Comments on “An Adaptive Terrain-Dependent Method for SRTM DEM Correction Over Mountainous Areas”

This paper comments on an article published in IEEE Access (C. Zhou et al., “An Adaptive Terrain-Dependent Method for SRTM DEM Correction Over Mountainous Areas,” IEEE Access, Vol. 8, pp. 130878-130887, 2020, DOI: 10.1109/ACCESS.2020.3009851) that presented a method for correcting the Shuttle Radar Topography Mission (SRTM) digital elevation data product. The present comments concern the proposed method and the data used by the authors of that paper. First, Zhou et al. ignored a fundamental accuracy limit of any digital elevation model (DEM) imposed by elevation data's discretization and quantization operations. These limits are intrinsic to DEMs and cannot be overcome. The current paper presents a detailed mathematical model of these limits. Second, Zhou et al. used GPS stations' elevation as a reference, which likely does not represent ground elevations as required. Third, the authors incorrectly named and interpreted the SRTM minus reference elevations by describing them as the “absolute error of the SRTM DEM. ” These shortcomings of the Zhou et al. paper warrant corrections to their findings.