The drainage of groundwater caused by grassland mining activities leads to the decrease of groundwater level, which caused a dramatic degradation of grassland vegetation in surounding area. Understanding how grassland mining activities impact on the intensity and range of groundwater depth, determining threshold of depth to the groundwater table for vegetation growth would be a crucial prerequisite to protect the ecological environment of grassland mining area. In this study, Yimin open-pit coal mine in Hulun Buir grassland was investigated as an example. By introducing remote sensing method, the quantitative response relationship between depth to the groundwater table and Normalized Difference Vegetation Index (NDVI) before and after mining was established. How the composition of vegetation community change to the groundwater level and how the depth to the groundwater table influence the vegetation in terms of types, species richness, vegetation coverage, aboveground biomass and summed dominance ratio in grassland mining areas was determined by analyzing the data of sample survey. Further, combined with the results from two methods, the threshold of depth to the groundwater table for maintaining the normal growth of vegetation in the study area was determined.The results are as follows: From Chedaminol Lake to the edge of the mining pit, the depth to the groundwater table decreased gradually from 0 m to 60 m, and the vegetation community was successively from salinized grassland to typical grassland to degraded grassland to degraded meadow grassland to salinized meadow grassland. A depth to the groundwater table of about 1 m was most suitable for growing vegetation; A depth to the groundwater table of 1-30 m was the threshold to maintain the normal vegetation growth in the study area. Finally, according to the above conclusion, the study area is divided into the first sensitive area, the second sensitive area and the third sensitive area and corresponding policy recommendations were proposed for different sensitive areas.