Snow cover is one of the important components of land cover, and it is necessary to accurately monitor the depth and coverage of snow cover. Using the GPS signal receiver data and the basic principle of snow depth detection based on GPS-MR technology, the snow depth of the three sites on the Greenland PBO network GLS1, GLS2, and GLS3 from 2012 to 2018 was obtained. The inversion snow depth is affected by site drift, which is a quite difference from the measured snow depth. Combined with the stable reference point, the velocity field distribution of Greenland Island and the U-direction component change value of the station can be obtained through GAMIT calculation. By analyzing the glacial flow and U-direction component, the influence of the site drift on the snow depth was deducted, and finally compared the corrected inversion snow depth and measured snow depth found that the two were better than before the correction, the results were significantly improved, and the consistency was good. The analysis of the experimental results showed that in extremely cold areas such as Greenland Island, affected by glaciers, the continuous, real-time, high-time resolution snow depth around the measured station obtained by ground-based GPS tracking stations has a large gap with the measured snow depth value, and the gap will gradually increase with time. By deducting the impact of glacier drift, the trend of the two is the same and the consistency is good. The correctness and feasibility of the application of ground-based GPS snow cover theory in the polar area further expand the application scope and practical value of ground-based GPS in snow monitoring.