Modeling the susceptibility of drinking water utilities to form high concentrations of trihalomethanes

Abstract

Chlorination is a key process in the production of drinking water which aims at ensuring an acceptable microbiological quality of water in urban distribution systems. But chlorine is also known to react with organic matter naturally present in water to form disinfection by-products such as trihalomethanes (THMs). Because epidemiological and toxicological studies have suggested THMs are potentially carcinogenic for humans, most industrialized countries have established maximum acceptable THM levels as part of their water quality standards. Modeling THM occurrence in water may be a valuable tool for decision-makers in dealing with these potentially hazardous by-products. This paper presents a modeling approach based on logistic regression analysis which is designed to estimate the susceptibility of water utilities (in Quebec, Canada) to generating high levels of THMs. The approach is aimed at estimating the probability to exceed established THM thresholds, based on general utility characteristics (type of water source, type of water treatment, geographical location) as well as information about water quality and operational parameters. The models are able to identify the worst conditions for THM formation. A sensitivity analysis using these models allows for the identification of strategies for THM reduction in distribution systems as well as an estimation of the capacity of utilities to comply with current and future regulations. The use of such models to generate data for regional epidemiological studies is also discussed.