Objective  The glaciers on the Qinghai-Tibet Plateau are an important freshwater reserve resource in China. In the context of global warming, glaciers on the Qinghai-Tibet Plateau are generally in retreat, affecting China’s water resources reserves. Therefore, it is necessary to monitor changes in glaciers on the Qinghai-Tibet Plateau. The existing DEM (Digital elevation model) for complex terrain applications in the Qinghai-Tibet Plateau region have insufficient descriptions of glacier details or data gaps, which cannot reflect glacier characteristics well and are not suitable for monitoring elevation changes in mountain glaciers and small ice caps. This article is based on the autonomous and controllable production of high-precision and finer grid DEMs in the glacier area of the Qinghai-Tibet Plateau using domestic stereo mapping satellites.

  Methods  When producing DEM using satellite remote sensing stereo images, field control points are often required. However, the research area is located in high mountain glacier areas, making it difficult for personnel to reach and collect control points. The GF-7 satellite laser altimetry data product has the characteristic of high elevation accuracy and can be used as elevation control points in complex terrain areas. The specific methods are as follows: Firstly, conduct preliminary screening of GF-7 laser points and select laser points in bare areas; Secondly, the matching algorithm based on phase correlation is used to determine the homonymous points of GF-7 laser points on the stereo image, establishing the connection between the laser points and the stereo image; Thirdly, establish the Affine transformation model between the laser point and the stereo image to realize the image square compensation of the stereo image, update the RPC (Rational Polynomial Coefficients, RPC) information of the stereo image, optimize the orientation parameters of the stereo image, and improve its stereo mapping accuracy; Finally, high-precision DEM is extracted through dense image matching to complete high-precision DEM extraction based on stereo images.

  Results and Discussions   In the research area, comparing the elevation accuracy of domestic three-dimensional surveying satellite DEM with medium spatial resolution DEM, as shown in Tables 4 and 5, it was found through ATL08 laser point verification that GF-7 DEM has the highest elevation accuracy, followed by ZY-3 DEM, AW3D, SRTM, and TanDEM has the worst accuracy, with a maximum error of over 20 m. Verify its accuracy with high spatial resolution HMA (High Mountain Asia, HMA) DEM data. From the validation results in Tab. 6, it can be seen that the elevation accuracy of GF-7 DEM is superior to HMA DEM, and the accuracy of HMA DEM is higher than ZY-3 DEM. The main reason is the limitation of basic data source resolution. HMA originates from sub meter resolution images, so it has more advantages in accuracy compared to ZY-3 DEM. From the perspective of DEM coverage, GF-7 DEM and ZY-3 DEM have comprehensive coverage, while HMA data has many data holes and poor quality, especially in glacier regions with less data coverage. In terms of detailed description of the glacier end, due to the finer grids of GF-7 DEM, ZY-3 DEM, and HMA, they have more advantages compared to other large grids, and the detailed texture features are clearer.

  Conclusions  Select laser data with similar collection times in the glacier area to verify the accuracy of GF-7 DEM and ZY-3 DEM. The accuracy of GF-7 DEM is better than 1.5 m when the slope is less than 6°, and the accuracy of ZY-3 DEM is better than 3.5 m when the slope is less than 6°. Both have high accuracy and can be applied to monitoring elevation changes in glacier areas. Domestic stereo mapping satellites have the advantage of being autonomous and controllable, providing stable data for glacier scientific research in the Qinghai-Tibet Plateau region. Over time, they can also form long-term time series data, better serving the monitoring of glacier changes in the Qinghai-Tibet Plateau.