You are here:
BACTERIA, BEACHES AND SWIMMABLE WATERS: INTRODUCING VIRTUAL BEACH
Citation:
FRICK, W. E. BACTERIA, BEACHES AND SWIMMABLE WATERS: INTRODUCING VIRTUAL BEACH. Presented at 4th International Conference on Marine Waste Water Discharges and Coastal Environment, Antalya, TURKEY, November 06 - 10, 2006.
Impact/Purpose:
A main objective of this task is to combine empirical and physical mechanisms in a model, known as Visual Beach, that
- is user-friendly
- includes point and non-point sources of contamination
- includes the latest bacterial decay mechanisms
- incorporates real-time and web-based ambient and atmospheric and aquatic conditions
- and has a predictive capability of up to three days to help avert potential beach closures.
Description:
Safe beaches meet water quality standards and are valued for their aesthetics and the recreational opportunities that they afford. In the United States recreational water quality assessments and beach closure decisions are presently based on samples of enterococci or Escherichia coli that typically take a day or more to culture and develop. Thus decisions to close beaches are generally based on old information that, due to the great daily variability in bacteria concentrations, leads to an unsatisfactory number of false positive and negative outcomes. Beaches are closed when they could be open and open when they should be closed, no one knowing their true status until the next day. Researchers have shown that various techniques, including empirical statistical models, can provide better results. To make such models generally available to the public, the EPA Office of Research and Development, in collaboration with the USGS, NOAA, and others, is developing the Virtual Beach model platform. Virtual Beach is designed to provide the public tools for developing site-specific predictive models for its beaches. Virtual Beach features both the multiple linear regression (MLR) and time-series forecasting techniques to relate indicator bacteria concentration to predictor variables. It emphasizes the value of data inspection, transformations, interaction terms, adjustment for time-series effects, identification of outliers, and model selection criteria. R2 and Cp statistics are joint criteria in the model selection process. The value of dynamic modeling, that explicitly recognizes the fact that beach conditions often reflect longer-term dynamic variations, is investigated and documented. Virtual Beach was tested with the Huntington Beach, Ohio 2006 summer bacteria data that was disseminated daily on the internet. The findings support the conclusion that readily available environmental data may be effective, where such data are available, in nowcasting and forecasting bacteria concentrations at locations around the world.