LOCATING POTENTIAL BACTERIAL SOURCES USING A COMPUTER MODELING APPROACH

The Problem Recreational beaches and shellfish beds are often closed due to exceedances of pathogen water quality criteria indicative of potential human health risks. One such example is the closure of shellfish beds in the Southport Beach area of Southport Harbor on Long Island Sound that are impacted by elevated levels of fecal coliform bacteria. However, the location and relative contribution of bacteria sources that result in elevated levels of bacteria over the shellfish beds have not yet been characterized and quantified. Although new bacterial source tracking technologies such as DNA fingerprinting provide a promising solution their use is often limited due to cost and required expertise. An alternative and complementary approach is the use of advanced computer models combining well-designed monitoring programs that hindcast the likely source locations that would affect a specified resource area such as a beach or shellfish bed. Goals and Objectives The project objectives were to:Determine the location of sources of bacteria responsible for closure of recreational shellfish beds in Southport Harbor,Determine the relative contribution from sources in order for state and local officials to focus BMPs where the greatest possible impact will be realized, andDevelop a model application that may be utilized to better understand other shellfishing waters impaired by bacteria.To achieve these objectives a combined field program and modeling effort was undertaken.Field Program ASA designed a field program to collect data necessary to develop and run the models for the summer of 2004. The field program was organized into three components: physical and chemical measurements, bacterial measurements and a dye study.Physical and chemical data were collected continuously from 19 July to 12 August 2004 in Southport Harbor. An acoustic Doppler current profiler (ADCP) was deployed on the bottom to measure the vertical structure of horizontal currents and tide height. Water column profiles of conductivity, temperature and depth were also measured at eleven locations in Southport Harbor during deployment and retrieval of the ADCP.A series of three one-day intensive bacterial surveys (20 and 27 July and 10 August 04) were conducted simultaneously with the deployment of moored instruments and dye studies. In each intensive survey, water samples were collected six times over the tidal cycle from fourteen locations and analyzed for fecal coliform concentrations.Two dye studies were conducted on 20 and 27 July, one for each of two dye release sites. Plume tracking in the harbor was performed by boat with an integrated fluorometer and DGPS system. Modeling For the Southport Harbor project, a modeling system that predicts the circulation and pollutant transport was utilized. A unique requirement was the ability of the model system to estimate source locations and strengths as well as the more typical pollutant distributions from those sources. Toward this end ASA used both the WQMAP hydrodynamic and pollutant modeling system with components from ASA's OILMAP modeling system. WQMAP includes a boundary-fitted, two-dimensional hydrodynamic model that generates water elevation and velocity and a pollutant transport model that generates resulting concentrations from known sources. OILMAP, is a particle-based model system that directly uses the WQMAP hydrodynamic model output to estimate either resulting concentrations from known sources (forward tracking) or probabilities of source locations from known impacted resources (backward tracking). This model was originally developed for use in oil spill tracking but can be similarly used to track other pollutants, i.e. fecal coliforms, as well. Collectively, the integrated model systems identify the location and relative contribution of sources that cause elevated fecal coliform densities over the shellfish beds in Southport Harbor. Summary A modeling and field program was conducted to assess likely sources of elevated fecal coliform levels at Southport Beach in Southport Harbor, Connecticut. The field program consisted of physical measurements (water level elevation and currents), bacterial measurements and dye studies. The hydrodynamic model was successfully calibrated to the field data and showed the complex nature of the tidal flows in Southport Harbor, varying from the primarily east-west flow in Long Island Sound south of the Harbor to the primarily north-south flow in the Mill River to the north. The particle based modeling was run in both forward and backward mode. A series of likely source locations was run in forward mode and plumes from Mill River were more likely to impact the Beach area than those from Sasco Brook. The backward mode results showed that the Mill River was the likely source based on six representative receptor sites located in the Beach area. Control of the Mill River sources is thus more likely to improve water quality at Southport Beach than control of Sasco Brook sources.