Distinguish The Duststorm Source In Eastern Of The Inner Mongolia Autonomous Region On Multi-source Remote Sensing Data

The eastern of Inner Mongolia Autonomous Region, which is located in the the northeast forest and grassland ecotone, the northern farming and animal husbandry ecotone. The region is drought and water shortage, land desertification area and desertification, which ecological environment is very fragile. In recent years, the eastern of Inner Mongolia Autonomous Region,especially Tongliao, Chifeng and Xilinguole Meng area, arable land expansion, animal husbandry and other reasons over grassland,has gradually become an important source of dust weather sandstorms, a direct threat to the ecological security of the eastern of Inner Mongolia Autonomous Region and the northeast region. In order to discuss the eastern Mongolia potential dust source area, using remote sensing technology, from previous experience, combined with the field survey, the establishment of index system of eastern Mongolia potential dust source areas. Using MODIS data, field survey data, Landsant8 OLI data and the second land survey data,combined with the actual situation in the study area, were studied to obtain further dust source area classification index. The study area of land cover types, soil humidity and vegetation coverage data, based on the index system of potential dust source areas, dust potential on the source of identification and classification, and the potential dust source areas by index system accuracy verify.The main conclusions are as follows:(1) Land cover types, soil humidity and vegetation coverage of three indicators can be obtained by remote sensing data, the pixel values accurate to more accurately identify potential dust source regions of eastern Mongolia, compared to other indicators to more accurately identify potential dust source areas.(2) ?By using remote sensing technology, from previous experience, combined with field investigation and study area land cover the actual situation of the study area land cover classification research, studies have unearthed overall classification accuracy of land cover type classification (Overall Accuracy) was 0.795, Kappa coefficient is 0.739,better classification accuracy, can provide the basis for the identification and classification of the potential dust source areas.?The selection of spring vegetation cover as the index,because the spring plant litter is an important index to identify research area spring dust source area. Remote sensing technology combined with the use of MODIS data and hyperspectral data, typical MODIS type spring vegetation coverage estimation model based on established, has provided an important support for the accurate identification of study area potential dust source areas.?The use of hyperspectral data,do the water control experiment of cultivated land in the study area, found that compared with other drought index, surface water content index is more suitable for the characterization of soil moisture in study area.(3) Through overlay analysis, the dust source area of 151 thousand and 100 square kilometers of the study area, the total area of 44%, 61 thousand and 300 of mild potential dust source area square kilometers, accounting for 18% of the total area of the study area,the study area of 83 thousand and 100 square kilometers of moderately potential dust source areas, accounting for 24% of dust research area, the research area of severe 49 thousand and 500 potential dust source area square kilometers, accounting for the accuracy of classification and grading index system of total area of 14%. established in the study of the potential dust source area is 0.745, better classification accuracy, can be used as the study area index to identify potential dust source areas on the basis of.