Stable spatial patterns of concentrations of antibiotic resistant bacteria in irrigation water

Abstract Irrigation waters serve as vectors for the spread of antibiotic-resistant bacteria which presents a worldwide health concern. Determining the concentrations of those bacteria in waters used for irrigation is therefore an important monitoring target. The objective of this work was to determine the existence of persistent spatial patterns in antibiotic-resistant bacteria in an irrigation waterbody and compare the patterns with patterns of water quality parameters. Escherichia coli , total coliforms, and total heterotrophic bacteria concentrations were measured in water samples collected from 20 sites across an irrigation waterbody on nine sampling dates. The Empirical Orthogonal Function (EOF) analysis was used to detect persistent spatial patterns of tetracycline- and cefotaxime-resistant bacteria as well as water quality parameters. The average concentrations of both generic and antibiotic-resistant bacteria were an order of magnitude higher in bank sampling sites than interior sites. From the EOF analysis results, sampling locations were identified which could provide the best estimates of the average concentrations for each bacterial group. Patterns in temperature (r s = -0.638 to -0.749), dissolved oxygen (r s = -0.173 to -0.675), and pH (r s = -0.483 to -0.770) showed the strongest negative correlations with patterns in bacteria concentrations while all other water quality patterns tended to be positively correlated. Results of this work show that the EOF analysis can be a powerful tool in the design of effective monitoring programs which assess antibiotic resistance in irrigation waters by identifying locations which are closest to the spatiotemporal average.