2015 GSA Annual Meeting in Baltimore, Maryland, USA (1-4 November 2015)

Paper No. 276-7
Presentation Time: 9:45 AM


MILES, Brian C., Institute for the Environment, University of North Carolina at Chapel Hill, 100 Europa Drive, Suite 490, Campus Box 1105, Chapel Hill, NC 27517 and BAND, Lawrence E., University of North Carolina, CB#6116, Chapel Hill, NC 27599-3220, brian_miles@unc.edu

Increases in stormwater runoff due to urbanization are driven by the presence of impervious surfaces. The level of stormwater runoff reduction required to meet contemporary goals of restoring pre-development hydrology may require retrofitting existing private residential areas where rooftops and driveways are a large component of impervious land cover. Rooftop disconnection is one approach to reducing stormwater volumes and peak flows by rerouting roof runoff from impervious to pervious surfaces (e.g. lawns, raingardens). Using an Urban Variable Source Area (UVSA) approach, we explore how residential rooftop hydrologic connectivity, as well as how the location of rooftop connectivity in a watershed, affects stormwater volume in two small medium-density urbanized headwater watersheds in Baltimore, Maryland and Durham, North Carolina. We hypothesize that decreasing residential rooftop connectivity will decrease daily peak flow while increasing baseflow, and that rooftop connectivity closer to streams will result in smaller storm flows at the watershed outlet than connectivity farther away. Simulations with the Regional Hydro-Ecological Simulation System (RHESSys) suggest that disconnecting all single-family residential rooftops from adjacent impervious surfaces will result in small decreases in daily peak flows and increased baseflow. This result was only statistically significant for the Durham, NC study watershed, which we hypothesize is due to this watershed’s greater density, more uniform distribution of single-family development, greater soil water storage capacity, and atmospheric evaporative demand. However, we did not discern a difference in streamflow response based on where (near-to vs. far-from stream) residential rooftops were connected or disconnected. We also identified a counter-intuitive result where rooftop downspout disconnection, a putative infiltration-based stormwater control measure, yielded decreased infiltration. This unintended consequence highlights the need for distributed watershed-scale ecosystem-based modeling approaches to stormwater management that include linkages between variable source area dynamics and the surface- subsurface flow processes in urbanized ecosystems.