Suspect-screening analysis of a coastal watershed before and after Hurricane Florence using high-resolution mass spectrometry
Graphical abstract
Introduction
Hurricanes (Kossin et al., 2017; Melillo and Richmond, 2014) and extreme rainfall (Knight and Davis, 2009) are projected to increase in intensity and frequency as evidenced by recent events in coastal U.S. communities from Texas to North Carolina (Moore et al., 2015; Paerl et al., 2019; Prat and Nelson, 2016). These extreme events can impact the transport and distribution of organic chemical contaminants via the deposition of aerosolized contaminants (Noyes et al., 2009), the disturbance and resuspension of sediments in water (Fisher et al., 2016a; Van Metre et al., 2006), surface run-off, infiltration of flood waters to groundwater (Bonneau et al., 2017; Jasechko et al., 2017; Jasechko and Taylor, 2015), and overflows from wastewater treatment plants, industrial and agricultural waste lagoons, and septic systems (Fisher et al., 2016a; Fisher et al., 2016b). The resultant disruption of chemical reservoirs therefore merits further understanding of both the extent of chemical redistribution and potential risk to surface and groundwater resources.
To date, studies of hurricanes have focused on chemical disturbance of nutrients (Fisher et al., 2016b; Bales et al., 2000; Presley et al., 2006), metals (Van Metre et al., 2006; Presley et al., 2006; Fox et al., 2009; Pardue et al., 2005), polycyclic aromatic hydrocarbons (Fisher et al., 2016a; Van Metre et al., 2006; Bales et al., 2000; Fox et al., 2009; Pardue et al., 2005), pesticides (Bales et al., 2000), pharmaceuticals and hormones (Fisher et al., 2016a; Fisher et al., 2016b), and other organic contaminants of concern (COCs) (Fisher et al., 2016a; Van Metre et al., 2006; Presley et al., 2006) using targeted analytical methods. We utilized suspect screening analysis via high resolution mass spectrometry (HRMS) versus targeted analysis to evaluate a broader array of regulated and non-regulated chemicals present in watershed sources before and after Hurricane Florence. This approach involves qualitative screening for a variety of chemical features (defined as detected m/z, retention time, and abundance), in which spectral output is compared with compound-specific parameters contained within curated chemical databases for tentative identification (Aceña et al., 2015; Gosetti et al., 2016). Our aim was to use HRMS to explore chemical features and tentatively identified chemicals prior to and after extreme rainfall and flooding from Hurricane Florence. Established sampling locations provided a unique opportunity to compare chemical profiles in ground- and surface waters temporally and spatially across a watershed with diverse reservoirs of chemicals including a municipal wastewater land treatment site.
On September 14, 2018, the eye of Hurricane Florence passed 32 km southwest of the study watershed near Jacksonville, North Carolina. The City of Jacksonville Land Treatment Site (LTS; Fig. A.1) comprises a large portion of the watershed; it irrigates secondary treated municipal wastewater across 890 ha of a 2193 ha mixed hardwood and loblolly pine forest (Birch et al., 2016; Hedgespeth et al., 2019; McEachran et al., 2016). Two weeks prior to Hurricane Florence's arrival, the LTS initiated emergency spraying to lower wastewater reservoir levels. The LTS did not irrigate on September 14, 2018, when Hurricane Florence delivered 686 mm of rainfall to the site and raised groundwater elevation to within 0.04–0.45 m below ground surface (Figs. A.2, A.3). On September 15, 2018, the facility resumed operation and emergency spraying continued until April 2019 (Jacksonville Land Treatment Site, personal communication). Emergency spraying involves more frequent rotation of irrigation zones, thus increasing hydraulic and chemical loading for spray fields. Permit requirements (e.g. avoiding pooling of irrigation water on soil surface) must be ensured, and daily surface water quality monitoring is required for microbiological, nutrient, and select macro-chemical parameters. Prior hydrological measurements and isotopic modeling have shown that irrigated LTS wastewater sustains surface water baseflow during inter-storm periods and constitutes a large fraction of streamflow during storm response peaks via groundwater discharge and lateral subsurface flow (Birch et al., 2016). Peak stream flow during a 2015 tropical storm resulted in peak export of select pharmaceuticals from the LTS (McEachran et al., 2017).
While lagoons did not overflow and accidental discharge of wastewater did not occur due to the storm at the LTS, we expected the extreme rainfall of Hurricane Florence to mobilize chemical reservoirs from LTS soils and groundwater to receiving surface water. However, we were cognizant of similar dynamics in upstream regions of the watershed where land use is dominated by residential development and septic systems, agricultural row crops, and a municipal airport (Birch et al., 2016). Because coastal forests provide important ecosystem services through stormwater mitigation and groundwater recharge (Callahan et al., 2017), suspect-screening evaluation of waters at the LTS and the larger watershed provides unique insight to pre- and post-hurricane changes in COC release for waste-, ground-, and surface waters for the facility and surrounding watershed.
Section snippets
Sampling
Grab samples were collected from LTS wastewater effluent (post-chlorination; i.e. irrigation water) and from two groundwater monitoring wells on the LTS site (LTS wells A and B) and LTS surface water. Reference watershed groundwater was collected to the immediate west and hydrologically-upgradient of the LTS (off site) as well as surface waters upstream and downstream of the LTS (Fig. A.1) (Hedgespeth et al., 2019). Samples were collected prior to Hurricane Florence on June 10–11, 2018, as part
Results and discussion
We detected ~4000 unique chemical features in LTS wastewater consistently across sampling events (Fig. 1a). Non-targeted analysis does not permit direct knowledge of compound concentration, but QC indicated relative stability of injections and matrix effects; we therefore make qualitative comparisons of relative chemical abundance, as a proxy for chemical concentrations, among individual chemicals and chemical use categories. Total relative abundance of all chemical features was nearly 30%
CRediT authorship contribution statement
Melanie L. Hedgespeth: Conceptualization, Investigation, Formal analysis, Visualization, Writing – original draft, Writing – review & editing. James P. McCord: Investigation, Resources, Writing – review & editing. Katherine A. Phillips: Resources, Writing – review & editing. Mark J. Strynar: Resources, Writing – review & editing. Damian Shea: Conceptualization, Writing – review & editing. Elizabeth Guthrie Nichols: Conceptualization, Investigation, Visualization, Resources, Writing – review &
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.
Acknowledgments
The authors would like to thank the City of Jacksonville and the Jacksonville LTS for their assistance. Special thanks to Amy Palaschak, Martina Gonzalez Bertello, Nancy Gibson, and Joshua Pil. This study was funded by USDA NIFA Award # 2016-68007-25069.
This document has been reviewed by the U.S. Environmental Protection Agency, Office of Research and Development, and approved for publication. The views expressed in this article are those of the authors and do not necessarily represent the
References (36)
- et al.
Non-target screening and prioritization of potentially persistent, bioaccumulating and toxic domestic wastewater contaminants and their removal in on-site and large-scale sewage treatment plants
Sci. Total Environ.
(2017) - et al.
Stormwater infiltration and the ‘urban karst’ – a review
J. Hydrol.
(2017) - et al.
Comparison of wastewater-associated contaminants in the bed sediment of Hempstead Bay, New York, before and after Hurricane Sandy
Mar. Pollut. Bull.
(2016) - et al.
The impact of onsite wastewater disposal systems on groundwater in areas inundated by Hurricane Sandy in New York and New Jersey
Mar. Pollut. Bull.
(2016) - et al.
Suspect screening and prioritization of chemicals of concern (COCs) in a forest-water reuse system watershed
Sci. Total Environ.
(2019) - et al.
Identification of novel micropollutants in wastewater by a combination of suspect and nontarget screening
Environ. Pollut.
(2014) - et al.
Pharmaceuticals in a temperate forest-water reuse system
Sci. Total Environ.
(2017) - et al.
The toxicology of climate change: environmental contaminants in a warming world
Environ. Int.
(2009) - et al.
Data-driven prioritization of chemicals for various water types using suspect screening LC-HRMS
Water Res.
(2016) - et al.
Environmental forensics in groundwater coupling passive sampling and high resolution mass spectrometry for screening
Sci. Total Environ.
(2016)
Advances in liquid chromatography–high-resolution mass spectrometry for quantitative and qualitative environmental analysis
Anal. Bioanal. Chem.
Hydrologic impacts of municipal wastewater irrigation to a temperate forest watershed
J. Environ. Qual.
Coastal forests and groundwater: using case studies to understand the effects of drivers and stressors for resource management
Sustainability
Characterization of storm flow dynamics of headwater streams in the South Carolina Lower Coastal Plain1
JAWRA Journal of the American Water Resources Association
Potential for chemical mixture exposures and health risks in New Orleans Post-Hurricane Katrina
Hum. Ecol. Risk. Assess.
Extended suspect and non-target strategies to characterize emerging polar organic contaminants in raw wastewater with LC-HRMS/MS
Environmental Science & Technology
Contaminants in water: non-target UHPLC/MS analysis
Environ. Chem. Lett.
Cited by (8)
Vulnerability of wells in unconfined and confined aquifers to modern contamination from flood events
2023, Science of the Total EnvironmentTarget and suspect analysis of liquid crystal monomers in soil from different urban functional zones
2023, Science of the Total EnvironmentCitation Excerpt :Therefore, it is necessary to determine whether novel LCMs are present in the environment. High-resolution mass spectrometry (HRMS) has been applied to explore unpredicted contaminants in various environments (Aalizadeh et al., 2021; Baesu et al., 2021; Caballero-Casero et al., 2021; Hedgespeth et al., 2021; Gong et al., 2022; Ren and Su, 2022; Ye and Su, 2022), due of its high mass-spectral resolving power, wide mass range, and high mass accuracy. Based on the above background, we combined the target and suspect analysis techniques, in allusion applied them to 39 target LCMs and 1173 suspect LCMs (Su et al., 2022), which were used for the synthetic estimation of the contamination status of LCMs in soil samples collected from five different functional zones.
Non-target and suspect-screening analyses of hydroponic soybeans and passive samplers exposed to different watershed irrigation sources
2022, Science of the Total EnvironmentCitation Excerpt :Because the land treatment system stores wastewater in large open reservoirs for 10–14 days and these reservoirs receive rainfall, consistent rainfall events can dilute the number of chemical features in wastewater to equivalent numbers detected in upstream surface waters (Hedgespeth et al., 2021). Aggregated ToxPI scores of TICs from multi-sampling campaigns were similar for chemicals of greatest abundance between upstream surface water samples and wastewater samples (Hedgespeth et al., 2021). After a catastrophic hurricane in 2018, the number of chemical features and total abundance in upstream surface water and off-site groundwater samples far exceeded those numbers determined for on-site and downstream surface waters and groundwater at two week and three months beyond the hurricane event (Hedgespeth et al., 2021).