Estimation, distribution, and development of a surrogate model for Escherichia coli in the New River, New River Gorge National Park and Preserve, West Virginia, 2021–23

Abstract

The New River Gorge National Park and Preserve in West Virginia receives more than 1 million visitors each year, many of whom come to enjoy the New River, which is known for its whitewater recreation. However, most of the tributaries within the New River Gorge are impaired by fecal-coliform bacteria, which are at concentrations that may exceed recreational-contact standards, posing a potential health risk to the public and, therefore, creating a need to better understand the spatial and temporal distribution of fecal-coliform bacteria and to communicate this information to park visitors. Concentrations of Escherichia coli , a species of fecal-coliform bacteria, were monitored in the New River and selected tributaries from October 2021 through September 2023, with emphasis placed on the primary recreational-contact season from May through October. Composite and cross-sectional water samples were taken from three U.S. Geological Survey (USGS) monitoring locations: the New River at Highway 41 at Prince, West Virginia (USGS 03184905), New River at Thurmond, West Virginia (USGS 03185400; hereafter, Thurmond), and New River at Fayette, West Virginia (USGS 03186000). Periodic longitudinal transects included water samples collected below seven major tributaries of the New River within the gorge. Water-quality parameters, including water temperature, pH, specific conductance, dissolved oxygen, and turbidity, were recorded with each E. coli water sample. During the 2 years of sampling, E. coli concentrations in samples collected from the New River ranged from less than 1 to 1,100 most probable number (MPN) per 100 milliliters (MPN/100 mL). The recreational-contact standard, which is based on the U.S. Environmental Protection Agency 90th-percentile statistical threshold value for E. coli concentrations (320 MPN/100 mL), was exceeded in 11 of the 110 samples collected from the New River during this study. Water-quality parameter measurements and E. coli concentrations in collected samples were generally consistent among USGS monitoring locations throughout the New River Gorge; however, storm events created notable exceptions because they increased tributary streamflow and E. coli concentrations in samples, particularly at the New River below Piney Creek at McCreery, West Virginia (USGS 03185208), and New River Below Arbuckle Creek at Thurmond, West Virginia (USGS 03185440), monitoring locations. Escherichia coli concentrations of cross-sectional samples tended to be consistent across the New River, except for a few nearshore samples. Sample E. coli concentrations and corresponding measurements of continuous water-quality parameters, streamflow, and precipitation data from Thurmond and the Piney Creek at Raleigh, West Virginia (USGS 03185000; tributary to the New River) monitoring locations were evaluated for use in a near-real-time E. coli surrogate model. The antecedent mean 24-hour turbidity at Thurmond was selected as the best variable for a simple linear regression surrogate model for the log 10 E. coli concentration in the New River and had an adjusted coefficient of determination of 0.556 and a p-value of less than 0.001. The regression equation surrogate model suggests that the recreational-contact standard is exceeded when the antecedent mean 24-hour turbidity at Thurmond is 23.6 formazin nephelometric units or higher (with a 95-percent confidence interval of 19.4–30.7 formazin nephelometric units). Evaluated against a turbidity duration curve, this standard is exceeded 7.5 percent of the time at Thurmond. This surrogate model could help New River Gorge National Park and Preserve staff provide near-real-time information about E. coli concentrations and related recreational-contact risks to the public.