Extraction and analysis of mountain glacier movement from GF-1 satellite data

With ongoing and accelerating global climate change, temperate glaciers are very sensitive to variations in the temperature and precipitation, and thus are in fact regarded as natural indicators of climate change. Glacier velocity, which is a combination of ice deformation, bed deformation, and glacier sliding, is an important parameter to better study the dynamics of glaciers and their interplay with climate changes in the region. In situ observations serve as one of the most accurate methods for measuring glacier velocity, but the remote areas where glaciers develop have prevented frequent visitation by people. Remote sensing is more effective in glacier monitoring and has been applied to study glacier velocity in many regions of the Tibetan Plateau, such as in Karakoram, Himalaya, West Kunlun, and other areas. In recent years, the Chinese high-resolution optical remote sensing images has gradually increased, but there was not much use of Chinese produced satellite remote sensing images for monitoring glacier flow parameters in mountain regions. In view of this situation, this study tried to apply the domestic "GaoFen-1" satellite images (GF-1) to the extraction of Yanong Glacier flow in southeast Tibet. By preprocessing and applying feature tracking on all available pairs within a defined period to derive the velocity, the reliable glacier velocities can be obtained by selecting stable ground control points from the ice-free areas to register these GF-1 data and the offset can be computed. The accuracy of the glacier flow velocity derived from GF-1 data was assessed by the residual displacements in non-glacial stable regions and glacier flow velocity along the longitudinal profile of the Yanong Glacier compared to that of Landsat-8 data in the same resolution and at the same Periods. The evaluation results showed that: The average deviation of GF-1 data in non-glacial stable regions was 7.48 m•a-1, which was higher than that of landsat 8 (the average deviation was 4.58 m•a-1), but less than 5% of the average velocity of glacier; The GF-1 data was consistent with Landsat 8 data in the change trend of glacier flow velocity along the longitudinal profile in the period from 2015 to 2016, and the mean square root of the deviation between them was 7.41 m, which was also less than 5% of the average glacier flow velocity. The results proved the feasibility of the application and the unique advantages of GF-1 satellite remote sensing data in monitoring of the mountain glacier velocity on the Qinghai-Tibet Plateau. © 2021, Science Press. All right reserved.