Viral, bacterial, and protozoan pathogens and fecal markers in wells supplying groundwater to public water systems in Minnesota, USA

Authors: STOKDYK, JOEL - Us Geological Survey (USGS); FIRNSTAHL, AARON - Us Geological Survey (USGS); WALSH, JAMES - Minnesota Department Of Health; Spencer, Susan; DE LAMBERT, JANE - Minnesota Department Of Health; ANDERSON, ANITA - Minnesota Department Of Health; REZANIA, LIH-IN - Minnesota Department Of Health; KIEKE, BURNEY - Marshfield Clinic Research; Borchardt, Mark

Submitted to: Water Research
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 4/8/2020
Publication Date: 4/12/2020
Citation: Stokdyk, J.P., Firnstahl, A.D., Walsh, J.F., Spencer, S.K., De Lambert, J.R., Anderson, A.C., Rezania, L.W., Kieke, B.A., Borchardt, M.A. 2020. Viral, bacterial, and protozoan pathogens and fecal markers in wells supplying groundwater to public water systems in Minnesota, USA. Water Research. 178(1):115814. https://doi.org/10.1016/j.watres.2020.115814.
DOI: https://doi.org/10.1016/j.watres.2020.115814

Interpretive Summary: Groundwater is the source of drinking water for many municipalities and rural public places such as schools, restaurants, and motels that have their own wells. These public water systems test the quality of their water using simple tests for bacteria called indicators. The indicators give only a partial picture of water quality as it is a single test and the indicators themselves do not cause illness. To fully characterize groundwater quality in Minnesota we took 964 samples from 145 wells supplying public water systems; the samples were analyzed for a variety of waterborne pathogens and microorganisms found in fecal wastes. Fecal contamination was found in 58% of samples and 96% of wells. We evaluated performance of the indicators for identifying pathogens in well water and learned the indicators were good at predicting pathogen absence but not presence. Some scientists use relationships between indicator and pathogen concentrations in sewage and extrapolate these to surface water to estimate health risk from activities such as swimming. However, we showed such an approach would not work for estimating illnesses from drinking fecal-contaminated groundwater. Compared to simple indicator tests, our approach provides a comprehensive picture of groundwater cleanliness, however, it is still difficult to interpret the data in terms of health risk.

Technical Abstract: Drinking water supply wells can be contaminated by a broad range of waterborne pathogens. However, groundwater assessments frequently measure microbial indicators or a single pathogen type, which provides a limited characterization of potential health risk. This study assessed contamination of wells by testing for viral, bacterial, and protozoan pathogens and fecal markers. Wells supplying groundwater to community and noncommunity public water systems in Minnesota, USA (n = 145) were repeatedly sampled over one or two years and tested using 23 qPCR assays. Eighteen genetic targets were detected at least once, and microbiological contamination was widespread (96% of wells, 58% of samples). The sewage-associated microbial indicators HF183 and pepper mild mottle virus were detected frequently. Pathogens were detected in 70% of wells and 21% of samples by qPCR, with Salmonella and Cryptosporidium detected more often than viruses. Samples positive by qPCR for adenovirus (HAdV), enterovirus, or Salmonella were analyzed by culture and for genotype or serotype. qPCR-positive Giardia and Cryptosporidium samples were analyzed by immunofluorescent assay (IFA), and IFA and qPCR concentrations were correlated. Comparisons of indicator and pathogen occurrence at the time of sampling showed that coliforms, HF183, and Bacteroidales-like HumM2 had high specificity and negative predictive values but generally low sensitivity and positive predictive values. Pathogen-HF183 ratios in sewage have been used to estimate health risks from HF183 concentrations in surface water, but in our groundwater samples Cryptosporidium oocyst:HF183 and HAdV:HF183 ratios were approximately 10,000 times higher than ratios reported for sewage. qPCR measurements provided a robust characterization of microbiological water quality, but interpretation of qPCR data in a regulatory context is challenging because few studies link qPCR measurements to health risk.