Abstract: The north slope of Tianshan Mountain is located on the Silk Road Economic Belt, with frequent human activities and many ecologically fragile areas. Exploring the temporal and spatial variations of vegetation coverage in this area is of a great significance for regional green development. Based on Google Earth Engine (GEE) cloud platform, the normalized difference vegetation index (NDVI) and vegetation coverage were calculated by using Landsat Image data from July to August of 1990, 2000, 2010 and 2020. The spatial agglomeration of vegetation coverage was analyzed by Moran’s I index, and the changes of high vegetation coverage were predicted by Grey Prediction GM (1,1) model. The results showed that: (1) Vegetation coverage on the north slope of Tianshan Mountain was generally low from 1990 to 2020, and that in 2020 showed an increasing trend compared with 1990. (2) According to the results of spatial autocorrelation analysis, the overall vegetation coverage on the north slope of Tianshan Mountain showed significant positive spatial autocorrelation (P<0.01), with an aggregation state. The results of local spatial autocorrelation analysis in different years were slightly different, which was mainly reflected in the increase of high-high aggregation in the middle part of the study area and low-low aggregation in the south edge of the study area. (3) The results of Grey Prediction GM (1,1) model showed that the average annual temperature in the study area would gradually increase from 2020 to 2040, the annual precipitation would decrease, and the area with high vegetation coverage showed an increasing trend. (4) The high efficiency of GEE cloud platform in processinglarge-scale and long-time series image data makes it an effective tool for vegetation coverage normalization monitoring.

Key words: Landsat, Google Earth Engine (GEE), grey prediction model, spatial autocorrelation, Tianshan Mountain.