H41G-1451
Watershed Scale Impacts of Stormwater Green Infrastructure on Hydrology and Nutrient Fluxes in the Mid-Atlantic Region.

Thursday, 17 December 2015
Poster Hall (Moscone South)
Michael J Pennino, Princeton University, Princeton, NJ, United States, Peter R Jaffe, Princeton Univ, Princeton, NJ, United States and Rob McDonald, The Nature Conservancy, Urban Sustainability, Arlington, VA, United States
Abstract:
Stormwater green infrastructure (SGI), including rain gardens, detention ponds, bioswales, and green roofs, is being implemented in cities across the globe to help reduce flooding, decrease combined sewer overflows, and lessen pollutant transport to streams and rivers. Despite the increasing use of urban SGI, there is much uncertainty regarding the cumulative effects of multiple SGI projects on hydrology and water quality at the watershed scale. To assess the cumulative effects of SGI, major cities across the mid-Atlantic were selected based on availability of SGI, water quality, and stream flow data. The impact of SGI was evaluated by comparing similar watersheds, with and without SGI or by assessing how long-term changes in SGI impact hydrologic and water quality metrics over time. Most mid-Atlantic cities have a goal of achieving 10-75% SGI by 2030. Of these cites, Washington D.C. currently has the highest density of SGI (15.5%), while Philadelphia, PA and New York, NY have the lowest (0.14% and 0.28%, respectively). When comparing watersheds of similar size and percent impervious surface cover, watersheds with lower amounts of SGI, on average, show up to 40% greater annual total nitrogen and 75% greater total phosphorus loads and show flashier hydrology (as indicated by 35% greater average peak discharge, 26% more peak discharge events per year, and 21% higher peak-to-volume ratio) compared to watersheds with higher amounts of SGI. However, for cities with combined sewer systems (e.g. Washington, D.C. and Philadelphia, PA), there was no relationship between the level of combined sewer overflows (CSOs) and the amount of SGI, indicating the level of SGI may not yet be sufficient to reduce CSOs as intended. When comparing individual watersheds over time, increases in SGI show no significant effect on the long-term trends in nutrient loads or hydrologic variables, potentially being obscured by the larger effect of interannual variability.