Unveiling Origins and Dynamics of Fecal Indicator Bacteria in an Urban Creek

Abstract

Urban waterways are highly vulnerable to bacterial contamination, which presents significant risks to public health and water quality. Common methodologies typically measure the total concentration of fecal indicator bacteria (FIB) but are unable to address the complex sources of contamination contributing to the overall bacterial load. This study established chip-based digital polymerase chain reaction (cdPCR) techniques for microbial source tracking to unveil the origins of Escherichia coli (E. coli). Along with a simultaneous analysis of physicochemical water quality indicators, an assessment was conducted using host-associated genetic markers that indicate fecal sources from humans (HF183/BacR287), ruminants (Rum2Bac), dogs (DG3), and birds (GFD) in the lower portion of Rock Creek River (RCR) in the District of Columbia, United States. Stream samples were collected twice a month (n = 24) and after rain events (n = 6) from three sites along the RCR in the district area that feature a mix of highly developed urban areas and park surface regions. Approximately 50% of the stream samples (n = 96) were found to be impaired, exceeding the district's single sample maximum assessment level (410 MPN/100 ml) for E. coli. Herein, we adopted a multi-scale characterization of the relationship of cultural E. coli with host-associated markers, the regression with in-stream physiochemical constituents, the distinction between sampling sites, and the correlation with sizeable land cover categories. In Chapter 1, a comprehensive overview of MST methods is presented. This chapter summarizes the development of MST, categorizes common MST techniques into library-dependent versus library-independent and culture-dependent versus culture-independent groups, and provides a brief history of the advancements in molecular instrumentation used for culture-independent methods. In Chapter 2, consistently elevated E. coli levels were observed at all sites during wet weather, highlighting the substantial impact of storm runoff on water quality deterioration. Among the four molecular markers tested, HF183/BacR287, which indicates human-associated contamination, was particularly prevalent, with the highest frequency found in one of the tributaries. The second marker, derived from avian sources (GFD), showed a moderate to low frequency across the sites. Detection of the ruminant- and dog-specific markers was sporadic at all three sites. Correlation and regression analyses involving E. coli, molecular markers, and physicochemical constituents revealed significant statistical relationships. Notably, turbidity and flow were useful indicators for quickly assessing bacterial contamination. These findings emphasize the importance of reducing microbial contributions from runoff in watershed areas to urban streams during wet weather. The methods and findings of this study are expected to assist stormwater management and regulatory agencies in developing best management practices (BMP) to protect the water quality of urban streams. In Chapter 3, a strong association of E. coli with low-intensity developed land was established, but this association to forested areas at smaller spatial scales. The HF183/BacR287 marker exhibited similar trends, reinforcing its role as a reliable indicator of E. coli contamination sources. This study highlights the value of MST markers in identifying sources of microbial contamination. It provides important insights for managing water quality across various land cover types and changing weather conditions. In Chapter 4, the scalability of cdPCR to cell equivalents was investigated. By transforming scaled cdPCR DNA copies, the study revealed that 3,153 DNA copies per 100 mL of human-associated HF183BacR287 corresponded to the same regulatory threshold as cultured E. coli, enabling direct comparison between cdPCR and Colilert methods for contamination detection. This approach highlights the potential of cdPCR as a complementary tool to traditional methods in MST studies, offering a more detailed and efficient approach for water quality monitoring and management. In Chapter 5, a summary of the results is presented, and a perspective of future research direction is proposed.