Modeling Climate Change Impacts on the Effectiveness of Stormwater Control Measures in Urban Watersheds

Abstract

Climate change (CC) science has made significant progress in development of predictive models. Despite these recent advances, the assessment of CC impacts in urban watersheds remains an area of active research, in part due to the small temporal and spatial scales needed to adequately characterize urban systems. Urban watersheds have been the focus of considerable efforts to restore hydrology and water quality, and the aquatic habitat of receiving waters, yet CC impacts threaten to reduce the effectiveness of these efforts. Thus, assessing the impacts of CC in urban watershed assessment are essential for assuring the success of water quality improvement programs and is an important research need. Simulations of CC for the 2041-2068 period were developed using downscaled Global Climate Models (GCMs) from the North American Regional CC Assessment Program (NARCCAP) and Coupled Model Intercomparison Project Phase 5 (CMIP5) to forecast precipitation and temperature time series. This data were then used to force a Storm Water Management Model (SWMM) of the Difficult Run watershed of Fairfax County, Virginia, a tributary of Potomac River, which flows into Chesapeake Bay. NARCCAP uses a scenario represents a medium-high greenhouse gas emissions assumption, A2; the latter, uses five GCMs, and two Representative Concentration Pathways (RCP 4.5 and 8.5) scenarios in an ensemble approach to better assess variability of model predictions in presenting precipitation, temperature, runoff quantity and quality. Then, the effects of CC on runoff peak, volume, and nutrient and sediment loads delivered to the Chesapeake Bay and on the treatment performance of a very common stormwater control measure (SCM), retention ponds, was assessed. Rainwater Harvesting (RWH) systems are an unusual SCM in that they recycle and reuse stormwater, normally from rooftops, and increase water supply and reduce runoff. The efficiency of RWH systems for projected CC for these dual purposes was assessed. NARCAAP data for selected locations across the U.S. were statistically downscaled using a modified version of the equiratio cumulative distribution function matching method to create a time series of projected precipitation and temperature. These data were used to force a simulation model, the Rainwater Analysis and Simulation Program (RASP) to assess the impacts of CC on RWH with respect to the reliability of water supply and runoff capture. To support CC modeling, an easy-to-use software tool, RSWMM-Cost, was developed. RSWMM-Cost automates the execution of SWMM, which is commonly used for simulating urban watersheds. Several features were incorporated into the RSWMM-Cost tool, including automated calibration, sensitivity analysis, and cost optimization modules; the latter can assist in identifying the most cost-effective combination of SCMs in an urban watershed. As an example, RSWMM-Cost was applied to a headwater subcatchment the Difficult Run watershed.