Framework for stochastic urban flood hazard mapping using coupled and industry-standard hydrologic and hydraulic models

Flood hazard mapping based on deterministic models does not represent the uncertainties inherent in the methods. Tools to characterize this uncertainty using industry-standard hydrologic and hydraulic models are lacking. This research presents SWMM-RASpy, an open-access Python tool to stochastically sample and analyze flood inundation using the widely-used Storm Water Management Model (SWMM) for hydrology and the Hydrologic Engineering Center’s River Analysis System (HEC-RAS) for channel hydraulics. Channel-floodplain hydraulics are represented in a two-dimensional, unsteady manner. The framework is tested in an urban watershed with stochastic sampling of flow roughness. For this watershed, it is shown that up to 4.5% more of the watershed and approximately double the number of buildings may be subject to flooding if roughness uncertainty is considered relative to a deterministic model. Flood hazard uncertainty is represented using an entropy map for clear communication, which could be used to improve flood risk management.