1. 1:1 million Xinjiang land use data; 2. 2010 TM data with a resolution of 30 meters; 3. DEM data with a resolution of 30 meters; 4. Auxiliary data 1:1 million Xinjiang grassland resource data (vector), 1:1 million Xinjiang forest land resource data (vector); 5. 2009-2010 Photos and recorded data obtained from field inspections and positioning in 2000.

In environments such as ERDAS, eCognition, and ArcGIS, through a combination of expert analysis and remote sensing automatic classification, combined with field field inspection results to verify the interpretation accuracy, it is completed by using thematic mapping according to the content and area. Main links: Based on 1:1 million Xinjiang land use data as the basic base map, as the framework data and control basis for mapping; Based on TM image data within a time range of 1-2 years before and after 2010, an orthophoto map is generated by superposing DEM data with a spatial grid accuracy of not less than 30 meters, and remote sensing image analysis software Ecogitation is used to remotely sense the orthophoto map. Interpret, realize attribute division and mapping of different land cover types; On the basis of the above three types of data, combined with manual visual interpretation and field inspection results, full digital mapping is carried out.

The total accuracy of the first-level category of land cover at the sample scale is>95%, the total accuracy of the second-level category is 85%, the accuracy of the second-level category of forests and shrubs reaches 85% respectively, and the lowest accuracy of other single categories is>65%; The total accuracy of the second-level category of each climate zone reaches 85%; the total accuracy of the second-level category of change detection is greater than 72%, and the lowest accuracy of the single category is greater than 42%.