UAV-Derived DEM Bathymetry for Enhancing 1D Hydrodynamic Modeling for Non-perennial River
摘要
Floods are recurring natural disasters whose frequency and intensity have significantly increased over the past few decades. Hence, effective flood assessment and management are crucial for informed decision-making and mitigation strategies. Hydrodynamic models rely heavily on Digital Elevation Models (DEMs) to extract critical data such as cross-sections in 1D and floodplain details in 2D modeling. Recent advancements in Unmanned Aerial Vehicle (UAV) technology have enabled the development of high-resolution DEMs, providing unprecedented detail with resolutions as fine as 10 m for flood modeling for a section of the Sabarmati River in Gujarat, India, specifically when the river was in a non-perennial state. This UAV-based high-resolution DEM was used to extract cross-sections for 1D hydrodynamic modeling of the river and compared with Cartosat and ALOS DEMs. The modeling process involved using the 2006 flood hydrograph as the upstream boundary condition and normal depth as the downstream boundary condition. Sensitivity analysis revealed a high degree of accuracy in the model outputs, with R2 values of 0.99 for discharge from UAV DEM v/s Cartosat DEM and a similar value of 0.99 from UAV DEM v/s ALOS DEM. The R2value of 0.71 for energy gradient elevation (E.G. Elev.) for UAV DEM v/s Cartosat DEM and 0.91 for UAV DEM v/s ALOS DEM. These results demonstrate that UAV-based DEMs can significantly enhance 1D hydrodynamic analysis, providing accurate predictions of flood characteristics such as extent, depth, and duration. UAV-based high-resolution DEMs offer substantial improvements in flood risk assessment by providing precise cross-sectional and floodplain data.