Modeling of Water Age in the Drinking Water Distribution System of Trondheim Kommune

Abstract: Drinking water provision is fundamental to human activities, and has major direct and indirect effects on society, economy and the environment. Drinking water quality is affected by raw water quality, as well as processes and incidents occurring in treatment and distribution. Using digital water network models, a wide range of investigations can be made to a drinking water distribution system (DWDS), with no risk for water users and no disruption of system operation. Water age is considered as an indicator that captures all system-specific degradation of water quality, and it is also one of the most easily derived results from water network model simulations. The main purpose of this Master's thesis was to provide estimates of the water age distribution in Trondheim kommune. This was done using EPANET, Matlab, and the water network model (WNM) of Trondheim kommune. Furthermore, the accuracy of model estimates has been assessed by conducting a chemical tracer study using sodium chloride (NaCl). Microbial water sample data was compared with mean water age values from a 120-day model simulation, and a statistical test was used to study the effect of site and season on heterotrophic plate count (HPC) data. Results from the tracer study showed a large difference between simulated and measured tracer peak arrival times. Statistical tests indicated a strong positive correlation (Pearson's R = 0:78) between average water age and HPC counts, and that these counts are significantly affected by site and season. A clear implication, based on results from the tracer study, is that the WNM of Trondheim kommune requires further calibration if it is intended for use in future water quality modeling