Abstract: Geological disasters in mining areas have become more and more serious in recent years. For accurate monitoring of surface subsidence with complex topography of tailings dam, based on the monitoring data of UAV(Unmanned Aerial Vehicle) low-altitude photogrammetry, the UAV original POS(Position and Orientation System) data error were improved, data from the error correction model was used to correct the original POS model and 7 kinds of control point layout were designed, high resolution evaluation was conducted on the orthogonal projection as well as the DEM(Digital Elevation Model) accuracy. The results show that when the number of image control points is 8, the data error can be controlled within 3 mm. The settlement map of the mining dam is generated by overlaying the ground model with the difference values of the two DEM data, and the profiles with Y=350 m, Y=100 m and X=60 m were made respectively. The measurement results indicated that the tailing dam has been subsided as a whole, and the southern mining dam has the largest subsidence area, which is within 0.16 m. This application verifies that the accuracy of UAV low altitude photogrammetry in mining dam surface monitoring is reliable. The high-precision mapping method of UAV is used to monitor the deformation of tailing dam, which plays a certain early warning role in the ecological disaster terrain of the oasis area and the surrounding rivers and lakes which may be caused by the emergency response of dam break and the safe production of mining area.