A remote sensing monitoring method for alpine grasslands desertification in the eastern Qinghai-Tibetan Plateau

Alpine grassland is the typical vegetation in the eastern Qinghai-Tibetan Plateau, which has important ecological service functions, and also supports the development of alpine stock farming. In recent years, under both the natural and human disturbance, alpine grasslands in this area have appeared to different degrees of desertification. A diagnosis of the desertification degree serves as the basis for grassland ecological restoration. This study constructs a comprehensive index based on remote sensing called alpine grassland desertification index (AGDI) to monitor the areas and degree of desertification. The most relevant indicators of desertification, namely, vegetation fraction, aboveground biomass, soil moisture, and land surface temperature, were selected to establish AGDI. The geographical detector is used to reselect and assess these indicators. The results show that the overall verification accuracy of AGDI is 82.05%. In particular, the accuracy of identifying severe desertification is the highest. Our study confirms that the desertification of alpine grasslands in the eastern Qinghai-Tibetan Plateau is characterized by fragmentation. Thus, Landsat-8 OLI data with a spatial resolution of 30 m is more suitable than MODIS data for alpine grasslands desertification monitoring. The research results can provide a methodological reference for monitoring desertification of alpine grasslands and other grassland regions in the world.