Abstract: Assessment of vegetation change and its causes in arid and semiarid regions are important for ecosystem protection project management. However, slow changes of vegetation coverage in arid and semiarid regions are difficult to explain through land use/cover type changes. Aimed to analyze the sensitivity of climate driving factors on vegetation changes, the human-induced and climate-driven vegetation changes in the Xilingol Grassland, Inner Mongolia from 2001 to 2014 were distinguished. Results show that vegetation coverage in the study area from 2001 to 2014 has an increasing trend, and the annual maximum normalized difference vegetation index (NDVI) is 0.0018. The decision coefficient ( R ²) of the vegetation index prediction model constructed by temperature and precipitation factors is 0.65, better than the results only using temperature factor ( R ²=0.46) or precipitation factor ( R ²=0.54). The contribution of human factors to the increase in vegetation cover is much greater than that of climate factors. The RESTREND method is a useful tool to improve the assessment of human-induced vegetation changes in arid and semiarid regions, and can provide scientific data support for the prevention and control of desertification work.