Authors: Reza Saleh Alipour – The University of Alabama
Title: Sensitivity of Flood Inundation Mapping to Digital Elevation Data, Building Footprints, and Land Use/Land Cover
Presentation Type: Poster
Abstract: The heterogeneity of urban landscapes, including buildings and land covers, challenges the accuracy of flood inundation mapping (FIM). Research advances to urban datasets and FIM techniques have improved results; however, the level of sensitivity to urban terrain features remains a question. Knowing the sensitivity will help to guide flood model choice, dataset requirements, spatial resolution, quantification of uncertainty, and other critical elements needed to produce actionable FIM. This study seeks to quantify the sensitivity of FIM to urban features. A sensitivity analysis is conducted using the HEC-RAS 2D model and the “zero-order” FLDPLN FIM technique to generate flood inundation maps using digital elevation models (DEMs) at 3, 10, 30, 60, and 90-meter spatial resolutions and multiple elevation model types: digital terrain model (DTM), digital surface model (DSM), and digital building model (DBM). The Amite River Basin in Louisiana serves as the case study. In HEC-RAS 2D, results are more sensitive to DEM type and LU/LC than DEM resolution. For FLDPLN, coarser DEMs slightly reduce accuracy but remain effective. Overall, flood depths align well across all DEMs, with coarser resolutions offering comparable performance and lower computational costs. The insights gained will help guide the application of FIM techniques for a range of purposes, from flood mitigation planning, flood control engineering, and flood forecasting.