Abstract: Seasonal frozen soil has the physical characteristics of periodic surface uplift/settlement, and traditional measurement methods can no longer meet the current high-precision and real-time monitoring needs. Ground-based GNSS is a low-cost, all-day, all-weather, and emerging ground-based remote sensing technology that can achieve continuous monitoring. In this paper, based on the observation data of SG27 station of the U.S. Continental plate boundary observation (PBO) GNSS netuork program from 2013 to 2021, the ground-based GNSS technology is used to interpret the snow in typical abnormal years in the Permafrost region of Barrow, Alaska, and the surface deformation, station deformation, soil moisture, and atmospheric water vapor changes in the last nine years of long time series snow free period. The retrieval accuracy of snow depth in abnormal years is verified through the PBO measured snow data, Verify that the inversion result is the deformation of the permafrost active layer through the deformation results of the measuring station, and conduct correlation analysis between water vapor and soil moisture. The results show that the absolute coefficient R² of retrieved snow depth and measured snow depth is 0.8155, the root mean square error (RMSE) is 0.0643, and the mean absolute error (MAE) is 0.0402; Through the trend chart of water vapor and soil moisture changes, it is found that there is a weak lag relationship between the two on the rainfall threshold, but only in the trend rather than the amplitude. It shows that ground-based GNSS has great application potential in long-term permafrost environmental monitoring..