Abstract: With the Google Earth Engine (GEE) cloud platform, this study uses Landsat imagery, temperature precipitation and land use types data as the basis, and Theil-Sen Median trend analysis, Mann-Kendall test, partial correlation and multiple regression residual analysis to analyze the 1999—2018 the spatial and temporal characteristics of vegetation cover, change trends, and the effects of climate change and human activities on different land use types on the Loess Plateau in northern Shaanxi. The resufts are that: ①The interannual FVC of the Loess Plateau in northern Shaanxi shows an improving trend from 1999 to 2018, and its average growth rate is 0.0049/a (P<0.01), and the area with an increasing trend of vegetation cover accounted for 74.43% of the total area. ②the partial correlation coefficients of vegetation cover with precipitation and temperature has obvious spatial differences, and vegetation growth is more sensitive to changes in precipitation. ③Climate change and human activities together are the main causes of vegetation growth, where the effect of climate change on vegetation FVC ranges from -0.0010 to 0.0036/a, while the effect of human activities on vegetation FVC ranges from -0.0461 to 0.0490/a.④Among the different land use types, climate change increases the most for water bodies and the least for coniferous and broad-leaved forests; while human activity changes increase the most for human occupied land and the least for broad-leaved forests.

Key words: Loess Plateau of northern Shaanxi, partial correlation, multiple regression residual analysis, climate change, human activities