Suspect screening and prioritization of chemicals of concern (COCs) in a forest-water reuse system watershed

Abstract

Much research has assessed organic chemicals of concern (COCs) in municipal wastewater and receiving waters, but few studies have examined COCs in land treatment systems. Many prior studies have implemented targeted methods that quantify a relatively small fraction of COCs present in wastewater and receiving waters. This study used suspect screening to assess chemical features in ground- and surface waters from a watershed where secondary-treated wastewater is irrigated onto 900 ha of temperate forest, offering a more holistic view of chemicals that contribute to the exposome. Chemical features were prioritized by abundance and ToxPi scoring across seasonal sampling events to determine if the forest-water reuse system contributed to the chemical exposome of ground- and surface waters. The number of chemical features detected in wastewater was usually higher than on- and off-site ground- and surface waters; in wastewater, chemical features trended with precipitation in which greater numbers of features were detected in months with low precipitation. The number of chemical features detected in off- and on-site waters was similar. The lower overlap between chemical features found in wastewater and downstream surface waters, along with the similar numbers of features being detected in upstream and downstream surface waters, suggests that though wastewater may be a source of chemicals to ground and surface waters on-site, dissipation of wastewater-derived features (in number and peak area abundance) likely occurs with limited off-site surface water export by the forested land treatment system. Further, the numbers of features detected on site and the overlap between wastewater and surface waters did not increase during periods of low rainfall, counter to our initial expectations. The chemical features tentatively identified in this watershed appear common to features identified in other studies, warranting further examination on the potential for resulting impacts of these on humans and the environment.

Keywords: Exposome; Forest; HRMS; Suspect screening; Wastewater.

Figure 1.

Map of the Jacksonville, NC, Land Treatment Site (LTS) showing wastewater (WW), groundwater (GW), and surface water (SW) sampling locations, along with soil types. Areas outlined in black indicate irrigated regions of the site. Arrows indicate general directional flow of water at the site.

Figure 2.

Mean number of chemical features detected in wastewater effluent and a.) groundwater and b.) surface waters, along with c.) cumulative precipitation and irrigation for each sampling event. Error bars represent the mean ± one standard deviation.

Figure 3.

Box-and-whisker plot of numbers of chemical features detected in groundwater samples across sampling events based on soil type with a description of soil drainage ranging from “well-drained” to “very poor drainage.”

Figure 4.

Top chemical features contributing to >1% of total abundance of each sampling event. Features with names had a match score of ≥90% and were tentatively identified as the top hit in the US EPA’s CompTox Chemicals Dashboard; those with “(n top hits)” listed before the chemical formula indicates that there was more than one top hit. Features listed as chemical formulae only had a match score of <90%, and the remaining are listed as “mass@retention time,” i.e. could not be assigned a chemical formula. Chemical features are ordered from most to least prevalent over sampling events (left to right), i.e. the first 14 listed were detected across >1 sampling event, and sub-ordered by total abundance. Note: there was only one hit in the CompTox Dashboard for two features both tentatively identified as hexamethylcyclotrisiloxane, labeled (a) and (b).

Figure 5.

ToxPi scores calculated across sampling events for tentatively identified chemical features within site types (score <1: black; ≥1: orange; ≥2: red; ≥3:dark red). Chemical features are listed in order from largest summed ToxPi scores (top) to lowest (bottom). *Features making up ≥1% of total abundance in one or more sampling events, also indicated by bold text. Note: there was only one hit in the CompTox Dashboard for the following chemical formulae, therefore two features were tentatively identified as: hexamethylcyclotrisiloxane, labeled (a) and (b); and octamethylcyclotetrasiloxane, labeled (a) and (b). Valsartan was detected in both ESI-positive and ESI-negative modes in the same samples, labeled (+) and (−), respectively.