Abstract
Bromide concentrations increased significantly in the Monongahela River in Pennsylvania in 2010–2011. These increases led to increases in trihalomethane (THM) and haloacetic acid (HAA) and increased bromination of these disinfection by-products (DBPs) in drinking water at plants on the Monongahela River. This study presents a statistical simulation model to evaluate the effect of the increasing source-water bromide on THM formation and speciation and analyzes the changing risks (by using cancer slope factors) in treated water from 2010 to 2012. Even very low bromide concentrations were associated with increased cancer risk from . increased with increasing source-water bromide as well, but the bromination factors of regulated decreased. Increasing source-water bromide in the Monongahela River favored bromine incorporation in rather than . Collection and analysis of bromide and THM and HAA data for the development of similar bromination factor regression analysis and human health risk assessment for regionally relevant elevated bromide in source water can be used in other areas experiencing or expecting similar source-water changes.
Introduction
Surface waters in the United States usually have very low bromide concentrations from natural sources (Flury and Papritz 1993), with elevated values reported near the coastlines (Chen et al. 2010; Holm et al. 2007). Anthropogenic bromide can enter source water from human activities, including agricultural applications, road runoff, and industrial discharges (Sollars et al. 1982; Wegman et al. 1983; Winid 2015). Recently, bromide discharges from oil and gas produced water disposal, and coal-fired power plant effluents, have been identified as new sources (Greune 2014; Landis et al. 2016; States et al. 2013; Weaver et al. 2015; Wilson and VanBriesen 2013). The discharge of bromide to surface waters is currently unregulated in the United States (DiCosmo 2012) because bromide has high human and ecotoxicity thresholds (Flury and Papritz 1993). In the fall of 2015, U.S. Environmental Protection Agency (EPA) released revised effluent limitation guidelines and standards (ELGs) for the steam electric power generating point source category. The EPA indicated that permitting authorities may set water quality–based effluent limitations on bromide to protect downstream drinking water plants and their customers from the effects of increasing bromide in source waters (U.S. EPA 2015). However, the methods for determining acceptable bromide levels for particular water bodies or drinking water intakes to ensure that drinking water safety is not compromised have not been reported.
Bromide in source water may affect finished drinking water quality and has been a concern for drinking water plants for many years (Heeb et al. 2014; Luong et al. 1980; Singer 1994; Westerhoff et al. 2004). Bromide reacts with applied disinfectants to form bromine, which then oxidizes organic matter, creating brominated and mixed bromochloro-disinfection by-products (DBPs). Higher concentrations of bromide increase the rate of DBP formation, leading to higher overall DBP concentrations (Heeb et al. 2014; Westerhoff et al. 2004) and increasing the incorporation of bromide into the formed DBPs (Diehl et al. 2000; Hellergrossman et al. 1993; Pourmoghaddas et al. 1993). Some DBPs have been associated with cancer and negative reproductive outcomes (U.S. EPA 1998; Villanueva et al. 2015).
Scientific studies have advanced understanding of the possible toxicological effects of DBPs, with the initial rodent toxicological studies linking chloroform to cancer no longer supported (Hrudey et al. 2015). Focus has instead turned to epidemiological studies suggesting a relationship between bladder cancer, other effects, and DBP exposure (Driedger et al. 2002; Grellier et al. 2015; Henry 2013; Hrudey et al. 2015). As the search for more specific DBP causal species and mechanisms continues (Hrudey et al. 2015), precautionary health-protective approaches to DBP regulation at the EPA and other environmental agencies remain on the basis of relationships estimated for the cancer endpoint. Brominated DBPs are in general more genotoxic, cytotoxic, and carcinogenic than chlorinated DBPs (Echigo et al. 2004; Plewa et al. 2004; Richardson et al. 2007), and the authors thus emphasize the effect of bromination on cancer risk in the analysis that follows.
Regulation of DBPs in the United States is based on an indicator approach (Richardson et al. 2007). The disinfectants/disinfection by-products (D/DBP) rule was established in 1998 to regulate the total trihalomethanes (TTHM) as a mass sum at a maximum contaminant level (MCL) of and to regulate a group of five haloacetic acids () as a mass sum at a MCL of (U.S. EPA 1998). The TTHM is defined as the sum of mass concentrations of chloroform, bromodichloromethane (BDCM), dibromochloromethane (DBCM), and bromoform. Because other trihalomethanes (THM) can form in drinking water, the sum of these four specific THM is sometimes referred to as . In the present work, TTHM is used when referring to the regulatory value, whereas the more explicit is used when referring to the measurements. Individual THM species were assigned maximum contamination level goals (MCLGs); MCLGs are nonenforceable contaminant levels at which no adverse health effects are likely to occur, allowing for a margin of safety (U.S. EPA 2006). The MCLGs for chloroform, BDCM, DBCM, and bromoform are 70, 0, 60, and , respectively (U.S. EPA 2006). The five regulated are monochloroacetic acid (MCAA), dichloracetic acid (DCAA), trichloroacetic acid (TCAA), monobromoacetic acid (MBAA), dibromoacetic acid (DBAA), which includes two bromine-containing compounds; the sum of the five regulated forms is often reported as . The MCL of is based on the MCLG for MCAA and TCAA, which are 70 and , respectively, considering the best technology and treatment techniques (U.S. EPA 2009). The four unregulated HAAs that are commonly measured in drinking water are bromochloroacetic acid, BCAA; dibromochloroacetic acid, DBCAA; bromodichloroacetic acid, BDCAA; tribromoacetic acid, TBAA; all contain at least one bromine. Neither MCLG nor MCL has been promulgated for these four HAAs. The sum of and these four HAAs is often reported as . Haloacetic acids that contain iodine may also be formed in drinking water; these were not measured in the present study. Because DBPs typically increase between the drinking water plant and the compliance sampling points in the distribution system, drinking water utilities often have a target of 80% of the MCL for water leaving the treatment plant (Becker et al. 2013; Roberson et al. 1995).
Occurrence studies show wide variation in DBP levels associated with different source-water bromide levels (Amy et al. 1994; McLain et al. 2002). Drinking water sources in the United States were characterized in the 1996 EPA information collection rule (ICR) (U.S. EPA 2000c). Source-water bromide concentrations were reported to range from (less than the detection limit) to (), with a mean of () and a median of () (McGuire et al. 2002), whereas varied from 2 to , and varied from 5 to (Obolensky 2007). Higher bromide source waters were associated with higher DBP formation and increased bromination of the DBPs that form; however, a wide variation in DBP levels associated with specific source-water bromide concentrations was observed (Amy et al. 1994; McLain et al. 2002).
Because source-water bromide concentration differences are generally assumed to be attributed to natural conditions, few studies have assessed the effect of changing bromide concentrations on particular utilities. The Metropolitan Water District of Southern California has studied changing bromide caused by seawater intrusion; approached when bromide was elevated (Krasner et al. 1991). Greune (2014) conducted a study of 78 water treatment facilities in North Carolina experiencing increasing bromination of THM and suggested that brominated THM species become dominant when source-water bromide concentrations exceeded (Greune 2014). Zhang et al. (2011) studied the formation potential of and HAAs in 13 source waters from four major water basins in China and suggested that the formation of brominated DBP is of concern when the bromide concentration exceeds (Zhang et al. 2011).
States et al. (2013) investigated bromide concentrations in the Allegheny River (through daily sampling for a year) and THM formation at the Pittsburgh Water and Sewer Authority treatment plant (with weekly finished-water sampling). They concluded that elevated bromide concentration in their source water led to increasing concentrations of , especially brominated THM in the utility’s finished water (States et al. 2013). Landis et al. (2016), studying the same utility through analysis of river water sampling and reported quarterly THM compliance data, noted that concentration demonstrated an upward trend for each quarter from 2010 to 2014 despite an average quarterly bromide concentration of only (Landis et al. 2016). Landis et al. (2016) conclude that, for this utility, increasing bromide increased bromination extent in TTHM but did not correlate with increasing reported for compliance purposes; however, they note that the nature of the data analyzed prevent understanding of the causality of the observed trends.
Thus, national occurrence studies and field studies indicate that plant-by-plant or regional analyses are necessary to determine protective in-stream bromide concentrations and discharge effluent guidance. Determining a single “safe” bromide level is likely to be impossible because of regional variations in DBP precursors (e.g., characteristics of natural organic matter) and differences in plant operational conditions (e.g., disinfectant type and dosing).
The present study is based on a 3-year field study in the Monongahela River in Pennsylvania. Organic carbon [measured as total organic carbon (TOC) and absorbance at 254 nm (UV254)] did not vary significantly in this river over this time period (Bergman et al. 2016), whereas bromide concentrations showed significant change. Thus, the present study focuses on evaluating the changing bromide concentration and its effects on brominated and formation and speciation in treated drinking water. Changing concentrations of regulated parameters (TTHM and ) and bromination extent are assessed in water leaving the treatment plants. Associated risk (based on cancer slope factors for THM species) for finished drinking waters with differing source-water bromide levels are compared.
Materials and Methods
Source-Water Sampling and Analysis
The Monongahela River is 206 km (128 mi) long and flows north from West Virginia into Pennsylvania. It meets with the Allegheny River in downtown Pittsburgh to form the Ohio River. Source-water sampling from six drinking water treatment plants (identified in this paper as A–F, with A as the farthest upstream near the West Virginia border and F near Pittsburgh) on the Monongahela River was described previously (Wilson and VanBriesen 2013).
Briefly, source-water samples were collected approximately biweekly (once every 2 weeks), collected in 500-mL polypropylene bottles, and stored in an ice cooler at 4°C during transport to the laboratory before laboratory analysis. Bromide concentration in samples was determined by using ion chromatograph (IC, Dionex-500 Ion Chromatography system, Sunnyvale, California) with a sample loop, which yields a bromide detection limit of (U.S. EPA 1997a).
Finished-Water Sampling and Analysis
Finished-water samples were collected at the same six drinking water plants. All the plants in this study used free chlorine disinfection during the study period, with two plants (Sites C and D) applying chlorine from the headworks through the full treatment train, and the other four plants applying chlorine after filtration. Sites B and C switched from chlorine to chloramine after filtration in April 2011 and June 2011, respectively. Finished-water data collected after the switch were not used in this study. Finished-water samples for were collected approximately biweekly from September 2009 to September 2012, and water samples were collected from January 2010 to September 2011. All samples were taken just before water entered the distribution system; thus, they do not represent regulatory compliance sampling locations. In general, concentrations increase with time in the distribution system; thus, finished-water values may underestimate concentrations in consumed water.
Although finished-water values cannot be assessed for plant compliance, comparing the levels of DBPs in finished water with regulatory compliance targets (MCL and 80% of MCL) is of interest. Weekly and biweekly values measured in each plant within each quarter were averaged to improve visualization of results; however, operational evaluation levels (OELs) and running annual averages (RAAs) for each quarter were not assessed in the present work. Thus, these results cannot be used to assess compliance status for the drinking water plants studied.
In the present work, all species in and were measured. Before collecting water samples for analysis, 60-mL amber vials were filled with a 1-g mixture of ammonium chloride () combined with phosphate buffer [1% sodium phosphate () and 99% potassium phosphate ()]. Before collecting water samples for analysis, 60-mL amber vials were filled with 6 g of ammonium chloride (). The added ammonia converts free chlorine residual to combined chlorine and prevents further formation of and . Water samples were stored at 4°C during transport to the laboratory and before laboratory analysis. All samples were analyzed within 14 days of collection. The EPA standard methods 551.1 and 552.3 were used to determine THM and HAA species concentrations, respectively (U.S. EPA 1990, 2003). Briefly, a liquid-liquid extraction using methyl-tertiary-butyl-ether (MTBE) was used, and and concentrations were analyzed by gas chromatograph (HP6890) with electron-capture detector (ECD), using a 0.25-mm inside diameter fused-silica capillary column. For quality control, a calibration check was run every 10 samples, and a duplicate sample was run every 10 samples. The calibration check standard was required to have a relative percent difference (RPD) of less than 10% when compared with the standard concentration, or the results were invalidated. Duplicates were required to have a RPD of less than 5%, or the primary and the duplicate results were invalidated for that sample.
Censored Data Handling
Censored data are concentrations not observed because of analytical detection limitations. Censored bromide and DBP concentrations were observed infrequently. Censored bromide data were handled by using the same procedures used previously (Wilson and VanBriesen 2013). Briefly, the observed data were used to fit a lognormal distribution, which was applied to extrapolate the values that were less than the detection limit (Travis and Land 1990). For censored THM and HAA data, EPA requires results that are less than the detection limit to be reported as zero (U.S. EPA 2012). Thus, to be consistent with the regulatory framework, censored THM and HAA species concentrations were replaced with zero before statistical data analysis in the present work.
Assessing Bromination Fraction
Bromine incorporation factor (BIF) is a method to evaluate the bromine substitution degree of DBPs in various studies (Krasner et al. 2008; Gould et al. 1983). When comparing different classes of DBPs, bromine substitution factor (BSF) is preferred because it ranges from 0 to 1 for all classes, regardless of the number of species (Bond et al. 2014; Hua and Reckhow 2012). Eq. (1) shows the calculation of , in which the THM concentrations are on a molar basis. Similarly, Eqs. (2) and (3) were used to compute .
(1)
(2)
(3)
In addition to the molar-based BSF, percent brominated on a mass basis has recently been used by the Pennsylvania Department of Environmental Protection (PADEP) (Handke 2009) and researchers at a drinking water utility in Pennsylvania (States et al. 2013) to assess the effect of changing bromide concentrations on DBP formation. It is easily computed from compliance data, which are reported in mass units. Percent brominated and were calculated by following Eqs. (4)–(6)
(4)
(5)
(6)
Statistical Data Analysis and Monte Carlo Simulations
The Mann-Whitney rank sum test, which allows evaluation of data sets with nonnormally distributed and censored data, was used by using the software Sigmaplot to evaluate the statistical significance of concentration differences across chemical species, locations, and time periods. Correlations among the source-water bromide concentration and individual DBP species concentrations in the finished water were computed in the statistical software Minitab using Spearman rank correlation analysis because of the non-normally distributed data (Obolensky and Singer 2005).
Monte Carlo simulation has been used previously by EPA to support regulatory impact analysis of DBPs (Gelderloos et al. 1992) and was used in the present work to generate an extension of DBP concentrations that consider correlation and interaction among DBP species (U.S. EPA 1997b). Monte Carlo simulation enables quantitative characterization of the uncertainty and variability in estimated concentrations and provides more information to forecast the full range of possible values in the future. By considering correlations among species, the simulated DBP concentrations better represent the true variability and uncertainty in finished-water quality and reflect the best available knowledge about drinking water quality from the data set (Grayman et al. 2008). In the present work, a correlation coefficient matrix based on field THM data was incorporated into the function of Monte Carlo simulation. A joint lognormal distribution was assumed and fit to the field-sampled concentrations of THM species in each of seven bromide concentration ranges that are greater than the detection limit: 10–20, 20–40, 40–60, 60–80, 80–100, 100–120, and . The data were aggregated across sites and sampling dates to estimate a separate multivariate lognormal model representative of each bromide interval. Each joint lognormal distribution was characterized by
•
() = means of the ln(concentrations) for each of the four THM species (four parameters);
•
() = standard deviations of the ln(concentrations) for each of the four THM species (four parameters); and
•
() = the correlation coefficients between the ln(concentrations) of each of the four THM species (six parameters).
As such, there were a total of 14 parameters fit for each bromide interval, with separate models fit for each of the seven intervals. Monte Carlo sampling for each interval was simulated with 10,000 iterations to generate correlated normally distributed values through @RISK for Excel. Correlation parameters for field sampling THM data and Monte Carlo simulated data and parameter estimates for each interval were developed (Tables S1 and S2 ).
A cumulative distribution function (CDF) plot represents possible values of a variable and the proportion of observations of that variable that are less than the value specified on the horizontal axis (DeGroot and Schervish 2012). The advantage of viewing a distribution with a CDF is that it clearly indicates the likelihood of having an observation that is equal to or less than a specified value of the variable. Cumulative distribution function plots were used with the Monte Carlo–simulated values to view the distribution and determine the probability of obtaining finished water that meets the TTHM standard.
Risk Analysis
The EPA recommends using the cancer slope factor (CSF) or unit risk in risk assessments to estimate the probability of an individual developing cancer resulting from an exposure to a particular level of chemical (U.S. EPA 2005a). Oral CSF is defined as an upper-bound estimate of the lifetime human cancer risk per milligram of chemical per kilogram of body weight per day (Hrudey and Charrois 2012). The EPA conducted a quantitative estimation of carcinogenic risk for THM species in drinking water (U.S. EPA 2005b), and the oral CSF of species (Faust 1992; U.S. EPA 2014a, b, c) are summarized in Table 1. The EPA recommends calculating the cancer risk level from drinking water ingestion by using oral CSF and life average daily dose (LADD) () as in Eq. (7) (Theodore and Dupont 2012). The LADD is calculated as the product of concentration (C), intake rate (IR), and exposure duration (ED) divided by body weight (BW) and lifetime (LT) as in Eq. (8) (Asante-Duah 2002)
(7)
(8)
Lifetime health advisories consider a 70-kg adult consuming 2 L of drinking water per day for 70 years (U.S. EPA 2009). Despite considerable uncertainty in cancer effects and dose-response functions, using CSF to estimate cancer risk levels from chemicals in drinking water remains the baseline approach for regulatory risk assessment (CDPH 2014; Chowdhury et al. 2011; Hrudey 2008; Huy et al. 2014; Kim and Han 2011; Mishra et al. 2014; Pawelczyk 2013; U.S. EPA 2015; Wang et al. 2007). The target risk for DBPs is [one person with cancer per 100,000 people for 70 years (WHO 1993)]; however, the TTHM and MCLs are mass-based class sum standards, and thus they cannot be directly linked to risk of the individual species. Rather, the EPA considered the risk-based MCLGs for each species and the technical and economic feasibility of reducing DBPs in setting the MCLs.
For chemical mixtures, such as DBPs in drinking water, the EPA recommends three approaches for quantitative health risk assessment. The first approach directly uses the available toxicity data of the mixture to evaluate the risk (U.S. EPA 2000b). In the second approach, when toxicity data are not available for the mixture, the EPA recommends using health effects data on a similar mixture (Rice et al. 2009). The third approach is to evaluate the mixture risk through its components by using dose addition or response addition on the basis of the assumption that interaction effects are insignificant and negligible to the risk estimate (U.S. EPA 2000b). This additivity approach is appropriate “when the effect of the mixture can be estimated directly from the sum of the scaled exposure levels (dose addition) or the response (response addition) of individuals components” (U.S. EPA 2000a). Dose addition is only applied when individual components exhibit similar toxicity (U.S. EPA 2000a). The response addition procedure requires determining the individual risk of each component to enable addition of the risk values (U.S. EPA 2000a). In the present work, the response additivity approach is used. The risk of each THM species is calculated by using its CSF. The risk of is then calculated by summing the calculated individual THM species risk values. In the present work, the authors assess the risk of , but not or , because there are limited CSF estimates available for or and no CSF estimates for individual brominated HAA species. The risk calculation was applied to individual THM species in each sample. Data were then averaged for all samples within a quarter to improve visualization of the results.
Although the previously described approach is routinely used for chemical mixture risk assessment and has been applied to THM data in the past (Wang et al. 2007), it is not the approach that was used for setting the regulatory compliance levels for DBPs in drinking water. Because individual DBP species were rarely measured in epidemiological studies that demonstrated elevated risk associated with consumption of chlorinated drinking water, attributable risk estimates could not be developed for individual DBP species. Rather, the EPA used epidemiological studies that show a link between TTHM levels and bladder cancer (U.S. EPA 2006), assuming a linear relationship between TTHM concentration and bladder cancer risk in the economic analysis (U.S. EPA 2005c). Recently, Regli et al. (2015) summarized the basis for the Stage 2 D/DBP rule analysis and used this formulation to estimate the risk associated with increasing bromide concentrations in source waters. However, this approach cannot be used directly to assess changing risk associated with changing speciation because it focuses exclusively on TTHM concentrations.
Results and Discussion
Source-Water Bromide and Finished-Water THMs and HAAs
Bromide concentrations in the source water changed significantly over the 3-year period from 2009 to 2012 in the Monongahela River basin, affecting the formation and speciation of DBPs in finished drinking water. As noted previously, the other major precursor for DBPs, natural organic matter, did not vary significantly (as measured by TOC and UV254) (Bergman et al. 2016).
Table S3 provides the quarterly statistical summary of source-water bromide, , , and at the six drinking water plants. Calendar quarters are defined (January–March, April–June, July–September, and October–December). In the EPA information collection rule for DBPs, a median bromide value of was reported for the national data set (McGuire et al. 2002), whereas bromide concentration in Pennsylvania had a median value of 53 (mean = 68; standard ). In the present study, bromide concentrations in the Monongahela River had a median value of and a mean of (standard deviation = 67). In the present study, the bromide concentration significantly varied spatially and temporally; the values ranged from less than the detection limit (less than ) at all sites to a high value of at Site D in June 2010. Amy et al.’s (1994) national bromide occurrence study previously reported that only 10% of drinking water sources sampled had bromide concentrations of greater than (Amy et al. 1994); however, in the present work, from 2009 to 2012, 18% of the source-water samples in the Monongahela River exceeded bromide. Bromide concentrations specific to Sites E and F were also reported in ICR data from 1997 to 1998, and there were no significant differences between the bromide data in the ICR (full 18 months) and those at Sites E and F during the present study (2009–2012). However, when considering the calendar year 1998 from the ICR, bromide concentrations at Sites E and F were significantly higher in 2010 than in 1998 ().
Fig. 1 shows quarterly bromide concentrations in the source water [Fig. 1(a)] and averaged quarterly concentrations of at Site D in the finished water [Fig. 1(b)]. The results for all sites were similarly assessed (Figs. S1 –S5 ). The source-water bromide concentration increased across all four quarters in 2010 and then decreased in 2011 and 2012. Flow is seasonally variable in this basin, with low flow in the summer (Quarter 3), and the bromide concentration would be expected to show similar seasonality (with highest concentration during low flow) if its load was constant. However, changes in management of discharging oil and gas wastewaters in the region likely account for the decreasing concentration trend after 2010 across all seasons (Wilson and VanBriesen 2013). In general, [Fig. 1(b)] shows expected seasonal behavior with higher values in the summer months (third quarter) when warmer water temperature increases DBP formation rates and higher chlorine doses are applied to ensure adequate residual chlorine in the distribution system (Clement et al. 2006). These quarterly values would be smoothed out if RAAs or OELs were computed for regulatory compliance assessment. Because these were finished-water samples rather than distribution system samples, this was not done, and these results cannot be used to assess the compliance status of these drinking water plants.
EE.1943-7870.0001175/asset/b8d7051a-8fcd-4afe-879e-8a7f43bad1b5/assets/images/large/figure1.jpg)
Shifting bromination patterns in DBPs are clearly visible, which would not be expected if source-water bromide levels were stable. In 2010, chloroform remained constant rather than increase seasonally as it did in 2011 and 2012. The increased BDCM, DBCM, and bromoform in the summer of 2010, caused by the increase in bromide in the source water (greater than ), likely accounts for the lower chloroform than would be expected. Fig. 1 shows that the quarters with greater than 80 μg/L show higher fractions of brominated THM (e.g., third and fourth quarters of 2010). The other sites in the basin (Figs. S1 –S5 ) show similar patterns; however, significant variability is seen. These results are expected because more rapid formation of brominated THM reduces the available organic carbon for chloroform formation, leading to higher overall brominated THM. The reaction of available bromide and hypochlorite acid (HOCl) producing hypobromous acid (HOBr) during treatment of drinking water is fast and irreversible (WHO 2004). After oxidization, HOBr is 25 times more reactive than HOCl (Brown 2009; Chang et al. 2001). In the distribution system, however, when excess chlorine is present while available bromide is low, chloroform will be dominantly produced among the THM species (Brown 2009).
Similarly, quarterly HAAs and quarterly bromide concentrations at Site D are shown in Fig. 2; other sites with similar results are shown in Figs. S6 –S10 . Fig. 2 shows that early in 2010, increasing bromide causes increasing and in finished water. However, in the second, third, and fourth quarters of 2010, no significant changes were observed in concentrations, whereas were higher in the third quarter than either the second or fourth quarters. When bromide decreased in the first quarter of 2011, the unregulated concentrations decreased significantly. However, unregulated concentrations in the second and third quarters of 2011 were again high despite lower source-water bromide. Across all the sites, bromide had a moderate association with increasing (), and a weak association with the unregulated ().
EE.1943-7870.0001175/asset/4943e926-74c1-4862-9bf5-69d06f37c1ec/assets/images/large/figure2.jpg)
Assessing Bromination
Fig. 3 shows the bromination fraction of as BSF [Fig. 3(a)] and percent by mass [Fig. 3(b)] for all quarterly data plotted against quarterly bromide concentrations. As expected, bromide concentration is predictive of bromination fraction in (). If a target goal of bromination (as BSF) were set to 0.25, these results indicate that bromide concentrations in the Monongahela River would need to remain less than . Alternatively, considering bromination on a mass basis as suggested by the recent work by the PADEP (Handke 2009), a linear regression () enables the determination of an in-stream bromide concentration if a target bromination percentage is selected. For example, to meet a goal of 25% bromination by mass in , bromide would need to remain less than , whereas to meet a goal of 50% bromination, bromide would need to remain less than .
EE.1943-7870.0001175/asset/86dac933-7f22-4c4b-b1dc-ad86bd4f725f/assets/images/large/figure3.jpg)
There is no clear linear relationship between quarterly bromide and () or percentage mass brominated (data not shown). However, increasing bromide in source water leads to a decrease in the percentage mass of brominated HAA species (DBAA and MBAA) in the regulated () and leads to a decrease of () (Fig. S11 ). This is expected. Increasing source-water bromide has been observed to increase the molar concentration of (Pourmoghaddas et al. 1993; Wu and Chadik 1998), whereas decreasing the molar concentration of (Hua and Reckhow 2006), which leads to a reduction in and in the regulated (measured by mass) because less brominated HAA weighs less.
These results also suggest that increasing source-water bromide in the Monongahela River favors increasing bromine incorporation in THM rather than in HAA. Similar results were observed by Sun et al. (2009), who report more bromine incorporation into THM compared with HAA at higher bromide levels (Sun et al. 2009).
THM Risk Analysis
Fig. 4 and Table S4 show the cancer risk level computed for each THM species for Site D in each quarter. The additive nature of the risk values computed allows the stacked results to represent the total risk by using this method. Although Fig. 1 shows that chloroform had the highest concentration among THM species in finished water in most quarters at Site D, Fig. 4 indicates that the risk related to chloroform only accounts for a small portion (8–19%) of risk. Fig. 4 shows that BDCM and DBCM together generally dominate the risk profile, ranging from 0% (when both were undetectable in Quarter 1 of 2012) to 95% in Quarter 4 of 2010. Similar results to those at Site D were observed at other sites (Figs. S12 –S16 ). Although DBCM concentrations were significantly lower than chloroform concentrations (), the calculated risk of DBCM is not significantly different from that of chloroform (), highlighting the importance of DBCM as a risk driver. Bromoform accounted for the smallest portion of the risk of because of its very low concentration in finished water throughout the study period and across the different drinking water plants. Although values suggest that finished-water quality was less than 80% of the MCL most of the time, Fig. 4 shows that increasing brominated THM species is associated with significant increased risk even while meeting this finished-water target.
EE.1943-7870.0001175/asset/1aa29106-ec8a-4286-bcb4-ac8a01336ee8/assets/images/large/figure4.jpg)
Monte Carlo Simulation Results
To assess the effect of increasing bromide concentrations to the risk of THM, the THM risk from six drinking water plants at different bromide concentration ranges (e.g., 10–20, 20–40,…, 100–120, ) were evaluated by using a Monte Carlo simulation. Fig. 5 shows the empirical CDF of the concentration [Fig. 5(a)] and the empirical CDF of the risk [Fig. 5(b)] for finished water when source-water bromide concentrations ranged from 20 to . Fig. 5 shows that 77% of finished-water samples were less than , and 72% met the finished-water goal of 80% of the MCL (). Fig. 5 also shows that only 40% of finished-water samples met a target of one person with cancer in 100,000 people. Thus, while the finished-water goal is likely to be met when bromide concentrations are less than , a target risk of is not likely to be met. The empirical CDFs of concentrations and oral cancer risk in higher bromide ranges were also computed (see results in Table S4 ).
EE.1943-7870.0001175/asset/1486c3d5-aa6b-4d2e-8016-f668de1ff4c6/assets/images/large/figure5.jpg)
With all the concentration ranges and sites considered, the likelihood of meeting a target risk of and meeting the TTHM standard (and 80% of the standard) are plotted in Fig. 6. Fig. 6 shows that the probability of meeting the finished-water goal at all bromide levels was significantly higher than the probability of meeting the target risk (). The probability decreased when bromide concentrations in source water increased. The probability of meeting the TTHM standard remained high (0.7–0.86) when bromide concentrations were less than and decreased to 0.61 when bromide concentrations were more than . The probability of meeting an 80% TTHM MCL value in the water leaving the plant was 0.68 when bromide concentrations were less than and decreased to 0.54 when bromide concentrations were more than . The probability of meeting the target cancer risk was always lower and generally decreased when bromide increased because of the increase in brominated THM that have higher unit risk values. The probability of meeting the target cancer risk decreased from 0.4 when bromide ranged from 20 to , to 0.04 when bromide exceeded . Fig. 6 shows that although the probability of meeting the target risk was always less than 0.4, the finished water had a high likelihood (0.7–0.86) of meeting the target finished-water concentration () when source-water bromide concentrations were less than . To increase the likelihood of meeting the risk target, the bromide concentration in the source water would have to be much lower than the level necessary to ensure meeting the finished-water goal. Thus, although the finished water may be meeting the goal most of the time, the risk associated with consuming this water is higher than it would be if the source water contained lower bromide levels.
EE.1943-7870.0001175/asset/b26e7e78-201b-47a9-bec1-3fb57e404c4a/assets/images/large/figure6.jpg)
Conclusion
Drinking water plants on the Monongahela River show variable responses to source-water bromide; however, to prevent adverse health effects associated with brominated THMs at all these drinking water plants, the bromide concentrations in this source water must be very low. Although organic precursors were observed to be stable during this study (measured as TOC and UV254), it is possible that the nature of the organic matter changed in a manner that affected DBP concentrations.
Regression analysis of bromination factors, combined with human health risk assessment, were used successfully to evaluate acceptable bromide concentrations in the Monongahela River. Similar methods may be used in other areas experiencing elevated bromide concentration in source water. To protect consumers of chlorinated drinking water, in-stream bromide concentration should be monitored, and discharges of bromide-containing wastewaters to surface water should be reduced where they are affecting drinking water sources. Identification of bromide discharges, proximity to drinking water intakes, and seasonal flow conditions in the river should all be considered in evaluating methods to control source-water bromide to reduce risks associated with brominated THMs in finished water.
Supplemental Data
Supplemental Materials
File (supplemental_data_ee.1943-7870.0001175_wang.pdf)
- Download
- 4.58 MB
Acknowledgments
The authors gratefully acknowledge the financial support of the Colcom Foundation and the Heinz Endowments and the cooperation of multiple drinking water utilities and their operators and engineers, without whom the sampling could not have been accomplished. The authors appreciate the many undergraduate and graduate researchers at Carnegie Mellon University who assisted with data collection and analyses during the field study, including: Dr. Jessica Wilson, Dr. Sandra Karcher, Ms. Alyssa Downs, Ms. Amrit Kaur, Ms. Stacie Lackler, Mr. Juan Medina, Ms. Rachna Sharma, Ms. Swapna Sridharan, and Ms. Zheng Wang.
References
@RISK 7.5 [Computer software]. Palisade, Newfield, NY.
Amy, G., Siddiqui, M., Zhai, W., DeBroux, J., and Odem, W. (1994). Survey of bromide in drinking water and impacts on DBP formation, American Water Works Research Foundation, Denver.
Asante-Duah, K. (2002). Public health risk assessment for human exposure to chemicals, Springer, New York.
Becker, W., Stanford, B., and Rosenfeldt, E. (2013). “Guidance on complying with Stage 2 D/DBP regulation.”, Water Research Foundation, Denver.
Bergman, L., Wilson, J., Small, M., and VanBriesen, J. (2016). “Application of classification trees for predicting disinfection by-product formation targets from source water characteristics.” Environ. Eng. Sci., 33(7), 455–470.
Bond, T., Huang, J., Graham, N., and Templeton, M. (2014). “Examining the interrelationship between DOC, bromide and chlorine dose on DBP formation in drinking water—A case study.” Sci. Total Environ., 470–471, 469–479.
Brown, D. (2009). “The management of trihalomethanes in water supply systems.” ⟨http://etheses.bham.ac.uk/364/1/Brown09PhD.pdf⟩ (May 19, 2016).
CDPH (California Department of Public Health). (2014). “Maximum contaminant level (MCL) determination for trichloroethylene.” ⟨http://www.waterboards.ca.gov/drinking_water/certlic/drinkingwater/documents/mclreview/TCE-MCLevaluation-12-01.pdf⟩ (Mar. 1, 2016).
Chang, E. E., Lin, Y. P., and Chiang, P. C. (2001). “Effects of bromide on the formation of THMs and HAAs.” Chemosphere, 43(8), 1029–1034.
Chen, W. H., Haunschild, K., Lund, J., and Fleenor, W. (2010). “Current and long-term effects of delta water quality on drinking water treatment costs from disinfection by-product formation, San Franciso estuary and watershed.” ⟨https://watershed.ucdavis.edu/shed/lund/papers/Chen2010.pdf⟩ (Feb. 2, 2015).
Chowdhury, S., Rodriguez, M., and Sadiq, R. (2011). “Disinfection byproducts in Canadian provinces: Associated cancer risks and medical expenses.” J. Hazard. Mater., 187(1–3), 574–584.
Clement, J., et al. (2006). “Predictive models for water quality in distribution systems.”, International Water Association (IWA), London.
DeGroot, M., and Schervish, M. (2012). Probability and statistics, Wiley, Boston.
DiCosmo, B. (2012). “EPA faces competing calls to revise ‘diesel’ definition in fracking guide.” ⟨http://insideepa.com/⟩ (May 12, 2015).
Diehl, A., Speitel, G., Symons, J., Krasner, S., Hwang, C., and Barrett, S. (2000). “DBP formation during chloramination.” J. Am. Water Works Assoc., 92(6), 76–90.
Driedger, M., Eyles, J., Elliott, S., and Cole, D. (2002). “Constructing scientific authorities: Issue framing of chlorinated disinfection byproducts in public health.” Risk Anal., 22(4), 789–802.
Echigo, S., Itoh, S., Natsui, T., Araki, T., and Ando, R. (2004). “Contribution of brominated organic disinfection by-products to the mutagenicity of drinking water.” Water Sci. Technol., 50(5), 321–328.
Faust, R. (1992). “Condensed toxicity summary for chloroform.” ⟨https://rais.ornl.gov/tox/profiles/chlorofo.html⟩ (Dec. 21, 2015).
Flury, M., and Papritz, A. (1993). “Bromide in the natural environment: Occurrence and toxicity.” J. Environ. Qual., 22(4), 747–758.
Gelderloos, A. B., Harrington, G. W., Owen, D. M., Regli, S., and Schaefer, J. K. (1992). “Simulation of compliance choices for the disinfection by-products regulatory impact analysis.”, EPA, Washington, DC.
Gould, J., Fitchhorm, P., and Urheim, E. (1983). “Formation of brominated trihalomethanes: Extent and kinetics.” Water chlorination: Environmental impact and health effects, R. L. Jolley, et al., eds., Vol. 4, Ann Arbor Science, Ann Arbor, MI, 297–310.
Grayman, W., Speight, V., and Uber, J. (2008). “Using Monte-Carlo simulation to evaluate alternative water quality sampling plans.” Proc., Water Distribution Systems Analysis Symp. 2006, ASCE, Reston, VA1–11.
Grellier, J., Rushton, L., Briggs, D. J., and Nieuwenhuijsen, M. J. (2015). “Assessing the human health impacts of exposure to disinfection by-products—A critical review of concepts and methods.” Environ. Int., 78, 61–81.
Greune, A. (2014). “Bromide occurrence in North Carolina and the relationship between bromide concentration and trihalomethane formation.” M.S. thesis, North Carolina State Univ., Raleigh, NC.
Handke, P. (2009). “Trihalomethane speciation and relationship to elevated total dissolved solid concentrations affecting drinking water quality at system utilizing the Monongahela River as a primary source during the 3rd and 4th quarters of 2008.” Water Program Specialist, Pennsylvania Dept. of Environmental Protection Bureau of Water Standards and Facility Regulation, Harrisburg, PA.
Heeb, M., Criquet, J., and Zimmermann, S. (2014). “Oxidative treatment of bromide-containing waters: Formation of bromine and its reactions with inorganic and organic compounds—A critical review.” Water Res., 48(1), 15–42.
Hellergrossman, L., Manka, J., Limonirelis, B., and Rebhun, M. (1993). “Formation and distribution of haloacetic acids, THM, and TOX in chlorination of bromide-rich lake water.” Water Res., 27(8), 1323–1331.
Henry, H. (2013). “EPA’s Stage 2 disinfection byproducts rules (DBPR) and Northern Kentucky water: An economic and scientific review.” Dose-Response, 11(4), 517–542.
Holm, L., Harader, S., and Femandez, P. (2007). “CALFED water quality program Stage 1 final assessment.” ⟨http://www.calwater.ca.gov/content/Documents/Draft_Final.pdf⟩ (Feb. 16, 2016).
Hrudey, S. (2008). “Chlorination disinfection by-products (DBPs) in drinking water and public health in Canada, a primer for public health practitioners reviewing evidence from over 30 years of research.” ⟨http://www.ncceh.ca/sites/default/files/Chlorination_Disinfection_By-Products_2008.pdf⟩ (Mar. 5, 2015).
Hrudey, S., et al. (2015). “Evaluating evidence for association of human bladder cancer with drinking water chlorination disinfection by-products.” J. Toxicol. Environ. Health Part B, 18(5), 213–241.
Hrudey, S., and Charrois, J. (2012). Disinfection by-products and human health, International Water Association (IWA), London.
Hua, G., and Reckhow, D. (2006). “Effect of bromide and iodide ions on the formation and speciation of disinfection byproducts during chlorination.” Environ. Sci. Technol., 40(9), 3050–3056.
Hua, G., and Reckhow, D. (2012). “Evaluation of bromine substitution factors of DBPs during chlorination and chloramination.” Water Res., 46(13), 4208–4216.
Huy, T. B., Tuyet-Hanh, T., Johnston, R., and Ngyyen-Viet, H. (2014). “Assessing health risk due to exposure to arsenic in drinking water in Hanam Province, Vietnam.” Int. J. Environ. Res. Public Health, 11(8), 7575–7591.
Kim, H., and Han, K. (2011). “Ingestion exposure to nitrosamines in chlorinated drinking water.” Environ. Health Toxicol., 26(1), 1–7.
Krasner, S., et al. (2008). “Bromine incorporation in regulated and emerging DBPs and the relative predominance of mono-, di- and trihalogenated DBPs.” Water Quality Technology Conf. and Exposition, American Water Works Association, Denver, 1692–1708.
Krasner, S., Gartrell, G., Gaston, J., Hoag, L., and Wadlow, W. (1991). “Delta water quality: A report to the legislature on trihalomethanes and the quality of drinking water available from the Sacramento-San Joaquin Delta.” ⟨http://www.water.ca.gov/pubs/waterquality/mwqi_reports/delta_water_quality__a_report_to_the_legislature_on_trihalomethanes_and_quality_of_drinking_water_available_from_the_sacramento-san_joaquin_delta/delta_water_quality_a_report_to_the_legislature_on_trihalomethanes_and_quality_of_drinking_water_available_from_the_sacramento-san_joaquin_delta_-_october_1991.pdf⟩ (Jan. 21, 2016).
Landis, M., Kamal, A., Kovalcik, K., Croghan, D., Norris, G., and Bergdale, A. (2016). “The impact of commercially treated oil and gas produced water discharges on bromide concentrations and modeled brominated trihalomethane disinfection byproducts at two downstream municipal drinking water plants in the Upper Allegheny River, Pennsylvania, USA.” Sci. Total Environ., 542, 505–520.
Luong, T., Peters, C., Young, R., and Perry, R. (1980). “Bromide and trihalomethanes in water supplies.” Environ. Technol. Lett., 1(6), 299–310.
McGuire, M., Hotaling, M., and Obolensky, A. (2002). “Information collection rule data analysis.” ⟨http://www.waterrf.org/publicreportlibrary/90947.pdf⟩ (Mar. 12, 2015).
McLain, J., Obolensky, A., and Shukairy, H. (2002). Disinfection byproduct speciation. Information collection rule data analysis, American Water Works Association Research Foundation, Denver, 141–168.
Minitab 17 [Computer software]. Pennsylvania State Univ., State College, PA.
Mishra, K., Kumar, S., and Sinha, A. (2014). “Human health risk analysis from disinfection by-products (DBPs) in drinking and bathing water of some Indian cities.” J. Environ. Health Sci. Eng., 12(73), 1–10.
Obolensky, A. (2007). “Analysis of information collection rule data to assess the impact of water quality and treatment on disinfection byproduct occurrence in drinking water.” ⟨https://cdr.lib.unc.edu/indexablecontent /uuid:4dedb159-25fd-4608-9871-c95b7f668497⟩ (Dec. 4, 2014).
Obolensky, A., and Singer, P. C. (2005). “Halogen substitution patterns among disinfection byproducts in the information collection rule database.” Environ. Sci. Technol., 39(8), 2719–2730.
Pawelczyk, A. (2013). “Assessment of health risk associated with persistent organic pollutants in water.” Environ. Monit. Assess., 185(1), 497–508.
Plewa, M. J., Wagner, E. D., and Jazwierska, P. (2004). “Halonitromethane drinking water disinfection byproducts: Chemical characterization and mammalian cell cytotoxicity and genotoxicity.” Environ. Sci. Technol., 38(1), 62–68.
Pourmoghaddas, H., Stevens, A., Kinman, R., Dressman, R., Moore, L., and Ireland, J. (1993). “Effect of bromide ion on formation of HAAs during chlorination.” J. Am. Water Works Assoc., 85(1), 82–87.
Regli, S., et al. (2015). “Estimating potential increased bladdar cancer risk due to increased bromide concentrations in sources of disinfected drinking waters.” Environ. Sci. Technol., 49(22), 13094–13102.
Rice, G., Teuschler, L., Bull, R., Simmons, J., and Feder, P. (2009). “Evaluating the similarity of complex drinking-water disinfection by-product mixtures: Overview of the issues.” J. Toxicol. Environ. Health Part A, 72(7), 429–436.
Richardson, S. D., Plewa, M. J., Wagner, E. D., Schoeny, R., and DeMarini, D. M. (2007). “Occurrence, genotoxicity, and carcinogenicity of regulated and emerging disinfection by-products in drinking water: A review and roadmap for research.” Mutat. Res., 636(1–3), 178–242.
Roberson, A., et al. (1995). “The D/DBP rule: Where did the numbers come from?” J. Am. Water Works Assoc., 87(10), 46–57.
Sigmaplot 13 [Computer software]. Systat Software, San Jose, CA.
Singer, P. C. (1994). “Control of disinfection by-products in drinking water.” J. Environ. Eng., 727–744.
Sollars, C. J., Peters, C. J., and Perry, R. (1982). “Bromide in urban runoff—Water quality considerations.” Effects of Waste Disposal on Groundwater and Surface Water (Proc., Exeter Symp.), International Association of Hydrological Sciences (IAHS), Wallingford, U.K., 101–112.
States, S., et al. (2013). “Marcellus shale drilling and brominated THMs in Pittsburgh, PA, drinking water.” J. Am. Water Works Assoc., 105(8), E432–E448.
Sun, Y.-X., Wu, Q.-Y., Hu, H.-Y., and Tian, J. (2009). “Effect of bromide on the formation of disinfection by-products during wastewater chlorination.” Water Res., 43(9), 2391–2398.
Theodore, L., and Dupont, R. (2012). Environmental health and hazard risk assessment: Principles and calculations, CRC Press, Boca Raton, FL.
Travis, C., and Land, M. (1990). “Estimating the mean of data sets with nondetectable values.” Environ. Sci. Technol., 24(7), 961–962.
U.S. EPA. (1990). “Method 551.1: Determination of chlorination disinfection by-products and chlorinated solvents in drinking water by liquid-liquid extraction and gas chromatography with electron-capture detection.” Washington, DC.
U.S. EPA. (1997a). “Guiding principles for Monte Carlo analysis.”, Washington, DC.
U.S. EPA. (1997b). “Method 300.1: Determination of inorganic anions in drinking water by ion chromatography.” Washington, DC.
U.S. EPA. (1998). “National primary drinking water regulations: Disinfectants and disinfection byproducts: Final rule.”, Washington, DC, 69390–69476.
U.S. EPA. (2000a). “Supplementary guidance for conducting health risk assessment of chemical mixtures.”, Washington, DC.
U.S. EPA. (2000b). “The risk assessment of mixtures of disinfection by products (DBPs) in drinking water.”, Washington, DC.
U.S. EPA. (2000c). “US EPA ICR auxiliary 1 database version 5.0, EPA number 815C00002.” ⟨https://ntrl.ntis.gov/NTRL/dashboard/searchResults/titleDetail/PB2001500031.xhtml⟩ (Jan. 2, 2015).
U.S. EPA. (2003). “Method 552.3: Determination of haloacetic acids and dalapon in drinking water by liquid-liquid microextraction, dericatization, and gas chromatography with electron capture detection.”, Washington, DC.
U.S. EPA. (2005a). “Chapter 7: Risk and hazard characterization.” ⟨http://www.columbia.edu/itc/sipa/envp/louchouarn/courses/env-chem/Risk%20Characterization%20Region%206.pdf⟩ (Dec. 22, 2015).
U.S. EPA. (2005b). “Drinking water citeria document for brominated trihalomethanes.”, Washington, DC.
U.S. EPA. (2005c). “Economic analysis for the final Stage 2 disinfectants and disinfection byproducts rule.”, Office of Water, Office of Ground Water and Drinking Water, Washington, DC.
U.S. EPA. (2006). “National primary drinking water regulations: Stage 2 disinfectant and disinfection byproduct rule.”, Washington, DC.
U.S. EPA. (2009). “Drinking water standards and health advisories table.” ⟨https://www3.epa.gov/region9/water/drinking/files/dwshat-v09.pdf⟩ (May 6, 2015).
U.S. EPA. (2012). “Organic chemicals, sampling and analytical requirements.” ⟨https://www.gpo.gov/fdsys/pkg/CFR-2013-title40-vol24/pdf/CFR-2013-title40-vol24-sec141-24.pdf⟩ (Jun. 1, 2015).
U.S. EPA. (2014a). “Bromodichloromethane; CASRN 75-27-4.” ⟨https://cfpub.epa.gov/ncea/iris/iris_documents/documents/subst/0213_summary.pdf⟩ (Mar. 6, 2015).
U.S. EPA. (2014b). “Bromoform; CASRN 75-25-2.” ⟨https://cfpub.epa.gov/ncea/iris/iris_documents/documents/subst/0214_summary.pdf⟩ (Feb. 1, 2015).
U.S. EPA. (2014c). “Dibromochloromethane; CASRN 124-48-1.” ⟨https://cfpub.epa.gov/ncea/iris/iris_documents/documents/subst/0222_summary.pdf⟩ (Feb. 1, 2015).
U.S. EPA. (2015). “Effluent limitations guidelines and standards for the steam electric power generating point source category.” ⟨https://www.regulations.gov/#!documentDetail;D=EPA-HQ-OW-2009-0819-5558⟩ (Dec. 20, 2015).
Villanueva, C., Cordier, S., Font-Ribera, L., Salas, L., and Levallois, P. (2015). “Overview of disinfection by-products and associated health effects.” Curr. Environ. Health Rep., 2(1), 107–115.
Wang, G.-S., Deng, Y.-C., and Lin, T.-F. (2007). “Cancer risk assessment from trihalomethanes in drinking water.” Sci. Total Environ., 387(1–3), 86–95.
Weaver, J., Xu, J., and Mravik, S. (2015). “Scenario analysis of the impact on drinking water intakes from bromide in the discharge of treated oil and gas wastewater.” J. Environ. Eng., .
Wegman, R., Hamaker, P., and De Heer, H. (1983). “Bromide-ion balance of a polder district with large-scale use of methyl bromide for soil fumigation.” Food Chem. Toxicol., 21(4), 361–367.
Westerhoff, P., Chao, P., and Mash, H. (2004). “Reactivity of natural organic matter with aqueous chlorine and bromine.” Water Res., 38(6), 1502–1513.
WHO (World Health Organization). (1993). “Guidelines for drinking water quality: Recommendations—Addendum.” 2nd Ed., Vol. 1, Geneva ⟨http://www.who.int/water_sanitation_health/dwq/2edaddvol1.pdf⟩ (May 4, 2015).
WHO (World Health Organization). (2004). “Part 2: Chemistry of disinfectants and disinfectant by-products.” Disinfectants and disinfectant by-products, Geneva ⟨http://www.who.int/ipcs/publications/ehc/216_disinfectants_part_2.pdf⟩ (Feb. 2, 2015).
Wilson, J., and VanBriesen, J. M. (2013). “Source water changes and energy extraction activities in the Monongahela River, 2009–2012.” Environ. Sci. Technol., 47(21), 12575–12582.
Winid, B. (2015). “Bromine and water quality, selected aspects and future perspectives.” Appl. Geochem., 63, 413–435.
Wu, W., and Chadik, P. (1998). “Effect of bromide ion on haloacetic acid formation during chlorination of biscayne aquifer water.” J. Environ. Eng., 932–938.
Zhang, J., et al. (2011). “Characterization of disinfection byproduct formation potential in 13 source waters in China.” J. Environ. Sci., 23(2), 183–188.
Information & Authors
Information
Published In
Journal of Environmental Engineering
Volume 143 • Issue 3 • March 2017
Copyright
This work is made available under the terms of the Creative Commons Attribution 4.0 International license, http://creativecommons.org/licenses/by/4.0/.
History
Received: Mar 17, 2016
Accepted: Aug 5, 2016
Published online: Oct 31, 2016
Published in print: Mar 1, 2017
Discussion open until: Mar 31, 2017
Authors
Metrics & Citations
Metrics
Citations
Download citation
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.