Microbial Source Tracking Using General and Host-Associated Bacteroidales 16S rRNA-Based Quantitative PCR Assays and Correlation with Environmental Parameters at an Urban and Rural Watershed Within The Edwards Aquifer

Identifying the sources and spatio-temporal trends of fecal pollution is critical to protecting water quality and implementing adequate pollution control and prevention strategies to mitigate pollution levels and risk to environmental and human health. Karst aquifers, such as the Edwards Aquifer in Bexar County, are known to be vulnerable to human and animal waste contamination, and presently, there is an incomplete understanding of the sources of pathogenic fecal contamination in the Edwards, a source of drinking water to more than 2 million people in south-central Texas. Conventional fecal indicator bacteria (FIB) are unreliable due to widely varying survival rates in the environment, failure to discriminate between sources, and inability to distinguish between fecal bacteria associated with recent contamination events and those adapted to secondary habitats.

The purpose of this study was to apply microbial source tracking (MST) techniques to evaluate fecal pollution at both an impaired urban and rural creek within the recharge and contributing zones of the Aquifer. MST can identify and quantify the source and species of fecal bacteria by targeting Bacteroidales 16S rRNA genes using polymerase chain reaction techniques. Bacteroidales do not have the ambiguous survival and sourcing issues present with conventional FIBs and can provide good spatio-temporal resolution. Quantitative PCR-based methods for general assays (BacUni, E. coli and Entero1) and five host-associated assays (HF183, BacHum, Chicken/Duck-Bac, BacCan and BacCow) were applied to resolve spatio-temporal patterns, seasonal variability, and to identify and quantify the sources of fecal contamination. In addition, a total of five storm event-related samples, water temperature, dissolved oxygen (DO), pH, dissolved nitrogen and rainfall data were collected. The effects of stormwater on marker levels and correlations among the markers and with environmental parameters were also investigated.

Water samples were collected bi-monthly from four sites on Balcones Creek and four sites on Leon Creek from January 2018 to February 2019. Levels for the three general markers were highest and exhibited a similar spatial and temporal distribution pattern across the sampling sites, and thus, it was determined that the sole use of these markers was not sufficient for specific fecal source identification in the Edwards. Among the host-associated markers, highest concentrations were observed for BacCan in the urban area and BacCow in the rural area. Additionally, Chicken/Duck-Bac, BacCan and BacCow all exhibited higher concentrations during the spring season and the end of fall/early winter. Relatively low concentrations were observed for the human-associated markers, however, levels were higher in the urban area and highest following rainfall seven days prior to sample collection, indicating that rainfall is an important factor in human fecal material loading in both rural and urban settings. Additionally, relatively higher levels in HF183 and BacHum were observed at sites located near increases in septic tank and SAWS/FAR sewer main densities and may be attributed to leaks or breaks in these infrastructures. Concentrations were highest for the stormwater samples, signifying that peak fecal concentration loadings occurred during storm-related events.