Microbial Source Tracking of Bacteroidalesand Human Viruses in Stockton, CA

Fecal contamination to a waterbody is a serious environmental problem that affects many coastal and inland waters worldwide. Human and animal fecal pollution impose risks to human health from exposure to waterborne pathogenic bacteria, viruses, and protozoa. PCR- and qPCR-based analyses of Bacteroidales 16S rRNA genes have emerged as a promising tool to identify sources of fecal contamination to a waterbody. Bacteroidales genetic markers, fecal indicator bacteria (FIB) and human Adeno- and Enteroviruses were evaluated from January 2006 to June 2010 in 139 ambient water samples in an urban watershed in the California Central Valley. Total coliforms, fecal coliforms and E. coli were positive for all samples. Overall, FIB exceeded water quality criteria more often during wet events than dry events, though FIB were in exceedence for all events over 50% of the time. Human and cow Bacteroidales markers were detected more often during dry events while dog Bacteroidales were detected more often during wet events. In general, concentrations of universal and human Bacteroidales markers were highest in waterbodies within the more densely populated urban area of the watershed compared to the less populated northern portion. Additionally, human Adenovirus C (AVC) and Adenovirus 40/41 (AV40/41) were detected in locations where land area was dominated by residential and industrial uses, suggesting human exposure to waterbodies in highly urbanized areas present a greater health risk to the public. All samples were qPCR negative for human Enterovirus. Kendall tau-b and Spearman rank correlation analyses were calculated for all Bacteroidales markers and indicator bacteria combinations. In general, when a relationship existed, Bacteroidales markers correlated positively with each other. Universal and human Bacteroidales markers positively correlated with total coliforms only. Cow Bacteroidales negatively correlated with total coliforms, fecal coliforms and E. coli when using Kendall tau-b correlation analysis. A probabilistic statistical model was used to eliminate the effects of false-positive and false-negative information on host-associated Bacteroidales concentrations and used to calculate estimated true predicted host-associated concentrations for human, dog, and cow Bacteroidales markers.