Evaluating associations between environmental variables and Escherichia coli levels for predictive modeling at Pawtuckaway Beach in Nottingham, New Hampshire, from 2015 to 2017

From 2015 through 2017, the U.S. Geological Survey in cooperation with the New Hampshire Department of Health and Human Services and the New Hampshire Department of Environmental Services studied occurrences of high levels of Escherichia coli (E. coli) bacteria at the Pawtuckaway State Park Beach in Nottingham, New Hampshire. Historic data collected by the New Hampshire Department of Environmental Services indicated that E. coli concentrations in the water typically increased through the beach season to levels considered potentially harmful to beachgoers. During the three beach seasons that were studied, E. coli samples were collected three to four times per week, and water-quality and meteorological data were collected continuously. The Virtual Beach software was used to generate a predictive model for each year of the study (2015–2017), and the model for each of these years was tested with data from the other two. Additionally, data from all study years were combined to generate a comprehensive model to help identify independent variables that might characterize environmental conditions relative to E. coli levels during multiple seasons. The accuracy of the models in predicting the occurrence of high E. coli levels was marginal, but the models did provide insights into the likely mechanisms for increased E. coli levels during the seasons. Variables most important in explaining high E. coli levels were the presence of geese at the beach, the progression of the season, the number of visitors at the beach, and wind vectors relative to beach orientation.