Search

Selecting the optimal monitoring points in a water distribution network is challenging due to the complex spatiotemporal variability of water quality degradation. The lack of a standardized methodology for monitoring point selection forces operators to rely on general recommendations, historical data and professional experience, which can mask water quality problems and increase the risk to consumers. This study proposes a new methodology to optimize the selection of monitoring points in distribution networks. The method considers the spatiotemporal degradation of water quality, the definition of representative zones and two selection criteria: global representativeness and potential health risk. Representative zones were determined for each node of the network based on hydraulic paths and their water quality spatial variability. Part of the distribution network in Quebec City, Canada was used as the case study, in which four water quality parameters were investigated: free chlorine residual (FRC), heterotrophic plate counts (HPC), trihalomethanes (THMs) and haloacetic acids (HAAs). Seasonal variabilities (summer and winter) were also analyzed. The results obtained for the two criteria and for both seasons were compared, and methodological and practical recommendations were established for dynamic monitoring programs that respond to the needs of operators., Competing Interests: Declaration of competing interest The authors declare the following financial interests/personal relationships which may be considered as potential competing interests:Manuel J. Rodriguez reports financial support was provided by Natural Sciences and Engineering Research Council of Canada. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 The Authors. Published by Elsevier Ltd.. All rights reserved.)

Cryptosporidium and Giardia are protozoan parasites responsible for gastrointestinal illnesses in humans and in animal species. The main way these parasites are transmitted is by ingestion of their (oo)cysts in drinking water. Monitoring (oo)cysts in water sources is beneficial to evaluate the quality of raw water supplying treatment plants. Currently, the only standardized protocol to enumerate these parasites from water samples is United States Environmental Protection Agency (USEPA) Method 1623.1. With this method, we monitored three major water sources in Quebec over a year to assess temporal and geographical variations of these parasite (oo)cysts. These three water sources have independent watersheds despite being in the same region. We found a general pattern for Giardia , with high concentrations of cysts during cold and transition periods, and significantly lower concentrations during the warm period. Cryptosporidium 's concentration was more variable throughout the year. Statistical correlations (Pearson's correlation coefficients) were established between the concentration of each parasite and various environmental parameters. The three study sites each showed unique factors correlating with the presence of both protozoa, supporting the idea that each water source must be seen as a unique entity with its own particular characteristics and therefore, must be monitored independently. Although some environmental parameters could be interesting proxies to the parasitic load, no parameter was strongly correlated throughout the whole sampling year and none of the parameters could be used as a single proxy for all three studies sources., Competing Interests: The authors have no competing interests.

Municipal water supply through truck-to-cistern systems is common in northern Canada. Household satisfaction and concerns about water services likely impact user preferences and practices. This case study explores household perspectives and challenges with regard to domestic access to water in a decentralized truck-to-cistern system. A case study was conducted in the Northern Village of Kangiqsualujjuaq, Nunavik (Quebec, Canada). A paper-based questionnaire was completed by 65 households (one quarter of the population). Many households (37%) reported not drinking tap water from the truck-to-cistern system. Chlorine taste was a frequently reported concern, with those households being significantly less likely to drink water directly from the tap ( p = 0.002). Similarly, households that reported a water shortage in the previous week (i.e., no water from the tap at least once) (33%) were more likely to express dissatisfaction with delivered water quantity ( r s = 0.395, p = 0.004). Interestingly, 77% of households preferred using alternative drinking water sources for drinking purposes, such as public tap at the water treatment plant, natural sources or bottled water. The study underscores the importance of considering household perspectives to mitigate the risks associated with service disruptions and the use of alternative sources for drinking purposes., Competing Interests: The authors declare there is no conflict., (© 2024 The Authors This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence (CC BY 4.0), which permits copying, adaptation and redistribution, provided the original work is properly cited (http://creativecommons.org/licenses/by/4.0/).)

The goal of this work was to investigate the occurrence of contaminants of emerging interest (CEI) in source surface water (SW; river water) and drinking water (DW; tap water) from five drinking water treatment plants (DWTPs) in the Province of Québec, Canada. A total of 28 sampling campaigns were conducted to collect SW and DW samples from each DWTP from June 2016 to July 2017. The seven targeted CEI, including acetaminophen, salicylic acid, caffeine, carbamazepine, ibuprofen, sulfamethoxazole and drospirenone, were analyzed using solid-phase extraction-ultra pressure liquid chromatography-mass spectrometry (SPE-UPLC-MS/MS) for all collected water samples. The selected CEI were detected in all SW and DW samples, with the exception of drospirenone, which occurred in amounts that were below the limit of detection in one DWTP in June and July 2016. In all the SW samples, caffeine was detected and had the highest median concentration range (12.3-91.0 ng/L), followed by acetaminophen (7.9-85.0 ng/L) and salicylic acid (21.6-39.0 ng/L). In the DW samples, salicylic acid was detected and had the highest median concentration range (20.5-50 ng/L), followed by caffeine (5.2-21.8 ng/L), and acetaminophen (5.0-7.7 ng/L). Carbamazepine, ibuprofen, and sulfamethoxazole primarily occurred in amounts between the limit of detection and limit of quantification in SW and occurred below the limit of detection in DW. All the DWTPs exhibited a similar trend in the removal of CEI, which include acetaminophen (≤97.6%), followed by caffeine (71.0-86.5%) and salicylic acid (<50.0%). Varying levels of efficiencies were observed among the removal strategies for CEI under study, which were mainly associated with the contaminant concentration in SW in the case of acetaminophen, and with the treatment processes in the case of caffeine and salicylic acid., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2020 Elsevier B.V. All rights reserved.)

The objective of this study is to propose and apply a novel approach to develop a drinking water quality index that enables water managers to routinely identify the vulnerabilities in a distribution system (DS) while taking into account the priorities of water managers and operators. Here, we propose an innovative approach based on the participation of eight researchers and water managers. The input from these participants was included through a combination of the Delphi and MACBETH (Measuring Attractiveness by a Categorical-Based Evaluation Technique) methods, which were applied to an index developed using fuzzy logic. The index was then validated by conducting sensitivity and comparative analyses on a database from field sampling campaigns conducted in various small municipal DSs in two Canadian provinces: Quebec and Newfoundland and Labrador. Finally, the drinking water quality index was applied to a database of the DS in Quebec City, Canada. Sensitivity and comparative analyses revealed that the developed index could allow for the characterization of water sample quality and could prove useful for prioritizing interventions in the DS. The index was also useful for representing the spatial variability of water quality, and provided nuanced information about water quality, especially when the water quality parameter values complied with guidelines and regulations. These results provide a transferable and integrated approach for developing customized drinking water quality indices. This study will help pave the way for water quality managers to better prevent episodes of possible water quality deterioration.

Perception of tap water is subject to a wide range of factors and interactions. These include risk perception, tap water quality and organoleptic perceptions, microbiological and chemical quality, prior experiences, information sources, trust in water companies and other groups, and perceived control and contextual factors, among others. The objective of this study is to assess the factors that influence and determine citizen behavior regarding drinking water. A phone survey was conducted among 1014 citizens living in the city of Québec, Canada. Five different domestic water consumption profiles were elaborated according to the citizens' preferences and behavior. Descriptive statistics and mediation analyses were carried out to analyse the survey results and assess the factors modifying the links between satisfaction and water consumption behavior. Results show that drinking water quality could be loosely linked with overall satisfaction with tap water. The water consumption profile was strongly linked with satisfaction levels related to the taste, odor and color of tap water. We observed that the association between an individual's tap water satisfaction and water consumption behavior was mediated by the water treatment strategies applied at home (filtering, cooling), knowledge about drinking water quality and its production, and risk perception. The mediating effects were shown to be significant mainly among bottled-water-only and tap-water-only consumers. Future interventions that aim to encourage the population's use of tap water as a primary source should prioritize cooling and filtering tap water in their messaging, in order to improve population satisfaction. The reduction of risk perception through targeted information campaigns is also of primary importance for decreasing the number of citizens who exclusively drink bottled water., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2020 Elsevier B.V. All rights reserved.)

As the third largest group of identified disinfection by-products (DBPs) by weight, halogenated acetaldehydes (HALs), were monitored for one year at numerous locations in two full-scale drinking water systems applying an ozone-chlorine sequential disinfection strategy. The HALs that were targeted included four trihalogenated acetaldehydes (THALs): chloral hydrate (CH), bromodichloroacetaldehyde (BDCAL), dibromochloroacetaldehyde (DBCAL) and tribromoacetaldehyde (TBAL). Three dihalogenated acetaldehydes (DHALs) were also included: dichloroacetaldehyde (DCAL), bromochloroacetaldehyde (BCAL) and dibromoacetaldehyde(DBAL). In addition to various sampling points in two distribution networks, this study also investigated the formation of HALs during water treatment and for the first time, reports the formation of DBAL before chlorine is applied. Low bromide levels in source waters from both systems resulted in the rare detection of DBAL and TBAL. CH accounted for >50% of total HALs (HAL7) with DHALs accounting for as little as 10% of HAL7, presumably due to the use of ozone-chlorine instead of ozone-chloramine. In the presence of chlorine residuals and with increasing water residence times, most HALs continued to form, more readily in warm water than in cold water. However, the spatial and temporal patterns for each HAL differed depending on speciation (THAL vs. DHAL) and water temperature. Compared to the relatively stable bromine incorporation factor (BIF) of THMs in the distribution systems, the decreasing BIFs of HALs according to water residence time increases suggested that bromine-containing THMs are more stable than their corresponding HALs. Re-chlorination at the extremities of the distribution networks demonstrated a significant impact on the occurrence and speciation of DBPs. In both full-scale systems, water temperature was shown to be the biggest contributing factor to HAL formation. The strong correlations between THM levels and THAL levels make it possible to predict the occurrence of THALs based on THMs., (Copyright © 2019 Elsevier B.V. All rights reserved.)

Disinfection by-products (DBPs) constitute a large family of compounds. Trihalomethanes and haloacetic acids are regulated in various countries, but most DBPs are not. Monitoring DBPs can be delicate, especially for small systems, because various factors influence their formation and speciation. Short-term variations of DBPs can be important and particularly difficult for small systems to handle because they require robust treatment and operation processes. According to our knowledge, for the first time, our study covers the short-term variability of regulated and non-regulated DBP occurrence in small systems in the summer. An intensive sampling program was carried out in six small systems in Canada. Systems in the provinces of Newfoundland and Labrador and Quebec were sampled daily at the water treatment plant and at six different locations along the distribution system. Five DBP families were studied: trihalomethanes, haloacetic acids, haloacetonitriles, halonitromethanes and haloketones. Results show that there were considerable variations in DBP levels from week to week during the month of study and even from day to day within the week. On a daily basis, DBP levels can fluctuate by 22% to 96%. Likewise, the large number of sampling locations served to observe DBP variations along the distribution system. Observations revealed some degradation and decomposition of non-regulated DBPs never before studied in small systems that are associated with the difficulty these systems experience in maintaining adequate levels of residual disinfectant. Finally, this study reveals that the short term temporal variability of DBPs is also influenced by spatial location along the distribution system. In the short term, DBP levels can fluctuate by 23% at the beginning of the system, compared to 40% at the end. Thus, spatial and temporal variations of DBPs in the short term may make it difficult to select representative locations and periods for DBP monitoring purposes in small systems., (Copyright © 2015. Published by Elsevier B.V.)

Background: Past studies have examined the effects of maternal exposure to water chlorination disinfection by-products (DBPs), such as trihalomethanes (THMs) and haloacetic acids (HAAs) during pregnancy. However, no human-based study has yet evaluated the effect of emerging DBPs, such as haloacetaldehydes (HAs) and haloacetonitriles (HANs) on small-for-gestational-age (SGA) status in newborns., Objective: This study aims to assess the association between maternal multiroute exposure to HAs and HANs during the third trimester of pregnancy and SGA status at birth, among neonates delivered by women residing in the Quebec City area (Province of Quebec, Canada). We also evaluated the interaction between exposure to these emerging unregulated by-products and regulated DBPs also found in drinking water (THMs and HAAs), for which a positive association with adverse reproductive outcomes has been suggested in previous studies., Methods: We conducted a population-based case-control study in the Quebec City area. SGA newborns (n=330) were compared to 1100 controls, with matching based on calendar week of birth. HA and HAN concentrations in drinking water at participant's tap were estimated using spatio-temporal strategy based on bimonthly measurements carried out at several locations in the participant's distribution system. A computer-assisted telephone interview was completed to collect information on individual habits of water consumption and water related activities in order to determine individual multiroute exposure. This enabled us to estimate the dose of HAs and HANs absorbed daily by each participant. Associations between total HA, HAN concentrations in drinking water and SGA were analyzed. Associations between the daily-absorbed doses of these emerging DBPs and SGA were also analyzed. Odds ratios (ORs) comparing the 4th quartile of exposure to the reference group (the first three quartiles) were obtained by means of conditional logistic regression, and controlling for potential confounders., Results: Globally, no evidence of increased risk of SGA was found with total HA and HAN concentrations in tap water when participants in the 4th quartile of exposure were compared to the first three quartiles (OR=1.0; 95% CI [0.7-1.5] and OR=0.8; 95% CI [0.6-1.2], respectively). Similarly, no association was found with the daily-absorbed doses of total HAs or HANs (OR=0.9; 95% CI [0.6-1.3] and OR=1.1; 95% CI [0.7-1.6], respectively). However, a small non statistically significant association was found between the dose of brominated HA and SGA (OR=1.4; 95% CI [0.9-2.1]). Also, in spite of the lack of interaction between other DBP classes, an unexpected negative interaction was observed between concentration of chloral hydrate (CH) (which represents the main HA species), and regulated DBPs (P=0.006)., Conclusion: In this population, exposure to low levels of HAs and HANs during the third trimester of pregnancy through drinking water was not associated to SGA status in newborns. Nonetheless, more research is needed to clarify possible effect of brominated compounds and interaction between different DBPs., (Copyright © 2015 Elsevier Inc. All rights reserved.)

Dissolved organic carbon (DOC) is a recognized indicator of natural organic matter (NOM) in surface waters. The aim of this paper is twofold: to evaluate the impact of geophysical characteristics, climate and ecological zones on DOC concentrations in surface waters and, to develop a statistical model to estimate the regional variability of these concentrations. In this study, multilevel statistical analysis was used to achieve three specific objectives: (1) evaluate the influence of climate and geophysical characteristics on DOC concentrations in surface waters; (2) compare the influence of geophysical characteristics and ecological zones on DOC concentrations in surface waters; and (3) develop a model to estimate the most accurate DOC concentrations in surface waters. The case study involved 115 catchments from surface waters in the Province of Quebec, Canada. Results showed that mean temperatures recorded 60 days prior to sampling, total precipitation 10 days prior to sampling and percentages of wetlands, coniferous forests and mixed forests have a significant positive influence on DOC concentrations in surface waters. The catchment mean slope had a significant negative influence on DOC concentrations in surface waters. Water type (lake or river) and deciduous forest variables were not significant. The ecological zones had a significant influence on DOC concentrations. However, geophysical characteristics (wetlands, forests and slope) estimated DOC concentrations more accurately. A model describing the variability of DOC concentrations was developed and can be used, in future research, for estimating DBPs in drinking water as well evaluating the impact of climate change on the quality of surface waters and drinking water., (Copyright © 2014 Elsevier B.V. All rights reserved.)

The spatiotemporal presence of eight N-nitrosamines in the water of seven supply systems in Quebec considered to be susceptible to these emerging disinfection by-products was evaluated. This is the first study on the presence of N-nitrosamines in drinking water utilities in Quebec. Seven sampling campaigns were carried out at several sampling points in each of the systems over a period of 1 year. The results show that N-nitrosamines, primarily N-nitrosodimethylamine (NDMA), were not commonly detected in the water of the facilities under study (10 % of samples). The concentrations measured were lower than those reported in recent North American studies. None of the 195 samples taken exceeded the Ontario standard of 9 ng/L for NDMA (maximum value observed of 3.3 ng/L). N-nitrosomethylethylamine and N-nitrosopiperidine were detected once, with concentrations of 3.7 and 6.0 ng/L, respectively. Chloramination was identified as being the main risk factor regarding the presence of N-nitrosamines, but water quality and some operating parameters, in particular disinfectant residual, also seem to be related to their presence. NDMA concentrations at the end of the distribution systems were generally higher than water leaving the plant. No seasonal trends were observed for the formation of N-nitrosamines in the investigated supply systems. Finally, an association between the presence of N-nitrosamines and the levels of trihalomethanes and haloacetic acids was observed in some facilities.

The non-regulated disinfection by-products (NrDBP) targeted in this study include four haloacetonitriles (trichloroacetonitrile (TCAN); dichloroacetonitrile (DCAN); bromochloroacetonitrile (BCAN) and dibromoacetonitrile (DBAN)); one halonitromethane (trichloronitromethane, better known under the name chloropicrin (CPK)); and two haloketones (1,1-dichloro-2-propanone (11DCPone) and 1,1,1-trichloro-2-propanone (111TCPone)). This study provides a detailed picture of the spatial and temporal variability of these NrDBP concentrations throughout a drinking water distribution system located in a region with major seasonal climate variations. The results obtained show that the concentrations of the investigated NrDBPs varied significantly according to time and location. The average concentrations of TCAN, DCAN, CKP and 111TCPone were significantly higher in summer. Surprisingly, the average concentrations of 11DCPone were significantly higher in winter. For BCAN and DBAN, the average concentrations observed in winter were higher, but not in a statistically significant way. On the other hand, the four HANs, CPK and 111TCPone generally had spatial profiles involving an increase of the concentrations along the network according to increasing water residence times, whereas 11DCPone overall had a profile where concentrations increased at the beginning of the network, followed by a drop in the concentrations towards the ends of the network. In spite of certain disparities in the individual spatio-temporal variation profiles, strong correlations were generally observed between NrDBPs, and trihalomethanes (THMs) and haloacetic acids (HAAs). Therefore, THMs and HAAs could be good statistical indicators of the presence of NrDBPs in the drinking water of the system under study., (Copyright © 2013 Elsevier Ltd. All rights reserved.)

Swimming is one of the most popular aquatic activities. Just like natural water, public pool water may contain microbiological and chemical contaminants. The purpose of this study was to study the presence of chemical contaminants in swimming pools, in particular the presence of disinfection by-products (DBPs) such as trihalomethanes (THMs), haloacetic acids (HAAs) and inorganic chloramines (CAMi). Fifty-four outdoor and indoor swimming pools were investigated over a period of one year (monthly or bi-weekly sampling, according to the type of pool) for the occurrence of DBPs. The results showed that DBP levels in swimming pools were greater than DBP levels found in drinking water, especially for HAAs. Measured concentrations of THMs (97.9 vs 63.7 μg/L in average) and HAAs (807.6 vs 412.9 μg/L in average) were higher in outdoor pools, whereas measured concentrations of CAMi (0.1 vs 0.8 mg/L in average) were higher in indoor pools. Moreover, outdoor pools with heated water contained more DBPs than unheated pools. Finally, there was significant variability in tTHM, HAA9 and CAMi levels in pools supplied by the same municipal drinking water network, suggesting that individual pool characteristics (number of swimmers) and management strategies play a major role in DBP formation., (Copyright © 2012 Elsevier Ltd. All rights reserved.)

Groundwater contamination by nitrate was investigated in an agricultural area in southern Quebec, Canada, where a municipal well is the local source of drinking water. A network of 38 piezometers was installed within the capture zone of the municipal well to monitor water table levels and nitrate concentrations in the aquifer. Nitrate concentrations were also measured in the municipal well. A Water flow and Nitrate transport Global Model (WNGM) was developed to simulate the impact of agricultural activities on nitrate concentrations in both the aquifer and municipal well. The WNGM first uses the Agriflux model to simulate vertical water and nitrate fluxes below the root zone for each of the seventy agricultural fields located within the capture zone of the municipal well. The WNGM then uses the HydroGeoSphere model to simulate three-dimensional variably-saturated groundwater flow and nitrate transport in the aquifer using water and nitrate fluxes computed with the Agriflux model as the top boundary conditions. The WNGM model was calibrated by reproducing water levels measured from 2005 to 2007 in the network of piezometers and nitrate concentrations measured in the municipal well from 1997 to 2007. The nitrate concentrations measured in the network of piezometers, however, showed greater variability than in the municipal well and could not be reproduced by the calibrated model. After calibration, the model was validated by successfully reproducing the decrease of nitrate concentrations observed in the municipal well in 2006 and 2007. Although it cannot predict nitrate concentrations in individual piezometers, the calibrated and validated WNGM can be used to assess the impact of changes in agricultural practices on global nitrate concentrations in the aquifer and in the municipal well., (Copyright © 2011. Published by Elsevier B.V.)

Background: The relationship between chlorination by-products (CBPs) in drinking water and human health outcomes has been investigated in many epidemiological studies. In these studies, population exposure assessment to CBPs in drinking water is generally based on available CBP data (e.g., from regulatory monitoring, sampling campaigns specific to study area). Since trihalomethanes (THMs) and haloacetic acids (HAAs) are the most documented CBP classes in drinking water, they are generally used as indicators of CBP exposure., Methods: In this paper, different approaches to spatially assign available THM and HAA concentrations in drinking water for population exposure assessment purposes are investigated. Six approaches integrating different considerations for spatial variability of CBP occurrence within different distribution systems are compared. For this purpose, a robust CBP database (i.e., high number of sampling locations selected according to system characteristics) corresponding to nine distribution systems was generated., Results and Conclusion: The results demonstrate the high impact of the structure of the distribution system (e.g., presence of intermediary water infrastructures such as re-chlorination stations or reservoirs) and the spatial variability of CBPs in the assigned levels for exposure assessment. Recommendations for improving the exposure assessment to CBPs in epidemiological studies using available CBP data from water utilities are also presented.

Heterotrophic plate count (HPC) constitutes a common indicator for monitoring of microbiological water quality in distribution systems (DS). This paper aims to identify factors explaining the spatiotemporal distribution of heterotrophic bacteria and model their occurrence in the distribution system. The case under study is the DS of Quebec City, Canada. The study is based on a robust database resulting from a sampling campaign carried out in about 50 DS locations, monitored bi-weekly over a three-year period. Models for explaining and predicting HPC levels were based on both one-level and multi-level Poisson regression techniques. The latter take into account the nested structure of data, the possible spatiotemporal correlation among HPC observations and the fact that sampling points, months and/or distribution sub-systems may represent clusters. Models show that the best predictors for spatiotemporal occurrence of HPC in the DS are: free residual chlorine that has an inverse relation with the HPC levels, water temperature and water ultraviolet absorbance, both having a positive impact on HPC levels. A sensitivity analysis based on the best performing model (two-level model) allowed for the identification of seasonal-based strategies to reduce HPC levels.

The increasing frequency of waterborne outbreaks demonstrates that classic indicators used for the surveillance of the microbiological quality of drinking water have several gaps and that routine public health surveillance seems insufficient to allow for the rapid detection of these outbreaks. The main objective of this study was to evaluate the possibility of using a regional health information telephone line, 'Info-Santé CLSC' (Info-Health Local Community Health Centre), for the surveillance of waterborne gastroenteritis. This study measured the incidence rate of calls for acute gastrointestinal illness (AGI) placed to the Info-Santé CLSC line, investigated the relationship between the frequency of calls for AGI placed to the Info-Santé CLSC line and the turbidity of the treated water in the Quebec City drinking water plant and evaluated the relevance and the conditions of use of the Info-Santé CLSC system for the surveillance of waterborne enteric illness. A relationship between the turbidity and the calls for AGI placed to Info-Santé CLSC line was observed. Significant time lags (11, 15 and 17 days prior to the outcome) were identified in the final model derived from a Poisson model using generalized additive models (GAM) as a time series analysis. Some recommendations to improve the system were formulated even though the system already seems to be useful for the surveillance of waterborne enteric diseases.

This study presents a comparative performance analysis of small drinking water utilities in Quebec (Canada). The investigation bears on 10 utilities that use surface water or groundwater under the direct influence of runoff and apply chlorination as the only treatment before distribution. The utilities under study were divided into two groups: four utilities that had never or rarely provided water violating provincial drinking water microbiological standards (relating to fecal and/or total coliform bacteria), called nonproblematic utilities, and six utilities that quite often violated the standards, designated as problematic utilities. The objective of the study is to develop utility performance indicators capable of explaining current and historical distributed water quality. Indicators are based on operational, infrastructure, and maintenance characteristics of utilities that are integrated using a multivariable weight-based index. Results show that utility performance indicators are systematically better for the nonproblematic group of utilities as compared to the problematic group. Disinfection-related, infrastructure, and maintenance variables are those that most contributed to indicator values. Sensitivity analyses served to assess the impact on indicator results of excluding variables and changing their weights.

The purpose of every water utility is to provide consumers with drinking water that is aesthetically acceptable and presents no risk to public health. Several studies have been carried out to analyze people's perception and attitude about the drinking water coming from their water distribution systems. The goal of the present study is to investigate the influence of water quality and the geographic location of consumers within a distribution system on consumer perception of tap water. The study is based on the data obtained from two surveys carried out in municipalities of the Quebec City area (Canada). Three perception variables were used to study consumer perception: general satisfaction, taste satisfaction and risk perception. Data analysis based on logistic regression indicates that water quality variations and geographic location in the distribution system have a significant impact on the consumer perception. This impact appears to be strongly associated with residual chlorine levels. The study also confirms the importance of socio-economic characteristics of consumers on their perception of drinking water quality.

The levels of trihalomethanes (THMs)--the main species of by-product from water chlorination--were monitored in the distribution systems of the five major drinking water utilities of the greater area of Québec City in order to investigate and model their occurrence on a spatial and seasonal basis. Data for THMs and other water quality and operational parameters associated with their formation were generated through a 16 month sampling program involving several sites representing variable water residence times, from the plant to the system extremity. The results demonstrate that the differences in measured THM levels between the five utilities are mainly due to the variable quality of raw waters, the type of water treatment process being used and the type and levels of applied disinfectant. Depending on the utility, average THM levels were from 1.3 to 2.5 times higher in the system extremities than in the water leaving the treatment plant. Also, average levels of THMs measured in summer at the distribution system extremities were, depending on the utility, from 2.5 to 5 times higher than the average levels measured in winter. The seasonal differences were found to be significantly greater than those observed by others in water utilities in the United States and Europe and are explained in large part by the considerable changes, over the year, in the quality and temperature of surface waters in Southern Quebec. For the live utilities under study, multivariate regression models were developed in order to predict spatial and seasonal variations of THMs. Both residual chlorine demand and temperature were found to be better, statistically, as predictors for THM occurrence. The usefulness of the developed models for routine and long term water quality management, as well as for assessment of human exposure to THMs, are also discussed.

Haloacetic acids (HAAs) and trihalomethanes (THMs) were generated in bench-scale chlorination experiments using treated waters (prior to final chlorination) of the three major drinking water utilities of the Quebec City area. The purpose was to investigate the formation and occurrence of these chlorination by-products (CBPs) on a seasonal basis. Data for HAAs, THMs and other physico-chemical parameters were produced through a six-month sampling program with variable conditions of water quality, water temperature, applied chlorine dose and reaction time. In waters from the three utilities, chloroform (THM specie), dichloroacetic and trichloroacetic acid (HAA species) were the most prevalent compounds due to the low concentrations of bromide in the utilities' raw waters. Significant differences in CBP occurrence were noted between the three utilities' chlorinated waters, mainly due to the type of disinfectant applied to raw water. The use of pre-ozonation, as opposed to pre-chlorination (or direct chlorination) in one of the utilities appears to be the major factor contributing to that utility's potential for compliance with current THM and future HAA standards. Seasonal variations in THMs and HAAs were mainly associated with variations in organic precursors and to changes in water temperature (two parameters which vary widely on a seasonal basis in surface waters of southern Quebec), with CBP occurrence at its highest in spring. Statistical correlations between HAAs and THMs were moderate and only temperature appeared to affect the preponderance of one CBP or the other. Finally, a regression analysis was carried out aimed at associating each CBP to water quality and the experimental parameters. Thanks to their predictive ability, multivariate models seem to be the tools with the best potential for decision-making purposes.