Evaluating the Impact of the Spatial Resolution of Digital Elevation Models on Flood Modelling

Abstract

Accurate flood modelling is crucial for effective disaster management. This study investigated the impact of different Digital Elevation Model (DEM) resolutions (1 m UAV, 10 m InSAR, and 30 m SRTM) on flood modelling within the Shiroro floodplain, Nigeria. Using the Shallow Water Equations (SWE) implemented in MATLAB to assess flood levels along 12 river channels within the study area, the findings demonstrated that the UAV-derived 1 m DEM provided the most accurate flood predictions, exhibiting lower RMSE values (0.249 m) compared to InSAR (0.352 m) and SRTM (0.455 m). The higher resolution of the UAV DEM captured topographic details more effectively, leading to more accurate flood predictions, in contrast to the low resolution of InSAR and SRTM, which resulted in an overestimation of flood risk due to the smoothing and generalization of topographic features. These results highlight the critical importance of using high-resolution DEMs, particularly those derived from UAV imagery, for accurate flood modelling and effective flood risk management, as these high-resolution datasets can significantly improve flood forecasting, refine flood hazard maps, and inform targeted flood mitigation measures, ultimately enhancing disaster preparedness and response within the Shiroro floodplain and beyond.