Abstract
Digital Elevation Models (DEMs) are essential tools for applications such as flood risk mapping, geoid modelling, and terrain analysis. The increasing availability of global digital elevation models (GDEMs), including the Shuttle Radar Topography Mission (SRTM), NASA Digital Elevation Model (NASADEM), and Copernicus DEM (CopDEM), has significantly improved access to elevation data worldwide. However, their accuracy varies depending on topography and data acquisition methods. This study evaluates the performance of SRTM, NASADEM, and CopDEM in the Perlis State region of Malaysia. Ground truth data were obtained from 38 Global Navigation Satellite System (GNSS) benchmarks and 1,177 points of Advanced Topographic Laser Altimeter System (ATLAS) onboard the Ice, Cloud, and land Elevation Satellite-2 (ICESat-2). Prior to analysis, ellipsoidal heights were converted to orthometric heights using the Earth Gravitational Model 1996 (EGM96) geoid model. DEM elevation values were interpolated at each validation point, and accuracy was assessed using Mean Error (ME), Standard Deviation (STD), and Root Mean Square Error (RMSE). Results indicate that NASADEM demonstrates the highest accuracy, with RMSE values of ±1.855 m (GNSS) and ±5.814 m (ICESat), followed by CopDEM and the SRTM DEM. All three (3) GDEMs show strong positive correlations with ground truth data (R > 0.98). These findings suggest that NASADEM provides the most reliable elevation data for the Perlis region, particularly in relatively flat areas. This study