The residence time of water in the distribution system (DS) is an important factor in determination of the extent of reactions. For example, disinfectant residual decreases and disinfection byproducts increase with increasing residence time. Loss of disinfectant residual can lead to serious bacterial regrowth problems. While chemical tracer studies are becoming more familiar to water utility personnel, a methodology has not been adopted for quantitative analysis of the resulting data from the DS. The alternative is water age calculations from hydraulic modeling but this requires development of the network model and its calibration. Moreover, the extent of skeletonization of the network may limit its utility for estimating water age in smaller pipes that are not included in the model. This study presents a method to calculate a mean constituent residence time (MCRT); the constituent is a conservative chemical tracer. The method is applied in the DS of Durham, North Carolina where the analysis was made more complex by inclusion of two water treatment plants (WTPs). A unique tracer study was developed todetermine simultaneously the percent contribution to flow from each WTP at each sampling station and the flowrate weighted average of the MCRT. This involved a switch in coagulants at one WTP and a change in fluoride feed concentration to the other. Includes 4 references, tables, figures.
Keywords: Distribution Systems;Tracers;Treatment Plants;Flow;Durham, North Carolina
