| Drinking water distribution systems are vulnerable to biological contamination via numerous potential intrusion locations; thus, a comprehensive water quality indicator is critical for detection of contamination events and protection of these systems. System vulnerability is examined using a water distribution system simulation model, EPANET, to predict chlorine residual profiles for a representative water distribution system. The analysis focuses on the identification of vulnerable areas where there is a high probability that the disinfectant residual will be below the required threshold. Public utilities can use this method to determine high vulnerability locations and ideal sensor placement sites.;Next, the seasonal variability of bacterial diversity in a free-chlorinated drinking water distribution system was examined using 16S rRNA gene clone library analysis. In four seasonal samples and one event sample (collected following a major line break), the majority of distribution system bacteria belonged to the Alphaproteobacteria, Betaproteobacteria, and Gammaproteobacteria classes. Winter samples showed greater bacterial diversity and the event sample showed significant changes from baseline samples with higher numbers of gammaproteobacteria. Bacterial populations in this distribution system appear to exhibit stable populations of predominately Proteobacteria, with seasonal changes in the relative distribution of classes due to changes in chlorine dosing.;The final study sought to confirm that bacterial diversity markers, specifically, the Alphaproteobacteria, Betaproteobacteria, and Gammaproteobacteria classes, could be used as an indicator of biological contamination in drinking water distribution systems. These bacterial groups were targeted using Quantitative Polymerase Chain Reaction (QPCR) to detect surrogate contaminants: Escherichia coli 01:K1:H7, Staphylococcus epidermidis, Mycobacterium aurum, Bacillus cereus, and Pseudomonas fluorescens, in bench-scale reactors. Baseline and contaminated water samples were analyzed using the QPCR-based method and the standard total coliforms and E. coli test method. Additions of surrogate bacteria yielding final concentrations of 103–104 CFU/ml in contaminated samples were detected in all five test cases contrasted against a single detection for E. coli by the current regulatory method. These preliminary experiments suggest that this novel technique could be applied in drinking water sensor technology to yield a faster and more comprehensive indicator than current water quality methods. |
