Pathways for Potential Exposure to Onshore Oil and Gas Wastewater: What We Need to Know to Protect Human Health

3.1.1. Geochemical Properties

Studies from several oil‐ and gas‐producing regions reported on the chemical composition of produced‐water quality (Figure 3 and Table S2 in Supporting Information S1). Produced water contains varying levels of salinity (approximately five to 10 times higher than seawater in some locations), total dissolved solids (TDS), dissolved organic matter, total suspended solids, metals, metalloids such as arsenic, critical minerals, volatile and semi‐volatile organic compounds (VOCs and SVOCs) including BTEX (benzene, toluene, ethylbenzene, and xylene) and PAHs (polycyclic aromatic hydrocarbons), naturally occurring radioactive material, ammonia, chemical additives such as surfactants and biocides, and per‐ and polyfluoroalkyl substances (Akob et al., 2015; Al‐Ghouti et al., 2019; Chen et al., 2023; Dresel & Rose, 2010; Gallegos et al., 2021; Jiang et al., 2022; Kumar et al., 2019; Lester et al., 2015; Liden et al., 2022; Luek & Gonsior, 2017; McDevitt et al., 2022; Murali Mohan, Hartsock, Hammack, et al., 2013, Murali Mohan, Hartsock, Hammack, et al., 2013; Orem et al., 2014; Rosenblum, Nelson, et al., 2017; Schreiber & Cozzarelli, 2021; Sun et al., 2019; Thakur et al., 2022; U.S. Geological Survey, 2022; Ziemkiewicz, 2013). In some instances, specific constituents in produced water reportedly exceeded U.S. Environmental Protection Agency (U.S. Environmental Protection Agency, 2020) drinking water standards (Chittick & Srebotnjak, 2017; Ziemkiewicz & He, 2015).

Certain properties, including ion ratios, isotopic compositions (e.g., ⁸⁷Sr/⁸⁶Sr), and the concentration and type of halogenated organic compounds, can serve as tracers to distinguish the influence of injected fluid versus formation water, produced water from different types of energy development (e.g., UOGD and COGD), and wells of varying ages (Barnaby et al., 2004; Cantlay et al., 2020; Goldberg & Griffith, 2017; Luek & Gonsior, 2017; Ogbuji et al., 2022; Osborn & McIntosh, 2010; Peterman et al., 2012; Phan et al., 2018; Sirivedhin & Dallbauman, 2004; Warner et al., 2014). Elevated concentrations of sodium, chloride, barium, strontium, lithium, strontium and radium isotopic compositions, and trace hydrocarbons were determined to be key markers of a large pipeline spill of produced water in North Dakota (Cozzarelli et al., 2017, 2021; Lauer et al., 2016). Strontium (Sr) isotopic compositions (⁸⁷Sr/⁸⁶Sr) can provide robust tracers of salinity sources, such as produced water, through the environment due to their conservative environmental behavior (Chapman et al., 2012; Cozzarelli et al., 2016; Harkness et al., 2017; McDevitt, McLaughlin, et al., 2020; Warner et al., 2014). Studies have indicated the ability to utilize ⁸⁷Sr/⁸⁶Sr to identify and quantify the accumulation of produced‐water contaminants in both mussel shells and tissues even with the application of low doses of produced‐water exposure (i.e., 0.15%) (Geeza et al., 2018; McDevitt, Geeza, et al., 2021). These data can be used to differentiate sources of contaminants that originate from spills and leaks of produced water from legacy contaminants from other oil and gas and industrial activities. This source apportionment is essential to evaluating the relative contribution from multiple potential routes of exposures. Thus, isotopic tracers such as ⁸⁷Sr/⁸⁶Sr can provide a unique identifier of potential produced‐water exposure pathways to humans.

The complex composition of produced water creates unique challenges to measuring chemical components, treatability for potential use, and understanding effects once released to the environment. The high salinity of produced water can make chemical analysis and treatment difficult. For example, salinity can be an obstacle to the accurate measurement of organic and inorganic constituents because methods are typically designed and optimized for surface water or groundwater that has relatively lower salinity. Consequently, even where produced‐water chemical characterization data exist, it may be incomplete or even unreliable, driving the need for method development to overcome these challenges (Jubb et al., 2020; Oetjen et al., 2017; Santos et al., 2019; Tasker et al., 2019). Furthermore, produced water's high salinity levels can foul treatment membrane systems and can be toxic to microorganisms commonly used in biological treatment systems (Acharya et al., 2020; Akyon et al., 2018). Constituents in produced water can affect the rate of biodegradation necessary to remove organic material (Akyon et al., 2018; Tinker et al., 2020, 2022). Studies recommend various approaches to address these challenges, including hybrid treatment approaches to address the complex assemblage of constituents (Acharya et al., 2020; Akyon et al., 2018; Hildenbrand et al., 2018; Tinker et al., 2020, 2022) and co‐treatment of abandoned mine drainage and produced water to manage radioactivity (He et al., 2016; McDevitt, Cavazza, et al., 2020; Ouyang et al., 2019).

3.1.2. Variability in Composition

The composition of produced water varies as a function of the geologic formations from which the oil or gas is sourced, the composition of the oil or gas resource, the amount of time since production began, the type of drilling, the composition and volume of hydraulic fracturing fluid used to complete the well, and well maintenance procedures (U.S. Environmental Protection Agency, 2020). For example, Figure 4a demonstrates the wide variability of produced‐water composition by displaying concentrations of TDS, which can range over six orders of magnitude across the U.S., and lithium across shale plays and geological basins (Blondes et al., 2019; Nicot et al., 2018; Wang, Lu, et al., 2019). Figure 4b shows known carcinogens radium‐226 (Ra‐226) (n = 727 ranging from non‐detect to 16,920 pCi/L) and arsenic (n = 539 ranging from non‐detect to 7 mg/L) concentrations in contiguous U.S. produced water. Highest Ra‐226 concentrations measured in produced‐water samples were collected from the Appalachian Basin while highest arsenic concentrations were collected from the Arkla Basin. Figure 4b additionally provides visual perspective on the sparsity of select trace element data important for understanding human exposure compared to the plethora of produced‐water sample data available for more commonly measured inorganic salt species such as TDS.

Produced‐water composition can also vary between samples retrieved from a single well, shale play, or basin (Abualfaraj et al., 2014; Akob et al., 2015; Blauch et al., 2009; Chaudhary et al., 2019; Goldberg & Griffith, 2017; Jiang et al., 2021; Kim et al., 2016; Macpherson, 2015; McMahon et al., 2018; Rowan et al., 2015; Varonka et al., 2020; Welch et al., 2021). Produced‐water sampling often occurs at a variety of collection points besides a specific wellhead or separator, and so can include an aggregated mixture of produced water from multiple wells and/or geologic units such as storage tanks situated on well pads (Mackey et al., 2024) or prior to injection into saltwater disposal wells (Jiang et al., 2022). Concentrations of several constituents can also vary as the ratio of formation water and injected fluid increases over the lifespan of a well (Hayes, 2009; Oetjen et al., 2018; Oetjen & Thomas, 2016). Part of this variability comes from the different hydraulic fracturing fluid additives used for the specific environmental conditions of a given well and how those additives react in the subsurface environment (Kim et al., 2019; Ziemkiewicz & He, 2015). Investigations continue to identify hydraulic fracturing fluid additives in produced water and whether they can be used as tracers in investigations of produced water as potential contamination sources (Lester et al., 2015; Nell & Helbling, 2019; Rosenblum, Thurman, et al., 2017; Sitterley et al., 2018; Thacker et al., 2015; Thurman et al., 2017).

3.1.3. Microbiological Characteristics

Microbial activity in produced water can cause corrosion of well equipment and potentially lead to environmental releases (Booker et al., 2017; Chilkoor et al., 2018; Daly et al., 2016; Liang et al., 2016; Lipus et al., 2017; Vikram et al., 2016). Microbial composition of produced water changes over time (Lipus et al., 2018) and can resist the efficacy of biocidal and disinfection processes, which can compromise well casing and grout integrity (Cluff et al., 2014; Hildenbrand et al., 2018; Struchtemeyer & Elshahed, 2012). Under thermophilic and anaerobic conditions, microbes can also metabolize hydrocarbons (Gieg et al., 2010) and polysaccharides in hydraulic fracturing fluid additives (Liang et al., 2016; Wang, Lu, et al., 2019). Pretreatment to mitigate solid formation and microbial activity before storage can help prevent corrosion and scaling (Murali Mohan, Hartsock, Bibby, et al., 2013; Thiel & Lienhard V, 2014).

3.1.4. Produced Water Toxicology

A subset of the literature focused on toxicological effects of produced water, including original research and reviews about the toxicity of chemicals associated with produced water. Several studies looked at the toxicity of produced water to aquatic organisms and systems. Larval amphibian tissue contained concentrations of selenium, vanadium, and other ions associated with contamination from produced water in the wetlands of Montana and North Dakota (Hossack et al., 2018; Preston et al., 2019; Smalling et al., 2019). The study of a contamination event from a pipeline spill in North Dakota used a biological multi‐level investigation approach and found increased mortality in fish and elevated estrogenic activity initially (6 months post‐spill), which decreased over time (3 years post‐spill), while changes in microbial community structure persisted (Farag et al., 2022). Based on liver tissue samples from Brook trout in watersheds with active natural gas development in the Marcellus region in Pennsylvania, McLimans et al. reported potential disruptions to endocrine activity and indications of stress responses (McLimans et al., 2022). Other studies reported accumulation of produced‐water constituents in the shell and soft tissue of freshwater mussels (Geeza et al., 2018; McDevitt, Geeza, et al., 2021; Patnode et al., 2015; Piotrowski et al., 2020; Wang, Kunz, et al., 2019). Wang et al. reported decreased survival and growth rates in both fish and mussels and McDevitt et al. reported accumulation of produced‐water constituents in freshwater mussels even at low doses of exposure to produced water, which could make mussels a good indicator of produced‐water contamination (McDevitt, Geeza, et al., 2021; Wang, Kunz, et al., 2019). However, investigators generally suggested conducting larger studies with greater replicability to confirm these results (Geeza et al., 2018; Patnode et al., 2015; Piotrowski et al., 2020).

A suite of studies examined health outcomes from exposing adult and pregnant mice during preconception, prenatal, postnatal, and lactational phases to (a) a mixture of 23 commonly used hydraulic fracturing fluid additives, (b) samples of UOGD produced waters from Colorado and West Virginia, and (c) related samples of surface and ground waters impacted by produced water in those regions (Balise, Cornelius‐Green, Parmenter, et al., 2019, Balise, Cornelius‐Green, Parmenter, et al., 2019; Boulé et al., 2018; Kassotis et al., 2015, 2016, 2018b; Nagel et al., 2020; Orem et al., 2017). They reported effects on hormone concentrations, adipogenic (fat cell‐related) activity, energy expenditure, and risk‐taking behavior in mice. They also reported associated effects such as decreased sperm counts, increased body weight, altered uterine and ovary weights, increased heart weights and collagen deposition, disrupted folliculogenesis, increased mammary gland ductal density, and increased preneoplastic lesions, representing adverse developmental and reproductive health outcomes in wildlife when exposed to this chemical mixture. Exposing human tissue culture cells to the same chemical mixture, studies reported impacts on estrogen, androgen, glucocorticoid, progesterone, and thyroid receptor activity, representing potential endocrine disruptions (Kassotis et al., 2014, 2018a; Nagel et al., 2020). Organic chemicals and additives were specifically associated with higher acute toxicity (Aghababaei et al., 2021) and tumor promotion (Crosby et al., 2018). Yao et al. reported that exposure to flowback from the Marcellus Shale can induce malignant cell transformation in vitro, suggesting carcinogenic effects (Yao et al., 2015).

Compiling data and knowledge on produced‐water toxicity to different species under varying environmental conditions allows for more refined risk and hazard assessments that clarifies possible effects to human health if exposed (Danforth et al., 2019; Elliott et al., 2016; Rose et al., 2019). Some investigators have recommended more study to assess the combined toxicity of produced‐water constituents (Crosby et al., 2018) and chronic toxicity resulting from National Pollutant Discharge Elimination System (NPDES)‐permitted discharges and the extent of exposure downstream (McDevitt, Geeza, et al., 2021; McLaughlin, Blotevogel, et al., 2020, McLaughlin, Borch, et al., 2020).

3.2.1. Accidental or Fugitive Release of Produced Water to the Environment

Evaporation, spills, or leaks from well equipment such as onsite storage tanks, spills during produced water transportation, and accidental discharges to surface water represent unintentional or intentional produced‐water releases to the environment (Abraham et al., 2023; Akob et al., 2016; Bain et al., 2021; Bean et al., 2018; Boulé et al., 2018; Drollette et al., 2015, 2015, 2015; Kashani et al., 2024; Kassotis et al., 2016; Llewellyn et al., 2015; Maloney et al., 2017; McMahon et al., 2017; Orem et al., 2017; Reilly et al., 2015; Rish & Pfau, 2018; Wang, 2021; Warner, Christie, et al., 2013; Wright et al., 2019). These releases may impact ecosystem health and could also result in human exposures to produced‐water constituents in air, soil, sediment, surface water, and groundwater.

Emissions to Air . Emissions from open‐top produced‐water storage impoundments can contribute to particulate matter (PM), VOCs, and other constituents of concern in ambient air (Bean et al., 2018; Ma et al., 2022). Repeated recycling of produced water from hydraulic fracturing concentrates constituents like radium in the “impoundment sludge” that accumulates at the bottom of storage tanks, which then subsequently radioactively decays and can be released into the air as carcinogenic radon gas. Zhang et al. investigated the potential for radiation exposure from impoundment sludge in the Marcellus (Zhang et al., 2015). To avoid risks to humans from unintentional radiation exposure, the researchers proposed management options such as transporting impoundment sludge to low‐level radioactive waste landfills, hazardous waste landfills, or municipal and industrial solid waste landfills (Zhang et al., 2015). In their exposure assessment, the authors further reported low carcinogenic risks associated with inhaling radon emissions from this sludge for on‐site workers and the surrounding public (Zhang et al., 2015).

Studies have also reported on fugitive emissions from produced water at the wellhead (Allen et al., 2013; Johnson et al., 2022). These studies reported that more efficient emission‐control equipment, and elimination of previously uncontrolled releases from individual wells, helped to reduce produced water‐related air emissions.

Leaks or Spills into Groundwater . In many studies, investigators emphasized the importance of comprehensive and consistent monitoring of produced water and its disposal to protect groundwater reservoirs, particularly reservoirs that overlay oil and gas production areas, have natural hydraulic connections to formation water, or can be contaminated by produced‐water leaks from impoundments (Abraham et al., 2023; Drollette et al., 2015; Llewellyn et al., 2015; McMahon et al., 2017; Reilly et al., 2015; Rish & Pfau, 2018; Wang, 2021; Warner, Kresse, et al., 2013; Wright et al., 2019).

In the case of potable drinking water wells, several studies conducted in the Eagle Ford, Fayetteville, Haynesville, and Marcellus shale regions, as well as in California, did not find evidence of produced‐water contamination. Instead, those studies reported contamination from other sources such as animal waste, septic effluent, and road salt, or attributed rapid rates of flushing in the aquifer system as to why contamination may have been minimal (McMahon et al., 2017; Reilly et al., 2015; Warner, Kresse, et al., 2013; Wright et al., 2019). One study analyzed samples from several domestic groundwater wells in Pennsylvania due to reported contamination following the development of five UOGD wells 1–2 km away (Llewellyn et al., 2015). The investigators hypothesized that a leak reported from a produced‐water pit at the nearest UOGD well could have been the source as measurements were consistent with UOGD produced water in the Marcellus region. However, they were unable to fingerprint the contaminant source due to the lack of drilling, pit, or hydraulic fracturing fluid samples from nearby development. Additional data gaps around such investigations include a lack of an up‐to‐date publicly available database of polymer‐lined containment pits for produced water, and limited volume and composition data in the Pennsylvania Department of Environmental Protection's (PADEP) disclosed violation reports for spills (Drollette et al., 2015; Rish & Pfau, 2018).

An additional concern for potential groundwater contamination is produced‐water leakage from abandoned oil wells. Scanlon et al. highlighted several data gaps around confirming evidence for this exposure pathway, including a lack of produced‐water composition data and standardized monitoring around known locations of abandoned wells to assess potential leakages (Scanlon et al., 2021).

Unlined produced‐water disposal ponds designed to passively treat produced water can pose exposure risks by leaking into surrounding or underlying groundwater reservoirs (DiGiulio et al., 2021; DiGiulio & Jackson, 2016; Preston et al., 2019). In the San Joaquin Valley in California, DiGiulio et al. reported elevated levels of TDS and salinity in groundwater around these ponds (DiGiulio et al., 2021). DiGiulio et al. suggested treating produced waters discharged to wastewater disposal ponds to the extent that chemical concentrations are reduced to nonhazardous levels (DiGiulio et al., 2021).

Leaks or Spills into Surface Water . Produced‐water leaks and spills into surface water can have long lasting effects on water quality and ecological health both in individual water bodies and at the watershed level (Bonetti et al., 2021; Casey et al., 2022; Cozzarelli et al., 2017, 2021; Lauer et al., 2016; Michaels et al., 2022). Moreover, produced‐water spills contaminating surface water bodies used for drinking water, irrigation water, and recreational activities can potentially pose health risks to nearby communities (Abualfaraj et al., 2018; Torres et al., 2017, 2018).

Produced water from UIC disposal facilities can also leak into surrounding surface water and streambed sediments (Akob et al., 2016; Kassotis et al., 2016). Toxicological analysis at these sites can help ascertain the contamination levels of stream sediment and water, related human exposures, and where in the watershed contaminant levels might be elevated (Orem et al., 2017). Additionally, produced water generated during flowback operations can corrode pipeline infrastructure; these corrosion deposits accumulate in the produced water that is eventually disposed through Class II wells, potentially elevating the toxicity of any accidental environmental releases at UIC facilities (Chilkoor et al., 2018).

Cozzarelli et al. examined a spill event that occurred in January 2015 where 11.4 million liters (or about 3 million gallons) of produced water leaked from a pipeline into Blacktail Creek in the Williston Basin in North Dakota (Cozzarelli et al., 2017, 2021). Cozzarelli et al. reported barium, strontium, and radium activities in downstream sediment 6 months after the spill in concentrations 15 times the upstream concentrations (Cozzarelli et al., 2017). Sediment has the potential to be a long‐term reservoir for constituents even in the surrounding flood plain as the investigators observed persisting barium, strontium, radium, and ammonium levels in sediment 7.2 km downstream from the spill even after 2.5 years. Investigators concluded that the effect of produced‐water constituents on the environment is still poorly understood and highly variable depending on hydrological conditions (Cozzarelli et al., 2021). Biological studies conducted 6‐month to 3 years later revealed continued changes in microbial community structures and ammonium levels attributable to the spill (Farag et al., 2022).

Maloney et al. analyzed spill records from state databases in Colorado, New Mexico, North Dakota, and Pennsylvania; the states chosen for the study were based on data accessibility and the various shale basins that underly the regions (Maloney et al., 2017). From 2005 to 2014, spill rates increased for all states except for Pennsylvania, which began to decrease after an initial increase to plateau in 2009. In Rossi et al., the California Governor's Office of Emergency Services' (CalOES) HazMat database showed that the number of spill incidents and their frequency decreased between 2006 and 2020 and that 16% of the reported incidences of produced‐water spills had documented effects on waterways, with no known impacts to drinking water after CalOES started monitoring drinking water in 2016 (Rossi et al., 2022).

Multiple knowledge gaps affect tracking produced‐water spills, including lack of standardization and transparency in spill reporting and monitoring, exact geographical coordinates of reported spills, information on potential impacts to people and ecosystems, variation in each state's laws and reporting requirements, and unreported spills (Maloney et al., 2017; Rossi et al., 2022).

Attributing changes in surface water quality at the watershed level to environmental releases of produced water is challenging due to the variety of potential pollutant sources, such as natural brine seeps, oil and gas development‐associated wastes besides produced water, coal extraction wastes, mine drainage fluids, agriculture, urban development, road salt, deforestation, and other land use changes (Casey et al., 2022; Johnson et al., 2022; Kassotis et al., 2020; Michaels et al., 2022; Pelak & Sharma, 2014). Some studies in the Marcellus Shale region reported surface water samples with contamination from constituents mimicking produced‐water composition in the region (Harkness et al., 2017; Johnson et al., 2015) or temporal correlations between TDS levels in surface water and drilling phases of oil and gas development in the vicinity (Bonetti et al., 2021). In the same region, Mumford et al. reported that the intensity of oil and gas development in the watershed was independent of changes in stream‐water biology and geochemistry (Mumford et al., 2020).

3.2.2. Authorized Release of Produced Water to the Environment

Methods to treat produced water are often basin‐specific and are primarily aimed at TDS and radium reduction. Treatment of produced water can involve use of a combination of pressure, gravity, and heat to separate the oil, water, and gas phases (Conrad et al., 2020). At the most complex level, treatment of produced water may include advanced treatment methods such as desalination (Conrad et al., 2020). Advanced treatment methods are used to target hazardous compounds to reduce the potential for human exposure; however, treatment byproducts (e.g., radioactive sludge or reject brines from desalination processes) could pose their own threats to human and environmental health.

Discharges from Wastewater Treatment Facilities . As UOGD became more prevalent in the United States in the early 2000s, produced water was initially sent to municipal wastewater‐treatment facilities even though these facilities were not designed for this type of wastewater. Investigators reported high concentrations of TDS, salinity, bromide and other ions, which can lead to the formation of disinfection byproducts in the receiving waters for the treatment facility discharges (Burgos et al., 2017; Ferrar et al., 2013; Lauer et al., 2018; Skalak et al., 2013; Van Sice et al., 2018; Warner, Christie, et al., 2013). DBPs are relevant to human and ecological health due to their carcinogenicity and toxicity (Hladik et al., 2014; Huang et al., 2018; Landis et al., 2016; Parker et al., 2014; States et al., 2013; Weaver et al., 2015; Wilson & VanBriesen, 2012). In 2011, the PADEP issued a voluntary request to drilling companies to recycle produced water from UOGD instead of disposing it through municipal wastewater treatment plants. In 2016, EPA issued a mandatory ruling that made the voluntary request law (U.S. Environmental Protection Agency, 2020). While produced water from COGD contain many constituents of concern similar to UOGD (Tasker et al., 2020), it can still be treated by municipal wastewater treatment plants (U.S. Environmental Protection Agency, 2020).

Discharges to Surface Water and Applications to Land . Water‐stressed regions in the western United States are utilizing, and considering expansion of, produced water for agriculture and other uses to conserve potable water (Texas Produced Water Consortium, 2022). Furthermore, under the Clean Water Act, minimally treated produced water can be discharged to surface water west of the 98th meridian if it is deemed of “good enough quality” for beneficial use (e.g., crop irrigation, livestock watering, and wildlife propagation). “Good enough quality” can vary greatly by state with a NPDES specific conductance effluent limitation daily maximum of 7,500 µS/cm for produced water discharged to surface water in Wyoming versus 1,000 µS/cm in California for produced water reclaimed for irrigation water (Kondash et al., 2020; McLaughlin, Borch, et al., 2020).

Several investigators sampled and analyzed treated produced water and surface water at the point of discharge in Wyoming, upstream of a drinking water reservoir (McDevitt, McLaughlin, et al., 2020, McDevitt et al., 2024; McLaughlin, Borch, et al., 2020, 2021). Using a gas chromatography‐based nontargeted analysis, McLaughlin et al. detected more than 50 organic chemicals in surface water at the point of discharge (McLaughlin, Blotevogel, et al., 2020). They reported that the chemicals were not specified in effluent limit guidelines of the specific NPDES permit for this discharge, and that few had health thresholds for humans, livestock, and aquatic species. While the regulated concentrations mostly met NPDES limits at the point of discharge, inorganic chemicals exceeded limits downstream due to evaporation, suggesting the need for regulations in arid and semi‐arid regions to account for the effect of climate on the composition of the discharge (McDevitt, McLaughlin, et al., 2020). In a companion study that measured chronic toxicity of this produced‐water release (McLaughlin, Borch, et al., 2020), the research team reported higher mutation rates in yeast cells in the water, which correlated with the concentrations of benzene and radium (known carcinogens) in the discharge stream reported on previously (McLaughlin, Blotevogel, et al., 2020). Minimal acute toxicity and no developmental toxicity were observed. The investigators suggested that such toxicological investigations are necessary to determine potential health impacts to downstream users, especially since reporting requirements for NPDES permits apply to the discharge site only rather than downstream changes in concentration, which are more relevant to adequately protect human health.

McDevitt, McLaughlin, et al. (2021) and McLaughlin et al. (2021) evaluated the potential for wetlands to attenuate potentially harmful constituents in produced water discharged for beneficial use. Wetland systems, both natural and constructed, employ vegetation and soil to naturally treat water through processes like sorption and biodegradation. After analyzing surface water and sediment samples, they reported that, under oxygenated conditions, such systems may reduce dissolved radionuclide and organic chemical additive loads from produced water (McDevitt, McLaughlin, et al., 2021; McLaughlin et al., 2021).

Irrigation . Studies examined the effect of produced water treated to various extents used to irrigate crops (Miller et al., 2020; Redmon et al., 2021; Sappington & Rifai, 2018; Sedlacko et al., 2022). The high salinity typically associated with many produced waters can decrease the water infiltration rate of water into soil and impact crop production (Oetjen et al., 2018; U.S. Environmental Protection Agency, 2020). Miller et al. reported reduced soil health and significant changes in microbial community diversity and structure that may shape soil biochemical cycling and yield (Miller et al., 2020). Sedlacko et al. observed metabolite changes in produced water irrigated wheat that indicated plant stress independent of salinity (Sedlacko et al., 2022). Sappington and Rifai observed symptoms of salt stress but saw positive effects on crop germination in cowpeas when irrigated with produced water treated to lower TDS levels (Sappington & Rifai, 2018). Combining field monitoring data of irrigation water, soil, and edible crops with information from state water and soil survey databases, Redmon et al. quantified minimal health risk from the consumption of trace metals in crops irrigated with oil field produced water that is pretreated to reduce salinity (Redmon et al., 2021). Generally, investigators suggested further, large‐scale study to determine long‐term effects on soil, plants, and human health.

Treatment of Roads . Farnan et al. analyzed multiple types of road applicants for efficacy as dust suppressants, including ten produced‐water samples from Pennsylvania, as produced water has been used for road treatment in a number of states for the past 70 years (Farnan et al., 2023). They reported road runoff during a rain event as a potential source of radium entering the environment. Exposure to constituents of produced water used to de‐ice and suppress dust on roadways could also occur through deposition of alkaline earth metals such as radium and barium in soil and sediment along roads, and via evaporation in residential areas where people may spend more time on sidewalks and driveways (Bain et al., 2021; Skalak et al., 2013). Additionally, Stallworth et al. reported that produced water used as a dust suppressant was less effective than commercial road maintenance products due to the destabilization of road aggregate with high sodium brines, potentially exposing humans to increased inhalation risks (Stallworth et al., 2021). Researchers suggested further investigation to clarify potential exposures from road treatment, particularly to better understand the actual exposure times for people occupying spaces in proximity to roadways and to consider the composition of produced water spread on roads.

3.2.3. Potentially Exposed Human Populations

Human exposure to produced water can vary depending on the variability of produced‐water composition (Figure 4) and management practices (Figure 1). Populations can be exposed to constituents in produced water by inhaling compounds that evaporate from storage units like impoundments or that are applied as road treatment for dust suppression or de‐icing. Accidental leaks and intentional releases of produced water from wells, storage units, and disposal facilities can contaminate surface water and groundwater, and the construction and maintenance deficiencies such as corroded pipeline infrastructure or improperly sealed well casing can result in spills or leaks from well pad equipment or pipelines (Chilkoor et al., 2018; Li & Carlson, 2014; Scanlon et al., 2021; Ziemkiewicz et al., 2014). Discharges of produced water from treatment facilities to surface water can represent another potential pathway of exposure, especially if the receiving water body is used for recreation or as a source of drinking water or irrigation water. The type and magnitude of exposure would depend on the treatment methodology and efficiency.

Many studies summarized in this review reported evidence of contamination from produced water in various environmental media or the potential for human exposure. However, whether the exposures represent an important health risk is difficult to assess given the unknowns about produced‐water composition and how its constituents migrate, transform, and degrade following release to environments with variable meteorological and hydrological conditions. Of the 236 studies (Figure 2), only seven conducted assessments of human exposure to produced water (Bain et al., 2021; Ma et al., 2019; McLaughlin, Borch, et al., 2020; Soriano et al., 2022; Torres et al., 2017, 2018; Zhang et al., 2015), and five included information characterizing the communities who may be exposed to produced water (Bain et al., 2021; Johnston et al., 2016; Reilly et al., 2015; Silva et al., 2018; Soriano et al., 2022).

Of the studies that examined and distinguished among specific exposed subpopulations, investigators generally reported that populations living near infrastructure such as disposal wells, commuter routes using produced water for deicing, and private residential groundwater wells hypothesized to be contaminated by flowback, tended to have attributes including high poverty rates, low median household income, high proportions of people of color, or be rural (Bain et al., 2021; Johnston et al., 2016; Reilly et al., 2015; Silva et al., 2018).