Spatial and hydrologic variation of Bacteroidales, adenovirus and enterovirus in a semi-arid, wastewater effluent-impacted watershed

Abstract

Bacteroidales and viruses were contemporaneously measured during dry and wet weather conditions at a watershed-scale in a semi-arid watershed impacted by a mixture of agricultural runoff, municipal wastewater effluent and municipal runoff. The results highlight the presence of municipal wastewater effluent as a confounding factor for microbial source tracking (MST) studies, and thus data were segregated into groups based on whether they were impacted by wastewater effluent. In semi-arid environments such as the Calleguas Creek watershed, located in southern California, the relative contribution of municipal wastewater effluent is dependent on hydrology as storm events lead to conditions where agricultural and municipal stormwater dominate receiving waters (rather than municipal wastewater, which is the case during dry weather). As such, the approach to data segregation was dependent on hydrology/storm conditions. Storm events led to significant increases in ruminant- and dog-associated Bacteroidales concentrations, indicating that overland transport connects strong non-human fecal sources with surface waters. Because the dataset had a large number of non-detect samples, data handling included the Kaplan–Meir estimator and data were presented graphically in a manner that reflects the potential effect of detection limits. In surface water samples with virus detections, Escherichia coli concentrations were often below (in compliance with) the recreational water quality criteria. In fact, sites downstream of direct inputs of municipal wastewater effluent exhibited the lowest concentrations of E. coli, but the highest concentrations of human-associated Bacteroidales and highest detection rates of human viruses. The toolkit, comprised of the four Bacteroidales assays and human virus assays used, can be successfully applied to inform watershed managers seeking to comply with recreational water quality criteria. However, care should be taken when analyzing data to account for the effect of non-detect samples, sources with differing microbial viability, and diverging hydrologic conditions.