This data includes six first-class types and 38 second-class types, including forest land, grassland, wetland, cultivated land, artificial surface and others, with a scale of 1:250000. The data reflects the land cover in Xinjiang in 2010. Taking 1:100000 Xinjiang topographic map, 1:1 million Xinjiang land use data, 1:1 million Xinjiang grassland resources data, 1:1 million Xinjiang forest land resources data, 2010 TM data with a resolution of 30 meters in Xinjiang area, DEM data with a spatial grid accuracy of no less than 30 meters as data sources, under the environment of ERDAS, ecognition and ArcGIS, the method of combining expert analysis and remote sensing automatic classification is adopted In accordance with the contents and regions, thematic mapping is used to complete the interpretation accuracy verification in cooperation with the field investigation results. See the data document for the definition and code meaning of each type. The data can comprehensively reflect the current situation of land cover / land use in Xinjiang in 2010. Combined with the data of land cover in 2000 and 2005, it can reflect the land cover changes in Xinjiang in the past 10 years, and provide scientific data support for Xinjiang land use research and national macro decision-making.
|collect time||2010/01/01 - 2010/12/31|
|collect place||Xinjiang Uygur Autonomous Region|
|data size||304.0 MB|
|Data time resolution||year|
|Projection||Orthogonal equal area double standard weft secant conic projection of Albers ellipsoid|
In 2010, Xinjiang land cover data was independently completed by Xinjiang Institute of ecology and geography, Chinese Academy of Sciences. On the basis of 1:1 million Xinjiang land use data and remote sensing classification data, it was completed in a full digital way. The main data sources are: 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 woodland resource data (vector); 5. The above data are from the accumulation of scientific research achievements over the years.
Under the environment of ERDAS, ecognition and ArcGIS, through the combination of expert analysis and remote sensing automatic classification, and with the results of field investigation, the interpretation accuracy is verified. According to the content and area, the method of thematic mapping is adopted. Main links: Based on 1:1 million Xinjiang land use data as the basic base map, as the frame data and control basis of mapping; Based on the TM image data within the time range of 1-2 years before and after 2010, the orthophoto map is generated by superposition of DEM data whose spatial grid accuracy is not less than 30 meters. The orthophoto map is interpreted by remote sensing image analysis software ecognition to realize the attribute division and mapping of different land cover types; On the basis of the above three kinds of data, combined with the results of artificial visual interpretation and field investigation, full digital mapping is carried out.
The results show that the total precision of the first class of land cover is 95%, the second class is 85%, the second class of forest and shrub is 85%, and the lowest precision of other single class is 65%; the total precision of the second class of each climate zone is 85%; the total precision of the second class of change detection is more than 72%, and the lowest precision of single class is 42%.
©Copyright 2005-2020. Northwest Institute of Eco-Environment and Resources, CAS.
Donggang West Road 320, Lanzhou, Gansu, China (730000)