Keeping Beaches Safe… with Statistics!
EPA researchers and partners are creating tools to help beach managers protect public health
For countless Americans, summertime is beach time: millions flock to the nation's coastal and inland beaches during the hot months of June, July, and August to enjoy sun, sand, and surf.
Beaches, however, are parts of complex ecosystems, and conditions in these ecosystems can sometimes lead to hazardous concentrations of certain kinds of bacteria. Typically found in sewage, these bacteria can multiply in the water and pose health threats to swimmers.
That's why beach managers are careful to monitor water quality at their beaches and issue swim advisories or close beaches in the case of dangerous bacterial contamination. The problem is that water sampling and analysis can take over 24 hours, meaning an existing contamination might go undetected for that long.
Working with collaborators in the U.S. Geological Survey (USGS), the National Oceanic and Atmospheric Administration (NOAA), state and local governments, and academia, EPA researchers have come up with a way to solve this problem: Instead of testing water to find indicators of bacterial contamination after the fact, create models that can predict contamination in real-time or even before the contamination occurs.
Faster Laboratory Testing Methods
In addition to Virtual Beach, EPA scientists have also developed a faster laboratory method to test water for bacterial indicators. This method is called quantitative polymerase chain reactions (qPCR), which can cut the lab time needed down from over 48 hours to only a few hours.
In Racine, Wisconsin, Julie Kinzelman and her scientific team are testing Virtual Beach against both bacterial culture methods and qPCR. They've found that qPCR is accurate and fast, but they still hope to use Virtual Beach to cut down on the costs of laboratory analysis.
“The model we developed with Virtual Beach has performed very well,” Kinzelman said. “The ultimate goal is to use the Virtual Beach model to predict bacteria exceedances, and then confirm those results with qPCR, which would create a big cost savings.”
The solution is a software suite called Virtual Beach. EPA scientists released the latest version of the program in March 2012.
"Virtual Beach allows users to develop custom models for their unique beach, which can be used to predict when levels of indicator bacteria in water will signal potentially dangerous contamination," explained Mike Cyterski, an EPA scientist who has worked on the development of the program. Beach managers can easily map their beach using an intuitive graphical interface, and import measurement data on wave height, precipitation, and other factors into the program.
By looking at data on the local watershed, Virtual Beach's models can find correlations between certain weather and water conditions and bacterial outbreaks, giving beach managers advanced warning of the kinds of contamination events that lead to swim advisories and beach closures.
"It's a real step forward in routine water quality monitoring," said Adam Mednick, a research scientist with the Wisconsin Department of Natural Resources, who has worked with EPA's scientists in testing and developing Virtual Beach for the last three years.
"Virtual Beach really improves ease of use," said Mednick. "The method of using predictive models for water quality has existed for a while, but this has made the process of developing the models much more efficient."
Virtual Beach has been successfully tested and used at beaches across the Great Lakes region, from Minnesota to Ontario. Researchers have also used it to successfully predict bacteria levels at marine beaches in Rhode Island, Mississippi, Alabama, Florida, South Carolina, and Puerto Rico.
EPA scientists are working on the next version of Virtual Beach, which will include more statistical methods for data-crunching, as well as a way for users to automatically retrieve data available online from institutions such as NOAA, USGS, and EPA.