Recently many water distribution systems (WDS) developed and calibrated extended period simulation model to assist in the Initial Distribution System Evaluation (IDSE). The IDSE is the first phase of the Stage 2 Disinfectants and Disinfection Byproducts (D/DBP) Rule promulgated by the United States Environmental Protection Agency (USEPA). The purpose of the IDSE was to identify locations within the WDS for Stage 2 D/DBP compliance monitoring. This research aided two northern Ohio distribution systems in completion of the IDSE requirements and further develops the models to predict chlorine residuals and trihalomethane (THM) concentrations.
Existing models were utilized as a base and updated with data collected during extensive field studies. A Form 4: Modeling Study Plan, which provides justification that the models meet minimum calibration requirements was submitted and approved by the USEPA. The updated models were utilized to select Stage 2 D/DBP future monitoring locations. Drafts of Form 5: IDSE Report for a Modeling SSS were completed for each system.
First-order reaction expressions were developed to estimate the effects of bulk reactions and reactions occurring along the surface of the distribution system pipes (wall) for free chlorine decay and THM formation. Most reaction expressions that utilize a global wall reaction rate coefficient in units of inverse time do not accurately represent the proportion of reactions occurring along the wall. As a chemical travels through a pipe, wall reactions within larger diameter pipes will have less of an effect on the overall reaction than when flowing through a smaller diameter pipe. This is because the ratio of the pipe surface to a unit volume of water is not constant throughout the distribution system. This is, however, constant given a specific flow path from the point of entry into the distribution system to any fixed point, such as a sampling location. In order to account for this wall reaction variability, a series of flow path diameter fractions are employed within the chlorine decay and THM formation expressions. A probabilistic approach using Bayesian statistics is then utilized to estimate the bulk and wall reaction coefficients. The flow path specific expressions better predicted actual data.