AWWA WQTC54999

The Goleta Water District wanted to upgrade and optimize its treatment plant and considered several alternative disinfection practices. In order to help choose treatment options that would remain in place for over a decade, the factors involved in DBP formation both within the treatment plant and in the distribution system were studied to identify trends. TTHM and HAA5 samples were taken at different steps in the treatment process at the same time of quarterly compliance monitoring. Data was examined and plotted against temperature, chlorine dose, plant flow, and other criteria to identify which factors were most strongly correlated with TTHM and HAA5 production. In addition, some plant and distribution system data were separated into pre-GAC and post-GAC data sets to see the effect GAC filtration had on various correlation factors. The proportion of total TTHM and HAA5 produced at each step in the treatment process was also evaluated. In general, the strongest correlations seen in treatment plant samples were with TOC, UV254, and chlorine residual in the previous sample location. For the distribution system samples, the strongest correlations were with TOC and UV254. THMs gave more correlations than HAAs did. Some interesting behavior was seen with brominated DBPs. Chlorine dose, temperature, and pH were found to have little effect on the formation of THMs and HAAs. Distribution system samples correlated most strongly with TOC and UV254. Overall, approximately 30% of the total TTHMs produced as measured in the distribution system are produced at the pre-chlorination step in the treatment plant (this does not account for any loss during residence in the plant and subsequent formation later). Another 20-40 % is produced prior to filtration. HAA5 treatment plant numbers were more varied, with sometimes lower numbers in later treatment steps, indicating that HAAs are more susceptible than TTHMs to destruction or microbial degradation during residence in the plant. Includes 4 references, figures.