changing rules of two periods. The main research results of this paper are as follows:(1) Reference on the normalized difference vegetation index (NDVI) construct ideas, through analysis the TM image data characteristics, puts forward using TM7and TM4to build the normalized difference rock index (NDRI), applying NDRI dimidiate pixel model method extracts the information of rock bare rate.(2) According to the karst rocky desertification classification index system in study area, respectively using decision tree method, comprehensive analysis method and quantitative synthesis method remote sensing monitoring the rocky desertification status, extracts the karst rocky desertification classification information in study areas, based on the field survey data to evaluate three methods accuracy, the verification results shows that the quantitative synthesis method works best.(3) Using1990,2000and2007three-phase remote sensing data, obtain the three phase karst rocky desertification status in study area by using quantitative synthetic method, and using the difference method dynamic monitoring karst rocky desertification in study area to obtain the two times rocky desertification changes status and distribution. According the statistical analysis shows, in study area the karst rocky desertification is intensifying, but its further aggravate rate gradually reducing, and improving in some area.(4) According the three-phase remote sensing monitoring results, analysis the correlation between karst rock desertification status and lithology, the results shows that limestone areas prone to rocky desertification and phenomenon is serious, carbonate rocks with clastic rock areas is relatively low incidence of rocky desertification, dolomite rock areas rocky desertification annual variation is small.