Comparative analysis of freely available digital elevation models for applications in multi-criteria environmental modeling over data limited regions

Abstract

Remotely sensed elevation data obtained from high-resolution Digital Elevation Models (DEMs) is one of the main sources of data required for environmental modeling and remediation especially when it has to do with identification of erosion sites and flood-prone areas. The focus of this study is to evaluate four recently improved and freely available high-resolution elevation data obtained from the Advanced Land Observing Satellite World 3D Digital Surface Model version 2.1 (ALOS W3D30), TerraSAR-X add-on for Digital Elevation Measurements (TanDEM-X 90), Multi-Error-Removed Improved-Terrain (MERIT) DEM and EarthEnv-DEM90 in order to determine the most suitable DEM that could be used as an input data in a multi-criteria environmental modeling to identify erosion sites and flood-prone areas in data limited regions such as Southeastern Nigeria. Intensive performance evaluation of these four DEMs was rigorously carried out using statistical and surface analyses tools (Root-Mean-Square Error (RMSE), Land Use/Land Cover, slope, and contour maps). The obtained RMSE range from 5.49 m to 19.79 m depending on the surface of contact. Slope values range from 0.00 to 89.49° depicting the complexity and ruggedness of the terrain. Analyses of the obtained RMSE showed that EarthEnv-DEM90 is more accurate (in terms of RMSE value) than the other three DEMs but further terrain analyses using slope and contour maps generated from each of the DEMs revealed that ALOS W3D30 represent the topography of the terrain (terrain variation) better than the other three DEMs. This study demonstrates the importance of carrying out rigorous and intensive performance evaluation of DEMs using many terrain parameters or attributes in order to avoid large errors that may be contributed by the use of a DEM in applications such as environmental modeling and remediation.