Using the EPA’s SUSTAIN Model to Assess the Capability of Best Management Practices (BMPs) to Improve Water Quality in the Los Angeles Basin
Friday, 19 December 2014
The desire to control the flow, storage and release of stormwater in urban systems distinctly alters the flow paths, water quality, and flood behavior of these once natural systems. The EPA and local municipalities have identified Low Impact Development (LID) and Best Management Practices (BMPs) as more natural methods to manage stormwater in urban areas, though the efficacy of these methods has not been quantified at the watershed scale. By holistically modeling stormwater flow at the watershed scale, engineered BMP solutions can be optimized for cost and evaluated on their ability to improve water quality and flood protection. In this research, the EPA System for Urban Stormwater Treatment and Analysis INtegration (SUSTAIN) model is applied to heavily urbanized watersheds in Southern California, specifically the Ballona Creek and Dominguez Channel watersheds (61% and 69% impervious land cover, respectively). Stormwater flow and pollutant load of selected metals are calibrated and validated over the gauged portions (89 sq. mi. in Ballona and 33 sq. mi. in Dominguez) and simulated over the full watersheds (~128 sq. mi. in both). Scenarios of BMP implementation are optimized for cost and pollutant load reduction. Results show that receiving water quality improvement is limited by BMP pollutant removal effectiveness. Consequently, larger BMP treatment capacity does not result in greater pollutant load reduction. The optimal scenario routed runoff from 90% of the watershed area through a variety of regional and distributed BMPs indicating that BMP implementation needs to focus on strategic placements to capture runoff from the largest drainage area instead of only maximizing BMP capacity. Secondary benefits such as peak flow reduction of 20-50% and recharge potential between 12,000 and 30,000 acre-ft per year were also demonstrated in selected Ballona Creek optimization scenarios.