Compound Flooding in River-Urban-Coastal Environments: Multi-factorial Drivers and Modeling Considerations

Abstract

In the National Capital Region, existing coastal flood guidance systems frequently underestimate total water levels (TWL), leading to inaccurate flood predictions. Flood forecasting in this region, located at the confluence of two major rivers (Potomac and Anacostia) with tidal connections to the Chesapeake Bay, faces TWL under-predictions due to missing physical processes, limited integration of hydrological and hydrodynamic models, and simplified operational model frameworks. This study introduces an integrated TWL framework using a high-resolution two-dimensional coastal storm surge model (ADCIRC) that includes multiple flood drivers (storm tide, river flows, urban runoff, and local wind forcing) as one-way input boundary conditions in the tidal Potomac River. This framework accurately replicates historical events based on observed data, with validations indicating a 0.1 m under-prediction at the NOAA Washington, DC tide station (WASD), representing a -5% deviation from observed maximum water levels. Through hypothetical simulations for 25-, 50-, and 100-year return periods, we emphasize the substantial impact of individual flood drivers. Local winds had the smallest impact on water levels at WASD compared to downstream storm surge from the Chesapeake Bay (Lewisetta, VA). Upstream major river discharges elevate water levels beyond the National Weather Service (NWS) major flooding level by 0.9 m, further amplified to 1.4 m above the threshold when urban discharges occur simultaneously in the National Capital Region. Unlike prior studies, our work offers a comprehensive evaluation of individual flood drivers' influence on TWL modeling, underscoring the imperative need for their inclusion in the framework to accurately estimate river, coastal, and compound floods in estuarine metropolitan areas.