Richard J. Weisman

Watersheds with relatively higher concentrations of disinfection by-product (DBP) precursors are of concern for public water systems (PWSs) seeking to control DBP exposure risk. In this study, the occurrence of bromide and total organic carbon (TOC), which are important DBP precursors, was evaluated in PWS source waters on a watershed basis at the hydrologic unit code 2 (HUC2) level. Variations were examined across source water type (e.g., surface water or groundwater) and watershed, and temporally based on seasonality as well as differences between the late 1990s and the years 2018–2020. The median bromide concentrations were higher in groundwater systems compared with surface water (by 25 μg/L), whereas the median TOC concentrations were higher in surface water systems compared with groundwater (by 1.7 mg/L). Surface water sources in multiple midwestern and western watersheds had a combination of elevated bromide and TOC (relative to median values of 21 μg/L and 2.5 mg/L, respectively), which was especially pronounced in the Texas-Gulf and Rio Grande watersheds. None of the HUC2 watersheds had elevated levels of both bromide and TOC in groundwater sources (median values of 50 μg/L and 1.0 mg/L, respectively). Source water bromide median watershed concentrations for very large systems were ∼20% lower in 2018–2020 than in 1998, however, more than 40% higher at the 90th percentile, whereas source water TOC median watershed concentrations for very large systems were ∼25% higher in 2018–2020 than in 1998 but less than 10% higher at the 90th percentile. These temporal differences may be related to the influence of climate change and other anthropogenic effects. In addition, many individual PWSs showed substantial change in bromide and TOC concentrations between those time periods. For example, about 7% of the PWSs had at least double the bromide concentrations, and 5% at least double the TOC concentrations, in 2018–2020 than in 1998.

Environmental Health Perspectives is an Open Access journal published by the National Institute of Environmental Health Sciences. Environmental Health Perspectives is an Open Access journal published by the National Institute of Environmental Health Sciences.

Temporal variations of de facto wastewater reuse are relevant to public drinking water systems (PWSs) that obtain water from surface sources. Variations in wastewater discharge flows, streamflow, de facto reuse, and disinfection by-products (DBPs – trihalomethane-4 [THM4] and haloacetic acid-5 [HAA5]) over an 18-year period were examined at 11 PWSs in the Shenandoah River watershed, using more than 25,000 data records, in gaged and ungaged reaches. The relationship of de facto reuse with DBPs by year and quarter at the PWSs was examined. A linear relationship was found between THM4 and de facto reuse on an annual average basis (p = 0.050), as well as in quarters 3 (July – September) (p = 0.032) and 4 (October – December) (p = 0.031). Using a t-test (p < 0.05), the study also showed that there were significant differences in DBP levels for PWSs relative to 1% de facto reuse. This was found for THM4 based on annual average and quarter 1 (January – March) data, and for HAA5 based on quarter 3 data during the period of record.

De facto reuse is increasingly being studied among the variety of stressors that are relevant to drinking water systems that obtain their source water from surface waters. De facto reuse may influence the levels and types of precursors relevant to formation of disinfection by-products (DBPs) in surface water systems. DBPs such as trihalomethanes (THMs) and haloacetic acids (HAAs) have been associated with bladder cancer and other health concerns in people who use drinking water provided by public water systems (PWSs). In this study, we used compliance monitoring data from conventional surface water PWSs in the Shenandoah River watershed to evaluate the relationship between de facto reuse in the watershed with DBP formation in those systems. The Shenandoah River watershed was selected for this study because it has a relatively small group of PWSs that draw their source water from surface waters in the watershed, the majority of whom treat their water using chlorine, and are less likely to have confounding factors (such as complex distribution systems or lengthy residence times) than other watersheds. We found that concentrations of THM4 and HAA5 increase in drinking water systems as de facto reuse increases in their source waters and that the relation is observed at both annual average and low streamflow conditions (annual average and low streamflow increases were statistically significant for THM4 with p values of 0.027 and 0,021, respectively). In addition, using a t-test, we found that a 1% level of de facto reuse was associated with significantly higher levels of THM4 and HAA5 (p < 0.05) under annual average streamflow conditions. While the concentrations of HAA5 were also higher under annual average and low streamflow conditions, we did not find that they achieved a level of a significant difference. Results from this research will be helpful to operators of PWSs and other researchers and stakeholders with an interest in water reuse and DBPs.