Water Resources Research Act Program

Details for Project ID 2013VT72B

Automated Mapping of Effective Impervious Areas (EIA) to Assess the Impact of EIA on Stream Health

Institute: Vermont
Year Established: 2013 Start Date: 2013-03-01 End Date: 2014-02-28
Total Federal Funds: $30,628 Total Non-Federal Funds: $63,312

Principal Investigators: Leslie Morrissey, Beverley Wemple

Project Summary: Although total impervious area (TIA) within a watershed has often been used as an indicator of development and its impact on stormwater runoff, recent studies have demonstrated that the effective impervious area (EIA) within a watershed, i.e. the impervious area that is hydrologically (directly) connected to the receiving stream network or stormwater system, may prove a far better predictor of the impacts of development on stream geomorphology and ecosystem health. The key issue raised by these results, therefore, is how best to protect stream systems in developing watersheds, i.e. by limiting total impervious area or the hydrologic connectivity linking impervious surfaces to receiving waters. Greater understanding of the impact of development on stream ecosystems is needed to address this issue. Unlike TIA, direct measurement of effective impervious areas has proven challenging and thus the number of watershed-scale studies remains extremely limited. We propose to take advantage of recent advances in geospatial technologies and available data to develop and evaluate a methodology to directly measure EIA and then apply that methodology to address how development impacts stream biological health at sub-reach to watershed spatial scales in a range of small watersheds in the Lake Champlain Basin. Our direct approach potentially marks a tremendous improvement over that of past efforts that have relied on coarser scale and less accurate methodologies, site-specific field surveys, or indirect estimates based on analysis of rainfall-runoff data and empirical relationships between TIA and EIA. The increasing availability of high spatial resolution digital imagery and LiDAR data in Vermont, cutting edge image classification techniques, and GIS-based hydrologic modeling, however, now offers the unprecedented opportunity to accurately, reliably, and cost effectively map EIA at sub-reach to watershed spatial scales. Our overall goal is to assess the relationship between EIA and stream biological health as determined by VT DEC within select watersheds in Chittenden County representing a wide range of development. To this end we will develop and evaluate a methodology to map EIA integrating high spatial resolution imagery, object oriented classification techniques, LiDAR-derived elevation data, and hydrologic flow modeling to identify hydrologic connectivity between impervious areas and nearby streams and stormwater conveyance systems. This effort will build on existing data, expertise, and our exploratory efforts to map EIA within an expert and automated classification system. Validation of resultant EIA maps will be made in comparison to independent image photointerpretation and VT DEC field survey data. Watershed-scale estimates of EIA, in addition, will be compared to estimates derived from rainfall-runoff data and USGS empirically defined TIA coefficients. We will then assess the relationship between our measures of EIA and VT DEC measures of biological stream health. Success in this effort will allow regulatory agencies to effectively target specific sub-watersheds where a reduction in hydrologic connectivity between impervious surfaces (i.e. EIA), rather than in TIA, might best minimize ecological and hydrological impacts to streams.