Accuracy Evaluation of Free Global DEMs for Topographic Mapping Applications Across Nigeria’s Diverse Physiographic Zones

Global digital elevation models available without charge, including SRTM, ASTER GDEM, AW3D30, NASADEM, and COP-DEM, are extensively used for topographic analysis in Nigeria due to the absence of national high-resolution terrain data, yet their vertical accuracy across Nigeria's diverse physiographic contexts has not been comprehensively characterised. This study evaluates the vertical accuracy of five free global DEMs against geodetically surveyed reference points distributed across six physiographic zones: the coastal lowlands, Niger Delta, basement complex plains, Jos Plateau, Benue Trough, and Sahel zone. A stratified random sample of 1,440 reference points collected using differential GNSS levelling in all six zones is compiled as the accuracy reference dataset. Vertical RMSE, mean error, and 90th percentile linear error statistics are computed for each DEM in each physiographic zone, and ANOVA tests assess the significance of accuracy differences between DEMs and between zones. Results demonstrate that COP-DEM achieves the highest overall vertical accuracy nationally with RMSE of 4.7 metres, followed by NASADEM at 6.2 metres, but accuracy is markedly lower in the dense vegetation canopy environments of the Niger Delta and rainforest zones where canopy heights contaminate surface models. ASTER GDEM performs least well in all physiographic zones. The study provides a comprehensive DEM accuracy reference for Nigerian practitioners selecting elevation data for engineering, hydrological, and cadastral applications. Keywords: global DEM, vertical accuracy, Nigeria, SRTM, physiographic zones.