Case Study in Pine Creek Watershed, Allegheny County, PA: Physical source tracking using molecular microbial methods

The Pine Creek Watershed is the focus of a large scale bacterial monitoring program sponsored by the Pennsylvania Department of Environmental Protection (PaDEP) and the United States Environmental Protection Agency with participation and support from the Pine Creek Watershed Coalition, Three Rivers Wet Weather Demonstration Project (3RWW), the Allegheny County Sanitary Authority (AlCOSAN), and the Water Quality in Urban Environmental Systems Center at Carnegie Mellon University (Water QUEST).Pine Creek is a tributary to the lower Allegheny River, and drains 67 square miles. Diverse land uses in the watershed and increased land conversion to residential use in the northern watershed provide stressors to the mainstem and its tributaries. Flooding is a significant problem, especially the lower watershed. Water quality impairment for contact recreation is suspect due to inputs from combined sewer overflows and failing septic systems. Other bacterial loads may exist in the watershed, but are not well characterized.While different species may be contributing to bacterial loads (e.g., wild turkey and deer), multiple different sources of human contribution are also suspected (e.g., combined sewer overflows (CSOs), failing septic systems). In the present work, we are developing a novel source tracking method that focuses on identification of physical sources – CSOs and septic fields – rather than focusing on species-specific indicators.Denaturing Gradient Gel Electrophoresis (DGGE) patterns for all bacterial species and for E.coli specific markers for surface water samples from 25 locations in the Pine Creek Watershed were compared to determine the utility of DGGE fingerprinting for source identification. Our results indicate that DGGE is an excellent tool for evaluating bacterial diversity and may have utility for evaluating specific sources of wet weather bacterial loading to watersheds. Additional statistical analyses are needed to determine if this method will be applicable to source identification.