关键词: NDVI_max, 趋势, 人类活动, 空间格局

Abstract: We analyzed the temporal and spatial changes of grassland vegetation cover and its response to human activities in Qinghai Province by trend analysis method, linear correlation analysis, and geographical weighted regression method, based on Landsat 8 OLI, GIMMS NDVI 3G data, MODIS data and statistical data from 1982 to 2018. The results showed that: (1) From 1982 to 2018, NDVI_max of grassland in Qinghai Province showed an overall decreasing trend, with an average rate of 1.2×10^-4. (2) The NDVI_max was sensitive to grazing intensity in the southwest and central parts of Qinghai Province, while it was not sensitive in the northeast. This may be related to the complementary advantages of agriculture and animal husbandry and the implementation of ecological engineering. (3) The regression coefficients between NDVI_max and per capita GDP, traffic, residential area, and total human population were 0.20, 0.11, 0.04 and 0.03, respectively. The local regression coefficients of per capita GDP and grassland NDVI_max were high in the west and low in the east, while that for other regression relationships were high in the west and east and low in the east region. This is mainly due to the rich mineral resources in the west and a transportation hub and population gathering area in the east.

Key words: NDVI_max, trend, human activity, spatial pattern.