AWWA WQTC64089

This study was undertaken to determine the effect of ultraviolet (UV) light in combination with chlorine based disinfection on the growth of bacteria in a simulated warm climate distribution system. The focus of this study was to determine the effectiveness of Cl₂, NH₂Cl, UV/Cl₂, and UV/NH₂Cl in suppressing biofilm growth and coliform occurrences in drinking water distribution systems through a field-scale study, and to compare treatment with UV to no UV treatment. Heterotrophic plate counts and coliforms were used to compare disinfection treatments for bacteria within the systems. In addition, water quality parameters were monitored throughout the experiments. Keller Water Treatment Plant is located in Pinellas County, Florida. The source for these experiments was groundwater treated for hydrogen sulphide removal, corrosion control, pH adjustment, and addition of fluoride. The water is disinfected using free chlorine followed by addition of ammonia at a ratio of 4.5:1 to produce monochloramine. Annular reactors containing coupons made of PVC material were used to simulate the distribution system in Pinellas County. Two ARs acted as controls and received treated water containing free chlorine or chloramines. Two more ARs received water with either Cl₂ or NH₂Cl and were additionally treated with UV light. For these experiments the water sources were chemically disinfected prior to exposure to UV at a 100 mJ/cm² dose, whereas in practice, UV disinfection would normally occur prior to chlorination. All non-opaque exposed surfaces of the ARs were covered to reduce the potential of phototrophic growth in the field systems. The study ran over a seven-month period from April to November, 2005. A general schematic of the annular reactor set-up is presented. Two streams, one containing monochloramine and one with free chlorine, were the primary source waters for the model distribution systems. The water collected from each stream was directed into a separate clearwell. The flow pumped from each of these clearwells was split to feed both a UV unit and one AR. Each water stream had a separate low pressure UV lamp (TrojanUV Max Model C) provided by Trojan Technologies. Once water streams passed through the UV treatment they were directed towards two additional separate clearwells which fed the remaining two ARs. During the study, the ARs were monitored once a week for heterotrophic bacteria counts (suspended and biofilm), disinfectant residual and coliforms, as well as other unreported water quality parameters. Bulk samples for HPC bacteria were collected in 100mL IDEXX bottles containing 10% w/v sodium thiosulfate to quench disinfectant residual. The PVC coupons were removed aseptically and placed in sterile 50mL glass containing PBS and 0.1% w/v sodium thiosulfate. Both samples were shipped overnight to Dalhousie University laboratory where the attached cells on coupons were immediately removed by the scraping method as described by Gagnon and Slawson (1999). The scrapings were vortexed and plated on R2A agar to determine heterotrophic plate counts. New coupons that were treated with ethanol were used to replace those that had been removed and shipped from each AR. The process for enumeration of HPC bacteria involved a standard spread plate technique as described in Standard Methods for the Examination of Water and Wastewater (21st edition) on R2A agar (Difco Laboratories). Coliforms were enumerated using the IDEXX Colilert® Quanti-tray® system. Commercially available sterile bottles containing sodium thiosulfate were used to collect the 100-mL samples. Statistical tests were performed and repeated for the various combinations of disinfectants. In addition, statistical tests compared the significant differences between the average influent and effluent values for the w