Baltimore Ecosystem Study Cary Institute of Ecosystem Studies

2012 BES Annual Meeting Presentation and Poster Abstracts

Storm-event hydrologic response across a spectrum of development patterns and watershed types among Gwynns Falls tributaries
Miller, Andrew
Co-Authors: Andrew J. Miller (1), Garth A. Lindner (1), Sierra Z. Shamer (1), Kevin M. Schmidt (2), Mark J. Kather (3), Daniel Jones (1), Matthew E. Baker (1), Claire Welty (1) : (1) - UMBC (2) - University of Delaware (3) - Baltimore County Public Schools

Abstract: We analyze characteristic watershed hydrologic response to pulse rainfall events for a subset of 15 current and formerly active stream gages along the Gwynns Falls and its tributaries. We examine hydrologic behavior at nested scales of 1-2 km2, 5-6 km2, and 14 km2 within the Dead Run watershed, and comparisons among five watersheds (Dead Run, Maidens Choice, Powder Mill Run, Gwynns Falls at Delight and Scotts Level) at scales of 9-14 km2. Streamflow records were compared with HydroNEXRAD radar rainfall data (1-km2 grid scale and 15-minute time intervals) and BES rain gage records to derive characteristic unit hydrographs for pulse events. Study watersheds span a range of development ages and patterns, with impervious cover varying between 28 and 64 percent and percent area draining to stormwater detention facilities varying between <1% and 48%. We examine network topology using an augmented drainage network that combines channels with impervious areas to derive a binary hydrography GIS layer for use in deriving width functions, or probability density functions for flowpath distance upstream of the watershed outlet. Simple assumptions about the ratio of channel:hillslope flow velocity are applied to develop idealized representations of the effect of watershed shape, drainage pattern and urban development on watershed hydrologic response for comparison with unit hydrographs. Preliminary results suggest this approach may have considerable explanatory power and also highlight where it fails to capture some aspects of characteristic behavior. Thus far we do not see clear evidence that stormwater management has a controlling influence on storm hydrologic response.