You are here:
Predictive models for Escherichia coli concentrations at inland lake beaches and relationship of model variables to pathogen detection
Citation:
Francy, D., E. Stelzer, J. Duris, A. Brady, J. Harrison, H. Johnson, AND M. Ware. Predictive models for Escherichia coli concentrations at inland lake beaches and relationship of model variables to pathogen detection. APPLIED AND ENVIRONMENTAL MICROBIOLOGY. American Society for Microbiology, Washington, DC, 79(5):1676-1688, (2013).
Impact/Purpose:
This study evaluates if predictive models can be used to provide real-time assessment at inland lakes.
Description:
Methods are needed improve the timeliness and accuracy of recreational water‐quality assessments. Traditional culture methods require 18–24 h to obtain results and may not reflect current conditions. Predictive models, based on environmental and water quality variables, have been used to estimate water quality, but their effectiveness has not been well studied in inland recreational waters. Results from sampling over three recreational seasons from eight inland recreational lakes in Ohio were used to investigate the use of predictive models for E. coli and the links between E. coli concentrations, environmental and water‐quality variables, and pathogens. Based upon sampling results from 21 beaches, models were developed for 13 inland lake beaches, and the most predictive variables were rainfall, wind direction and speed, turbidity, and water temperature. Models were validated at nine beaches during an independent year. At three of the nine beaches, the use of a model resulted in an increase in correct responses, sensitivities, and specificities as compared to using the previous day’s E. coli concentration (the current method). At one beach, model performance was not improved over that of the current method and, at the remaining five beaches, there were too few exceedances during the validation year to adequately assess model performance. At five sites, a subset of samples were analyzed for bacterial,protozoan, and viral pathogens. Cryptosporidium, adenovirus, eaeA‐E. coli, ipaH‐Shigella, and spVC‐Salmonella were found in at least 20% of samples collected for pathogens. The presence orabsence of the three bacterial genes were related to some of the model variables, but were notconsistently related to E. coli concentrations.
