Detecting Illicit Connection to Stormwater: A Primer on Statistics and Fecal Indicators

As Clean Water Act programs have been increasingly effective for point source control, efforts to detect and eliminate sources of pollution from non-point and stormwater sources must increase. These sources have become the predominant cause of non-attainment of water quality criteria. In particular, recreational use criteria are often impaired by bacteria either from background sources or exfiltration from aging sewer infrastructure. Efforts to identify and control these sources are limited by the statistical power of data collection efforts and are compounded by the use of indicator species absent a cost effective method for true pathogen detection.Our study examined the effectiveness of sampling for pathogen indicators and the statistical methods required to measure illicit connection detection and removal in an urban stormwater collection system. The project is funded through the settlement of a State lawsuit against the municipality so the names of the City and the State are being kept confidential. Our staff had worked with the City and the State to develop and implement a sampling program to identify illicit connections, and define and target basins for other investigative work such as smoke and dye testing. Upon detection, illicit connections were removed or sewers were repaired as required by the type of problem identified.The goal of the study was to refine the sampling program to provide a more quantitative measure of the success of illicit connection removal. More intensive sample replication was proposed and tested as a method of statistically defining improvements to the system. Collection of samples for DNA analysis was also recommended. These and other recommendations were implemented in the next phase of the program and better definition of the success of illicit connection removal was achieved. The role of background sources of bacteria was also better defined both from upstream and from in system wildlife sources.