A Global-Scale DEM Elevation Correction Model Using ICESat-2 Laser Altimetry Data

Spaceborne laser altimetry technology assists global digital elevation models (GDEMs) to improve the accuracy of elevation data due to its highly accurate range and wide coverage. Compared to the previous laser altimeter systems used for Earth observation, Ice, Cloud, and land Elevation Satellite (ICESat-2) has a sensitivity for photon detection that can provide more accurate and denser surface elevation observations. This article proposed a DEM correction model using ICESat-2 data. The model used the altimetric data to verify the DEM elevation errors in ICESat-2 coverage areas first. Then, an attribute set was constructed to evaluate the error sources of the global-scale DEM. The evaluations of the error sources include the location/positioning of the platform, atmospheric conditions, topographic relief, land cover, heterologous infill data, and so on. Finally, a regression model was constructed by the attribute set and the DEM elevation errors within ICESat-2 coverage areas, in order to correct the DEM in areas without ICESat-2. In the validation experiments, this study conducted elevation correction experiments using the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) GDEM and the shuttle radar topography mission (SRTM) in three regions around the world and applied the airborne light detection and ranging (LiDAR) data in each region to verify the corrected results. The results showed that the proposed model was suitable for the elevation correction of global-scale DEMs and can be applied to more than 90% of global land, i.e., land areas with a slope
${ < } 25^{\circ }$
. The accuracy improvement ratios of the corrected GDEM were 17.89%–33.33% across the different types of topography, and the accuracy improvement ratios of the corrected SRTM were 27.77%–44.64% across different types of topography.