Can Watershed-Scale Green Stormwater Infrastructure Mitigate Floods? A Hydrologic Modeling Study

Stormwater management has traditionally focused on improving rapid (or “efficient”) transport of runoff from roadways via combined or storm sewers known as grey stormwater infrastructure. Alternatively, green stormwater infrastructure (GSI), a nature-based solution to flooding, attempts to capture, treat, and infiltrate stormwater at the site scale. The Stormwater Management Model (SWMM) was used to study GSI’s mitigative capacity to prevent pluvial flooding in Clear Run, a headwater tributary to Bradley Creek in Wilmington, North Carolina. The Clear Run watershed is approximately 80 ha and drains a mixed development of commercial and residential. Land use was digitized from aerial images for pervious (35.1%) and impervious (64.9%) surfaces. The existing 15 km of grey conveyance infrastructure were surveyed to obtain location, pipe inverts, diameters, and materials. High-resolution (1.5-m by 1.5-m) digital elevation models (DEMs) which were burned with pipe invert elevations were used to delineate watersheds. Stage-storage curves for the 17 wet retention basins and 1 dry infiltration basin were generated from DEM derived elevation contours. Models were bounded by existing rainfall records collected hourly at Wilmington International Airport (KILM) and every two minutes at nearby research stations during the duration of the monitoring. Models were calibrated to flow gauge depths collected, starting July 2020, at two-minute resolution upstream of a 1.8-m corrugated steel double barrel culvert at College Acres Drive. Existing conditions of the watershed were used for calibration and to serve as a baseline comparison. Modeled GSI retrofit scenarios included the following: (1) convert the vegetated traffic islands into infiltration basins with the parking lot as a multi-use floodplain detention basin, and (2) reduce residential runoff by disconnecting directly connected impervious surface (e.g., disconnect rooftop downspouts). Combinations were evaluated for flood resiliency to continuous rainfall records and discrete floods.