Soil Moisture and Salinity Inversion Based on New Remote Sensing Index and Neural Network at a Salina-Alkaline Wetland

In arid and semi-arid regions, soil moisture salinity are important elements to control regional ecology climate, vegetation growth land function. Soil salt content more in wetlands. The Ebinur Lake wetland is an part of the ecological barrier Junggar Basin Xinjiang, China. a representative area climate degradation central Asia. It great significance study spatial distribution its causes for restoration Basin. Based on field measurement Landsat 8 satellite data, variety remote sensing indexes related were tested compared, prediction models established, accuracy was assessed. Among them, D1 D2 latest ones that we proposed according research data. maps retrieved from correlation analysis between performed. several indexes, normalized index NDWI had highest with moisture, salinity, reaching 0.600 0.637, respectively. BP neural network model estimating higher than one other models; R2 = 0.624, RMSE 0.083 S/m. effect cubic function also network, support vector machine 0.538, 0.230. regularity changes seemed be consistent, degree 0.817, synchronous change higher. generally low surrounding area, high middle lake coverage area. slightly decreased outward as center west lower east. However, mutation rate stronger heterogeneity salinity.