Land use and dog park associations with Escherichia coli in the Chattahoochee River National Recreation Area watershed

A recent study in the Chattahoochee River National Recreation Area (CHAT) indicated that dogs were a primary source of fecal contamination in the Chattahoochee River and that at least some of the contamination in the river was coming from locations outside of CHAT. The study herein sought to determine if dog parks in the CHAT watershed were sources of dog fecal contamination in streams within the watershed. Escherichia coli (E. coli) data were compiled from the Chattahoochee Riverkeeper Neighborhood Water Watch (NWW) program for sites within the CHAT watershed. Information about dog park locations within the Atlanta metropolitan area was compiled through online searches. Wilcoxon rank-sum tests, forward stepwise linear regression, and Spearman rank correlations were used to investigate the relations between seasonal E. coli levels (E. coli concentration and the proportion of samples that exceeded the U.S. Environmental Protection Agency beach action value [BAV]) and dog parks within the drainage basins. NWW sites with dog parks within the drainage basins had higher E. coli concentrations in the summer and winter, and samples exceeded the BAV more frequently in the winter than sites without dog parks within the drainage basins. Escherichia coli levels in the summer and winter were positively correlated with the number of dog parks within the drainage basins, indicating that E. coli concentrations and the frequency of BAV exceedances were seasonally higher at sites with more dog parks than at sites with fewer dog parks within the drainage basins. Escherichia coli concentrations in the summer were negatively correlated with distance to the nearest dog park in the drainage basin, indicating that sites with at least one dog park in close proximity had higher E. coli concentrations in the summer than sites for which the closest dog park was more distantly located. However, results of this study may have been influenced by the high degree of spatial autocorrelation in the data caused by overlapping drainage basins. Additionally, E. coli occurs in the gut systems of many species, so concentrations of E. coli may not represent levels of dog fecal contamination. Dog waste in residential yards and neighborhoods is a possible source of contamination in the watershed that could be investigated in future studies on sources of fecal contamination in the CHAT watershed. Utilizing dog-specific genetic markers in future studies would help reduce ambiguity in data interpretation.