Abstract: Based on a light-use efficiency model and MODIS data, the net primary productivity (NPP) of grassland during the growing seasons from 2000 to 2018 in native grassland in Inner Mongolia was continuously estimated, and the spatial-temporal change trend of NPP over the past 20 years was also analyzed using a least square method at pixel scale to determine the degree of grassland degradation. It was found that NPP increased from west to east across Inner Mongolia, with an annual average of 198.04 g C·m^-2·yr^-1. An area of potential grassland degradation of 162200 km² was identified. The areas of severely and heavily degraded grassland identified were 2000 and 11100 km², respectively. The latter were mainly distributed in areas with intensive human activities, such as mining areas, or construction land and its surrounding areas. In terms of grassland types, the potentially degraded areas of temperate steppe, temperate meadow, temperate desert and temperate desert-steppe were 52200, 14000, 40400 and 22100 km² respectively. Through analyzing the correlation between NPP and climate factors in the last two decades, it was found that NPP is significantly correlated with precipitation, but has no correlation with temperature. The response of NPP to precipitation was the most sensitive in temperate meadow, that of temperate desert steppe was the second, and that of temperate steppe was the lowest. Based on the above results, this paper puts forward some suggestions for grassland ecological protection.

Key words: light-use efficiency model, MODIS, net primary productivity of grassland, climate factors