Watershed-Scale Impacts from Surface Water ... - ACS Publications

Jul 12, 2017 - Watershed-Scale Impacts from Surface Water Disposal of Oil and Gas. Wastewater in Western Pennsylvania. William D. Burgos,*,†...
1 downloads 0 Views 3MB Size
Download PDF
Article pubs.acs.org/est

Watershed-Scale Impacts from Surface Water Disposal of Oil and Gas Wastewater in Western Pennsylvania William D. Burgos,*,† Luis Castillo-Meza,† Travis L. Tasker,† Thomas J. Geeza,† Patrick J. Drohan,‡ Xiaofeng Liu,† Joshua D. Landis,§ Jens Blotevogel,∥ Molly McLaughlin,∥ Thomas Borch,⊥ and Nathaniel R. Warner† †

Department of Civil and Environmental Engineering, The Pennsylvania State University, 212 Sackett Building, University Park, Pennsylvania 16802, United States ‡ Department of Ecosystem Science and Management, The Pennsylvania State University, 116 Agricultural Sciences and Industries Building, University Park, Pennsylvania 16802, United States § Department of Earth Sciences, Dartmouth College, 6105 Fairchild Hall, Hanover, New Hampshire 03755, United States ∥ Department of Civil and Environmental Engineering, Colorado State University, 1320 Campus Delivery, Fort Collins, Colorado 80523, United States ⊥ Department of Soil and Crop Sciences, Department of Soil and Crop Sciences, Colorado State University, 1170 Campus Delivery, Fort Collins, Colorado 80523, United States S Supporting Information *

ABSTRACT: Combining horizontal drilling with high volume hydraulic fracturing has increased extraction of hydrocarbons from low-permeability oil and gas (O&G) formations across the United States; accompanied by increased wastewater production. Surface water discharges of O&G wastewater by centralized waste treatment (CWT) plants pose risks to aquatic and human health. We evaluated the impact of surface water disposal of O&G wastewater from CWT plants upstream of the Conemaugh River Lake (dam controlled reservoir) in western Pennsylvania. Regulatory compliance data were collected to calculate annual contaminant loads (Ba, Cl, total dissolved solids (TDS)) to document historical industrial activity. In this study, two CWT plants 10 and 19 km upstream of a reservoir left geochemical signatures in sediments and porewaters corresponding to peak industrial activity that occurred 5 to 10 years earlier. Sediment cores were sectioned for the collection of paired samples of sediment and porewater, and analyzed for analytes to identify unconventional O&G wastewater disposal. Sediment layers corresponding to the years of maximum O&G wastewater disposal contained higher concentrations of salts, alkaline earth metals, and organic chemicals. Isotopic ratios of 226Ra/228Ra and 87Sr/86Sr identified that peak concentrations of Ra and Sr were likely sourced from wastewaters that originated from the Marcellus Shale formation.

■

INTRODUCTION

Much more water is required to drill and stimulate an unconventional O&G well via HVHF as compared to a conventional well.19 Larger volumes of water used for HVHF leads to larger volumes of wastewater returning to the surface (a.k.a. flowback and produced water). For example, in Pennsylvania in 2015, 10 000 unconventional Marcellus Shale gas wells produced 6.4 billion liters of wastewater while 100 000 conventional O&G wells produced 1.1 billion liters of wastewater.20 Wastewater produced from both conventional and unconventional O&G wells contain a variety of contaminants of concern−salts, metals, naturally occurring radioactive material (NORM), and both reservoir-derived and anthropogenic organic compounds.21−24 Waters produced from

Combining horizontal drilling with high-volume hydraulic fracturing (HVHF) has increased the extraction of hydrocarbons from low-permeability, unconventional oil and gas (O&G) formations across the United States.1 In 2015, 51% of oil and 67% of U.S. domestic gas were produced from tight formations where HVHF was used.2,3 The dramatic increase in gas production has shifted the major fuel source for electricity generation from coal to gas.4 Along with the benefits of increased domestic production of O&G, there are environmental concerns such as methane dissolution into groundwater,5,6 methane migration into buildings,7 methane emission to the atmosphere,8 contamination of shallow groundwater from surface spills,9−11 induced seismic activity from subsurface injection of O&G wastewater,12−14 and surface water or drinking water impairment from O&G wastewater disposal through centralized waste treatment (CWT) plants.15−18 © XXXX American Chemical Society

Received: April 3, 2017 Revised: June 15, 2017 Accepted: June 16, 2017

A

DOI: 10.1021/acs.est.7b01696 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Article

Environmental Science & Technology

Figure 1. Study area in the Allegheny River watershed in western Pennsylvania. (A) Volumes of oil and gas wastewater discharged from centralized waste treatment (CWT) plants in 2009. Circles represent CWT plants with NPDES permits, circle size represents annual volume. (B) Blacklick Creek watershed marked in red and Conemaugh River watershed marked in orange. Dashed rectangle shows location of Conemaugh River Lake (CRL). (C) Extent of CRL at a pool elevation of 906 feet m.s.l. denoted with red contour lines. Location of sediment cores shown as yellow circles. The median and mode for the CRL pool elevation from February 1, 1989 to May 21, 2015 (n = 9606) are 905 and 904 feet m.s.l., respectively. Figures generated using ArcGIS 10.4; HUC 10 watershed boundaries in (A) and (B), and aerial photograph in (C) from the Geospatial Data Gateway (https://datagateway.nrcs.usda.gov/); topographic contour lines in (C) from Pennsylvania Data Access (http://maps.psiee.psu.edu/ ImageryNavigator/).

carcinogens.31−33 Other observed adverse health effects caused by HVHF additives include mutagenicity, developmental toxicity, neurotoxicity, and endocrine disruption.32,34,35 After entering the stream, many of these contaminants will become particle-associated and be removed from the water column, but contaminant accumulation in stream sediments may continue to pose long-term risks. Lake sediments provide a suitable physical setting for reconstructing the pollution history and evaluating the corresponding long-term, watershed-scale impacts on sediments.36−39 The fate and transport of contaminants discharged into surface waters are affected by a variety of physicochemical processes and chemical characteristics of the contaminants. The forensic value of surface water grab samples is limited by the discrete representation of environmental conditions at one location, and one time. Aquatic sediments capture and accumulate contaminants that tend to become particleassociated (e.g., via sorption and ion exchange reactions) and, therefore, provide temporally composited information on watershed conditions. Layers of temporally coherent sediments deposited into a lake or reservoir can then provide a geochronological record of watershed conditions. The objectives of this study were to determine if chemical signatures in the sediment record can be used to quantify the

Marcellus Shale gas wells are exceptionally salty and radiogenic compared to shale plays in other parts of the U.S.25 Depending on the geographic location, O&G wastewaters are typically disposed of into underground injection control (UIC) wells, treated to some extent for in-field reuse, or sent to CWT plants for treatment and eventual discharge to surface water.18,21,26,27 Across the U.S., several states, including California, Michigan, Montana, Ohio, Oklahoma, Pennsylvania, Texas, West Virginia, and Wyoming, allow produced waters from O&G wells to be discharged to surface water.26,28 In Pennsylvania, the majority of CWT plants (10 of 11 plants operational in 2015; 19 of 38 total plants since 2004) are located in the Allegheny River watershed (Figure 1). While all of these plants have been issued National Pollution Discharge Elimination System (NPDES) permits, the chemical components and volumes of their effluents are not well characterized. Because CWT plants function as a regional collection point for O&G wastewater and often provide limited treatment,15,16 these facilities pose an elevated risk to surface water quality. If the receiving stream is used as a drinking source downstream, human health could be impacted via the consumption of water with elevated brominated disinfection byproducts.18,29,30 Furthermore, radium as well as some HVHF organic additive (breakdown) products such as formaldehyde are known B

DOI: 10.1021/acs.est.7b01696 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Article

Environmental Science & Technology

on May 20−21, 2015. Multiple cores were collected, each 1−2 m from the bank and ∼10-m between cores. The precise location of the three cores analyzed and details of the field sampling protocols for this study are provided in the SI. Table SI2 provides a summary of measurements made with the various cores. Porewater was extracted from the sediment, and pH and conductivity were measured inside an anaerobic chamber on unfiltered samples. Filtered porewater samples were acidified with HNO3 before analysis by inductively coupled plasma atomic emission spectrometry (ICP-AES) (Ba, Ca, Na, and Sr), ICP mass spectrometry (ICP-MS) (Pb)), or multicollector ICP-MS (MC-ICP-MS) (87Sr/86Sr). Filtered, unacidified porewater samples were analyzed for anions (Br, Cl) by ion chromatography (IC). Sediment samples were dried, digested in aqua-regia, and analyzed by ICP MS at the ALS Geochemistry Laboratory (Reno, NV). Total organic carbon and total organic nitrogen were measured by flash combustion. Samples were first pulverized and acidified to remove inorganic forms of carbon and nitrogen. Particle size distributions were determined by laser diffraction on sediment passed through a 0.84 mm sieve. Radioisotopes were measured by gamma spectroscopy using direct measurements at 186.1 keV (226Ra) that require spectrum deconvolution from the 185.74 keV emission from 235 42,43 228 U. Ra was measured indirectly at the same time, by daughter 228Ac using 911.16 keV.44,45 Activities of 210Pb, 7Be, and 137Cs were determined using 46.5 keV, 477.6 keV, 661.7 keV, respectively. All samples were measured in the same geometry as calibration standards. Applying both GC- and LC-coupled mass spectrometry methods, a nontargeted analysis was conducted to screen sediment samples for a broad range of organic contaminants commonly associated with O&G wastewater. Dried sediments were extracted with acetone/hexane solvent (1:1 v/v) on a vibration shaker and allowed to rest for 1 h afterward. Extracts were analyzed using a liquid chromatograph coupled with a time-of-flight mass spectrometer. Peaks were identified by accurate mass and potential chemical formulas, which were then verified with polyoxyethylene tridecyl ether and 4nonylphenol-polyethylene glycol standards (Sigma-Aldrich, Saint Louis, MO). Extracts were also analyzed for (semi-) volatile organics by a gas chromatograph equipped with a mass selective detector. Sediment Age Modeling. We combined 7Be and 210Pb using the linked radionuclide accumulation model (LRC)46 to provide explicit correction to conventional 210Pb age models (e.g., constant rate of supply (CRS)47) because of the “nonideal”48 depositional process during the sedimentation history of a river-lake (see SI for full details). With its short half-life of just 54 days, 7Be records deposition but not longer-term sedimentation. The LRC model uses 7Be to normalize the observed 210Pb inventory, and where the flux ratio of 7Be:210Pb can be constrained the ingrowth time of 210Pb in the soil or sediment can be interpreted. Storm-driven, turbidity-laden surface water samples were collected from Blacklick Creek and Two Lick Creek to directly measure the 7Be:210Pb ratio of “fresh” sediments. The LRC model does not require a steadystate assumption, but instead tracks the ingrowth of each of the 7 Be and 210Pb sediment inventories according to

historical impacts of O&G wastewater disposal. To accomplish these objectives we (1) identified a watershed with significant amounts of O&G wastewater disposal that included a reservoir with hydrodynamic conditions for sediment retention; (2) developed a sediment transport model to reconstruct the sediment stratigraphy in the reservoir to identify sampling locations most likely to retain a coherent temporal record; (3) developed an age model to date the sediment layers in cores collected from the reservoir; and (4) measured contaminant depth profiles in the sediment and porewater.

■

MATERIALS AND METHODS Site Selection. Oil and gas operators record wastewater volumes generated (every 6 or 12 months) from individual wells and how the wastewater is disposed (58 PA C.S. § 3222; 25 PA § 78.121) and report values to the Pennsylvania Department of Environmental Protection (PADEP) Office of Oil and Gas Management.20 We used these data to calculate the annual volumes of O&G wastewater sent to CWT plants. We focused on two facilities (identified by their NPDES permit numbers − CWT plant #1 = PA0095273 and CWT plant #2 = PA0206075) on Blacklick Creek upstream of the Conemaugh River Lake (Figure 1B). NPDES-permitted CWT plants are required to report the parameters specified in their permits to the PADEP in monthly discharge monitoring reports (DMRs). These DMRs can be submitted to the PADEP in hard copy or electronically through the eDMR online reporting system.40 For CWT plants #1 and #2, DMR data not available through the eDMR system were gathered during file reviews providing a complete data set from 2008 to 2015. Total yearly waste volumes of O&G wastewater discharged to sub-basins of the Allegheny River watershed were determined by summing contributions from all NPDES facilities within the watershed. Blacklick Creek was selected for study because it has historically received high volumes of O&G wastewater from two CWT plants, conveys a lower baseflow that provides less wastewater dilution compared to the main stem of the Allegheny River,21 and because it discharges into the Conemaugh River Lake, a dam-controlled reservoir. Sediment Transport Modeling. The Conemaugh River Lake is a reservoir formed by an U.S. Army Corps of Engineers (USACE) dam built for flood control. The reservoir receives water from two major tributaries, the Conemaugh River and Blacklick Creek. The One-Dimensional Sedimentation and River Hydraulics (SRH-1D) model developed by the US Bureau of Reclamation41 was used to simulate sediment deposition in the Conemaugh River Lake from 1982 to 2015. The data for model calibration and simulation included the hourly pool elevation, hourly discharge in Conemaugh River and Blacklick Creek, sediment grain size distributions at various locations, and river cross section surveys (both main channel and floodplain). The available data covered the period from the reservoir impoundment in 1952−2015 (sediment cores collected in May 2015). The model initial condition was setup by using the bathymetric survey in 1982 (a time when little sediment accumulated at the sampling locations) and the model was calibrated using the bathymetric survey in 2007 (significant sediment accumulation began after the minimum pool elevation was raised in 1989). A full description of the sediment transport model is included in the Supporting Information (SI). Sediment and Porewater Collection and Analyses. Sediment cores were collected from the Conemaugh River Lake

It = C

D (1 − e−λt ) λ

(1) DOI: 10.1021/acs.est.7b01696 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Article

Environmental Science & Technology where the total inventory I observed at any time t reflects depositional flux D and decay constant λ. To derive depth-dependent ages the accumulative 7Be:210Pb ratio of the sediment, RBe:Pb, is summed from the sediment− water interface over increasing sediment depths according to 0

R nBe:Pb = ,0

∑n IiBe 0 ∑n IiPb

≈

DBe λPb (1 − e−λBet ) DPb λBe (1 − e−λPbt )

(2)

for depositional flux D and decay constant λ of both 7Be and 210 Pb, with I as the activity inventory (Bq m−2) for each i depth interval, and n,0 denotes sediment layers summed from a depth n to the surface 0. The observed inventory for buried 210Pb was first corrected for 210Pb decay subsequent to its burial and decoupling from new 210Pb deposition for the accumulation time of upper layers, that is, to estimate the inventory at its time of burial. The inventory of any buried layer of 210Pb at the time of burial, IB, was estimated from classic decay law, according to IB = I /e−λt

(3)

where IB is the present-day measured inventory, and t is the age of the sediment layer. The accumulation age of the 210Pb peak was then solved as t from eq 1.

■

Figure 2. Historical oil and gas wastewater discharge to Blacklick Creek upstream of the Conemaugh River Lake (CRL). (A) Summed total volumes of conventional and unconventional O&G wastewater sent to CWT plants #1 and #2, compiled from ref 22 (B) Barium and TDS loads (kg/year) calculated from NPDES discharge monitoring reports (DMRs) for CWT plant #1 (furthest upstream). (C) Chloride and TDS loads (kg/year) calculated from NPDES DMRs for CWT plant #2 (closer to CRL). Error bars in (B) and (C) represent standard deviations associated with analytes reported in the DMRs (n = 4−12/year). Highlighted bands denote years of maximum discharge of unconventional O&G wastewater.

RESULTS AND DISCUSSION Wastewater Loading to Streams. Data from the Oil and Gas reporting site and DMR reports were used to calculate waste volumes discharged to the Allegheny River watershed from all NPDES-permitted facilities (Figure 1A) and from the two facilities located on Blacklick Creek upstream of the Conemaugh River Lake (Figures 2B,C). A review of monthly discharge monitoring reports (DMRs) for the CWT plants in the Allegheny River watershed found that some facilities reported TDS, Cl, Ba, oil, and grease, although no effluent limits were established for any of these analytes, and that facilities typically reported maximum daily flow rates as specified in their permits. No plants, however, were found to continuously record discharge flow rates and only a few reported 30 day-average flow rates. DMRs for both CWT plants were obtained through on-site file reviews at the Southwest Regional Office of the PADEP in Pittsburgh, PA. While the PADEP has initiated an electronic submission system for DMRs (eDMR) where these data are publically available online, the PADEP is required only to keep DMRs submitted in hardcopy for up to 10 years. Thus, our file reviews did not include data from before 2006. Data included in each facility’s DMRs were dependent on parameters included in their specific NPDES permit. Both CWT plants reported TDS concentrations, presumably representative of the whole 30-day reporting period, whereas only CWT plant #1 reported total barium concentrations. The barium load from CWT plant #1 peaked in 2009 and was coincident with the peak TDS load, both driven by the corresponding peak in O&G wastewater volume received. The TDS load from the CWT plant #2 also peaked in 2009. While effluent barium concentrations were not reported in the DMRs from this CWT plant, the barium load from this plant also likely peaked in 2009. Volumes of O&G wastewater sent to CWT plants were determined from waste reports submitted to the PADEP Office of Oil and Gas Management. O&G operators are required to submit waste reports (every 1, 6, or 12 months depending on

well type) that document the volume of wastewater generated from each well and where the wastewater is disposed. Total annual waste volumes sent to any NPDES-permitted facility were calculated using these data (Figure 2A). Contaminant concentrations (mass/volume) and the total volume of the effluent (volume/year) reported in DMRs were multiplied to calculate total contaminant loads (mass/year) entering the receiving stream (Figure 2B,C). Based on these data, the volume of unconventional O&G wastewater and associated contaminant loads (e.g., TDS, Cl, Ba) discharged to Pennsylvania surface waters peaked in 2009−2010. In August 2010, the Pennsylvania legislature forced new or expanding CWT plants to meet effluent water quality standards of 500 mg/L TDS, 250 mg/L Cl, 10 mg/L Ba, and 10 mg/L Sr. Up until this point, all but one CWT plant in Pennsylvania had only to monitor and report effluent TDS, chloride or osmotic pressure. Eight permitted facilities were listed as exempt from the new TDS standard.18 In April 2011, the PADEP requested that O&G operators no longer deliver wastewater from unconventional gas wells to CWT plants exempt from the new TDS effluent standard.18 The net effect of these policy changes dramatically reduced the volume of unconventional O&G wastewater sent to CWT plants and spurred the reuse of produced waters for continued hydraulic fracturing of new wells.27 Sediment Deposition. We collected sediment cores from Blacklick Creek within the pooled section of the reservoir at the D

DOI: 10.1021/acs.est.7b01696 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Article

Environmental Science & Technology upstream end of the Conemaugh River Lake (Figure 1C). Sediment transport simulations for the period between 1982 and 2015 show that our sampling location provided a coherent record for our analysis period. The simulated results indicate that, despite reservoir elevation fluctuations due to occasional high-flow floods, the sediment core locations show a persistent depositional pattern since the impoundment of the reservoir (more details in the SI). Starting in February 1989, the USACE raised the minimum pool elevation of the lake from 890 to 901−904 feet above mean sea level to accommodate a hydropower station near the dam. This operational change moved the upstream margin of the lake farther into the Blacklick Creek and created a second aggrading topset farther upstream. Consequently, the sediment at our sampling locations showed a recent layer of finer sediment on top of relatively coarser material (SI Figure SI3). In this case, coarse sediments below 120 cm were likely deposited before 1989. The deposition of >100 cm of fine sediments since 1989 is consistent with USACE sediment surveys that reported rapid deposition rates of up to 8 cm/yr.49 Similar trends were observed in the modeling results, that is, semiequilibrium prior to 1989 followed by rapid sedimentation after the pool elevation was increased. Sediment Age. While sediment depositional modeling can reconstruct stratigraphy, the LRC sediment age model was required to reconstruct the Conemaugh River Lake sediment history. The depth distribution of excess 210Pb (defined as 210Pb − 226Ra) in Core 4 appeared as a series of three subsurface peaks, each consistent in morphology with what we expect from nonideal deposition46,48 (Figure SI 6). Insofar as 210Pb chronology reflects sedimentary regime, the sedimentation history at this site appears to be episodic, with massive emplacement of sediment followed by quiescent periods as 7Be and 210Pb buildup by hyporheic exchange with the overlying water column. Measured 7Be activity in Core 4 exhibits a surface maximum and an exponential decline to maximum depth of ca. 30 cm. The total 7Be inventory observed in Core 4 is 6- to 10-times larger than might be expected for atmospheric equilibrium, which implies an unreasonable atmospheric flux for the region (16,000 ± 400 Bq m−2 y−1 versus ∼2000 that might be expected).43,50 This overly large 7Be inventory implies massive scavenging or focusing of atmospheric 7Be to the sediment bed. Thus, while the 210Pb inventory (8600 ± 200 Bq m−2 y−1) is comparable to or up to two-times higher than what might be expected for atmospheric equilibrium,43,50 7Be demonstrates that this 210Pb inventory has, in fact, accumulated over a much shorter period of time than 210Pb equilibrium would require. Using the LRC model, sediment age could be assigned to discrete depths (Figure 3A). The predicted ages are sensitive to the flux ratio of 7Be:210Pb of freshly suspended/deposited solids. This ratio was measured (7Be:210Pb = 4.4) for suspended solids collected during a storm event on May 25, 2017 but could vary over time. Therefore, a date range is provided to correspond to the maximum physically possible range of 7 Be:210Pb ratios of 2 to 7 calculated for this system. We found that 28.5 cm below the sediment−water interface corresponded to 2012 (range of 2010−2014). Similarly, we found that 66 cm below the sediment−water interface corresponded to 2006 (range of 1999−2010) and that 89 cm below the sediment− water interface corresponded to 2005 (range of 1997−2011). Additional details on model sensitivity are provided in the SI.

Figure 3. Depth profiles of sediment concentrations of alkaline earth metals. Zero on y-axis is sediment−water interface. Highlighted bands denote 2006−2011. (A) 226Ra/228Ra ratios greater than 1.0 are indicative of Marcellus Shale formation water.21,24 (B) Sediment ages of Core 3 were not modeled but synchronous peaks in concentrations of alkaline earth metals likely correspond to period of maximum surface water disposal of unconventional O&G wastewater. Cores collected May 21, 2015. Symbols represent single measurements.

Oil and Gas Record in Sediments. Alkaline earth metals such as strontium, barium, and radium occur naturally in Marcellus Shale and can be mobilized by high-salinity water associated with HVHF. Within each core, peak concentrations of strontium, barium, and radium in sediments all co-occurred at the same depths (Figure 3). This depth interval (highlighted from 45−80 cm below the sediment−water interface in Core 4; and 30−70 cm below the sediment−water interface in Core 3) corresponds to the years of maximum discharge of unconventional O&G wastewater disposal into Blacklick Creek (2006− 2011). Elevated concentrations of strontium, barium, and radium have all been detected in flowback and produced waters from unconventional Marcellus Shale gas wells,23,24 in CWT plant effluents, and in river sediments downstream of CWT plants.21 Elevated concentrations of these alkaline earth metals react with high-sulfate surface waters common in western Pennsylvania (>100 mg SO4/L)21 and likely precipitate or coprecipitate as Barite or strontium sulfate.51,52 Alkaline earth metals could also be removed from the water column and into the sediments via adsorption to clay minerals, iron oxides, or manganese oxides. The co-occurrence of peak concentrations of strontium, barium, and radium at the same depths was consistently observed in all sediment cores analyzed (Figure 3 and Figure SI5). In Core 3, this depth interval was 30−70 cm below the sediment−water interface. Because 7Be and 210Pb isotopes used to date the sediments were not measured at the fine spatial resolution as used in Core 4, the exact assignment of sediment E

DOI: 10.1021/acs.est.7b01696 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Article

Environmental Science & Technology age is less certain in Core 3. However, peak concentrations of strontium and barium were particularly sharp in this depth interval. Lead concentrations measured in sediment displayed no peak at depths during the years of maximum discharge of unconventional O&G wastewater disposal (Figure SI 5). The general absence of elevated lead concentrations at depth also suggests that the bottom-most sediments were deposited no earlier than ca. 1980, consistent with our age model and periodic sediment depth surveys measured by the USACE. The U.S. Environmental Protection Agency (USEPA) began to limit the lead content in gasoline in the mid-1970s (totally removal by 1996), and the associated drop in lead emissions is often recorded in undisturbed soils and sediments. The peak concentrations of barium in the sediment record (Figure 3) corresponded to the period of time when maximum barium and TDS loads were released by the upstream CWT plants (Figure 2B). CWT plant #1 is located 19 km upstream of the reservoir on Blacklick Creek and historically has accepted both conventional and unconventional O&G wastewater. CWT plant #2 is located 10 km upstream of the reservoir on Blacklick Creek and historically has accepted only conventional O&G wastewater. Based on the TDS load discharged into Blacklick Creek during 2008−2015, CWT plant #1 has typically contributed 25−100 times more pollution than CWT plant #2. The highest activity of radium (∼100 pCi/L) in the effluent from CWT plant #1 was reported in voluntary sampling results reported to the USEPA during this same period.53 Concentrations of major and trace elements in sediment porewaters also displayed strong depth-dependence (Figure 4). In general, concentrations of all major elements of interest (Cl, Br, Sr, Ba) were higher in porewaters than the overlying surface water (0 cm depth). The interval of elevated concentrations of strontium, barium, and chloride in porewaters corresponded with the highest strontium and barium concentrations in the sediments (Figure 4B). Both conductivity and bromide concentrations in the porewaters were highest at roughly 30 cm below the sediment−water interface in Core 3 (Figure 4A) and were offset from the peak concentrations of Sr, Ba, and Ra (Figures 3 and 4B). Compared to cationic alkaline earth metals, anionic halogens (Cl and Br) and monovalent cations (Na) are likely more mobile in the sediments. Cationic alkaline earth metals may be less mobile due to sorption and cation exchange reactions with sediment minerals. Concentration gradients between the porewater and overlying surface water could drive upward vertical transport. The offset between peak conductivity and peak strontium may be driven by differing extents of vertical transport that has occurred over 5−10 years since burial. Alternatively, high concentrations of Cl and Br may have been contributed from flue gas desulfurization facilities located upstream.54 Isotope Signatures of Marcellus Shale. The peak total Ra concentration (226Ra + 228Ra) in sediments corresponded to the peak 226Ra/228Ra ratio (Figure 3A). In highly saline fluids such as O&G wastewaters, Ra activities have been reported to range from 7000−21 000 pCi/L.24,55 Isotope ratios of 226 Ra/228Ra help identify the source of the radium in these sediments.21,24 Radium released into well stimulation fluids, conveyed to the surface with produced water, and ultimately discharged to surface water in the effluent from CWT plants retains a 226Ra/228Ra isotope ratio that reflects the host rock. The 226Ra/228Ra ratio for radium from Marcellus Shale reflects

Figure 4. Depth profiles of porewater concentrations in Core 3. Zero on y-axis is sediment−water interface, and white symbols on x-axis represent values from overlying reservoir. Highlighted bands denote 2006−2011. (A) Conductivity and salt ions. (B) Alkaline earth metals. 87 Sr/86Sr ratios of 0.710−0.711 and elevated Sr/Ca molar ratio are indicative of Marcellus Shale formation water.57 Symbols represent single measurements.

the higher U/Th ratio of the host rocks (i.e., shales) as compared to shallower horizons with lower U/Th ratios (e.g., coal beds and conventional O&G-bearing units in Upper Devonian-aged sandstones in western Pennsylvania) where the ratio typically does not exceed one. Importantly in this study, peak radium concentrations corresponded to higher 226 Ra/228Ra ratios (>1) and likely reflect Marcellus Shale formation water. Using XRF spectroscopy on a core logger system collecting data at every 5 mm, a similar peak in strontium concentrations was measured in Core 5 (Figure SI5). The maximum activities of total radium in all sediment cores did not exceed 5 pCi/gram, a threshold used by some states (e.g., Ohio) to differentiate solid wastes that can be disposed of in a landfill versus those that must be disposed of in radioactive waste facilities. To better resolve the source of the strontium in these porewaters, isotope ratios of 87Sr/86Sr were analyzed. Strontium 87 Sr/86Sr ratios do not fractionate as a result of biological or chemical reactions and are often employed in hydrologic studies.56 Within the Conemaugh River Lake watershed the total Sr concentration and the 87Sr/86Sr ratio in the background sediments and porewaters are determined primarily by weathering reactions of the parent minerals. In contrast, O&G wastewaters have been found to be highly enriched in strontium (>1 g/L) relative to background surface waters.21 The 87Sr/86Sr ratio in Marcellus Shale wastewater is distinct F

DOI: 10.1021/acs.est.7b01696 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Article

Environmental Science & Technology from conventional O&G wastewaters in western Pennsylvania as well as surface water and mine-impacted water.57,58 87Sr/86Sr ratios of 0.712−0.714 measured in the porewaters from the upper (0−30 cm below the sediment−water interface) and lowest (75−250 cm) portions of Core 3 (Figure 4B) are typical of western PA surface waters and coal mine drainage.21,57 However, the lowest 87Sr/86Sr ratios measured in the porewater (ca. 50 cm) corresponded with elevated strontium concentrations in both the porewater and sediments. Low 87Sr/86Sr ratios of 0.710−0.711 are indicative of Marcellus Shale formation water.57 Furthermore, elevated molar ratios of Sr/ Ca in this same interval are indicative of Marcellus Shale formation water57 indicating the preservation of a distinct geochemical signature from the time of peak unconventional O&G wastewater disposal. Mixing of the porewaters was evident based on plots of 1/Sr concentration versus 87Sr/86Sr ratios. A one-dimensional transport model was developed to examine the extent of possible dispersion of strontium within the sediment column (details in SI). Diffusive and advective transport were modeled in the sediment based on measured grain size analysis, sediment depth, trace metal concentrations, and reported values for the diffusive coefficient, porosity, and tortuosity of similar analytes and sediments. Results indicated that the transport distance of strontium in the sediment assuming a point-source at the depth of highest concentration was approximately 75 cm in the direction of groundwater flow, consistent with the observed dispersion of the strontium peak in Figure 4B. Organic Signatures. Organic compounds in O&G wastewater could also pose long-term environmental concerns. Because organic compounds can be transformed, or even completely mineralized over long time periods, their use for environmental forensic analysis is more limited as compared to inorganic contaminants. The selection of unique organic compounds to target for forensic analyses is further complicated because of limited information on organic chemicals used in hydraulic fracturing fluids,59 and less information on the transformation of these chemicals under high pressure and temperature.31,60 One study, targeting over 90% of disclosed additives to hydraulic fracturing fluids, detected petroleum distillates and nonylphenol ethoxylate surfactants in CWT plant effluents.16 The two major classes of organic compounds detected in the sediment extracts were nonylphenol ethoxylates (NPEs) and polycyclic aromatic hydrocarbons (PAHs) (Figure 5). These findings are in agreement with the almost exclusive detection of nonionic polyethoxylate surfactants and petroleum distillates in CWT plant effluents in Pennsylvania.16 Both NPEs and PAHs fall within the more hydrophobic and persistent fraction of these two broad chemical classes,61 likely explaining their accumulation in sediment. NPEs are widely used as corrosion inhibitors, solvents, surfactants, and nonemulsifiers in hydraulic fracturing fluids59 but also as well-maintenance chemicals in both conventional and unconventional wells. NPEs are known to form the more persistent endocrine disrupting compound (EDC) nonylphenol during biotransformation.37,62 In contrast, PAHs in O&G wastewater are likely sourced from naturally occurring, reservoir-derived organics. While both chemical classes can be sourced from more than just O&G wastewater, it is worth noting that maximum sediment concentrations of NPEs and PAHs coincided with the period of maximum surface water disposal of unconventional O&G wastewater (2006−

Figure 5. Depth profiles of sediment concentrations of organic contaminants in Core 3. Zero on y-axis is sediment−water interface. Highlighted bands in (A) and (B) denote 2006−2011. (A) Nonylphenol ethoxylate (NPE) surfactants measured by liquid chromatography/time-of-flight mass spectrometry. (B) Summation of 16 U.S. EPA priority polycyclic aromatic hydrocarbons (PAHs) measured by gas chromatography/mass spectrometry. (C) Extracted ion chromatogram of NPE species detected in sample from 49−54 cm below sediment−water interface. Symbols represent single measurements.

2011) and did not correlate with maxima in sediment organic carbon concentration (included in SI). Environmental Significance. Large quantities of O&G wastewater with high loads of chloride, barium, strontium, radium, and organic compounds have been discharged into the Conemaugh River watershed. Stream sediments in Blacklick Creek immediately downstream of CWT plant #1 were found to contain 226Ra+228Ra levels that were ∼200 times greater than activities measured in upstream and background sediments.21 Elevated concentrations of radium and other alkaline earth metals have now been detected in reservoir sediments ∼19 km farther downstream of this CWT plant. Despite several other sources of contaminants such as coal bed methane, coal mine drainage, and flue gas desulfurization releases that can impact surface water quality,54 we document multiple lines of evidence (Br, Cl, Ba, Sr concentrations, radium activities, isotopic ratios of 226Ra/228Ra and 87Sr/86Sr, and organic chemicals) that indicate the legacy of unconventional O&G wastewater disposal has impacted stream sediments and porewater on a watershedscale, likely >20 km. The unconventional O&G wastewater signal was likely derived from a relatively small volume of O&G water relative to the volume of the stream, but has nonetheless had a measurable impact. Risks posed by the pollutants buried in the sediment and porewater of the Conemaugh River Lake are difficult to assess. USEPA Region III uses a soil screening level (SSL) to evaluate if contaminated sediments may threaten groundwater quality.63 Using a risk-based model, the SSL for Sr is 422 mg/kg and the SSL for Ba is 1600 mg/kg. Using a maximum contaminant level-based model, the SSL for Ba is 82 mg/kg. In this study, the maximum Sr sediment concentration was 117 mg/kg, whereas the maximum Ba sediment concentration was 531 mg/ kg. These comparisons suggest that Ba in the sediment could threaten the quality of neighboring groundwater. For some G

DOI: 10.1021/acs.est.7b01696 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Article

Environmental Science & Technology states like Ohio, solids containing total 226Ra+228Ra concentrations greater than 5 pCi/gram must be disposed of in radioactive waste facilities (O.R.C. § 3734.02 (P)(2))). In this study, the maximum 226Ra+228Ra concentration was 4.3 pCi/ gram. Drinking water standards are not directly applicable because the Conemaugh River is not used as a public water source. Effective management and disposal practices for wastewater from unconventional O&G wells are critical for development to occur with limited environmental impact. Subsurface disposal into UIC wells appears to raise concerns regarding increased seismic activity12 and thus could be limited in the future. Infield wastewater reuse for additional well completions will (eventually) become less available as drilling activity in any one region slows down. Therefore, the volume of O&G wastewater sent to CWT plants for treatment and disposal to surface water is likely to increase in the future. As highlighted by this study, large volumes of effluent from CWT plants can lead to sediment contamination far downstream that persists for long periods of time. We note that these discharges are permissible with respect to permits that stipulate no contaminant concentration limits. The overall risk associated with these sediment contaminants has not yet been determined, but to minimize potential impacts, regulatory agencies should develop and apply more restrictive effluent discharge limits for CWT plants that protect human health and the environment.

■

(Final Report); U.S. Environmental Protection Agency: Washington DC, 2016. (2) EIA. Hydraulic Fracturing Accounts for about Half of Current U.S. Crude Oil Production; U.S. Energy Information Administration: 2016. (3) EIA. Hydraulically Fractured Wells Provide Two-Thirds of U.S. Natural Gas Production; U.S. Energy Information Administration: 2016. (4) EIA. Natural gas expected to surpass coal in mix of fuel used for U.S. power generation in 2016; US Energy Information Administration: 2016. (5) Osborn, S. G.; Vengosh, A.; Warner, N. R.; Jackson, R. B. Methane contamination of drinking water accompanying gas-well drilling and hydraulic fracturing. Proc. Natl. Acad. Sci. U. S. A. 2011, 108, 8172−8176. (6) Sherwood, O. A.; Rogers, J. D.; Lackey, G.; Burke, T. L.; Osborn, S. G.; Ryan, J. N. Proc. Natl. Acad. Sci. U. S. A. 2016, 113 (30), 8391− 8396. (7) Christopher, N.; Osher; Finley, B. Fatal home explosion reignites debate over drilling setbacks, but with a twist. Denver Post April 30, 2017 2017. (8) Caulton, D. R.; Shepson, P. B.; Santoro, R. L.; Sparks, J. P.; Howarth, R. W.; Ingraffea, A. R.; Cambaliza, M. O. L.; Sweeney, C.; Karion, A.; Davis, K. J.; Stirm, B. H.; Montzka, S. A.; Miller, B. R. Toward a better understanding and quantification of methane emissions from shale gas development. Proc. Natl. Acad. Sci. U. S. A. 2014, 111 (17), 6237−6242. (9) McLaughlin, M. C.; Borch, T.; Blotevogel, J. Spills of Hydraulic Fracturing Chemicals on Agricultural Topsoil: Biodegradation, Sorption, and Co-contaminant Interactions. Environ. Sci. Technol. 2016, 50 (11), 6071−6078. (10) Maloney, K. O.; Baruch-Mordo, S.; Patterson, L. A.; Nicot, J.-P.; Entrekin, S. A.; Fargione, J. E.; Kiesecker, J. M.; Konschnik, K. E.; Ryan, J. N.; Trainor, A. M.; Saiers, J. E.; Wiseman, H. J., Unconventional oil and gas spills: Materials, volumes, and risks to surface waters in four states of the U.S. Sci. Total Environ..201758158236910.1016/j.scitotenv.2016.12.142 (11) Cozzarelli, I. M.; Skalak, K. J.; Kent, D. B.; Engle, M. A.; Benthem, A.; Mumford, A. C.; Haase, K.; Farag, A.; Harper, D.; Nagel, S. C.; Iwanowicz, L. R.; Orem, W. H.; Akob, D. M.; Jaeschke, J. B.; Galloway, J.; Kohler, M.; Stoliker, D. L.; Jolly, G. D. Environmental signatures and effects of an oil and gas wastewater spill in the Williston Basin, North Dakota. Sci. Total Environ. 2017, 579, 1781−1793. (12) Ellsworth, W. L. Injection-Induced Earthquakes. Science 2013, 341 (6142), 142. (13) Keranen, K. M.; Weingarten, M.; Abers, G. A.; Bekins, B. A.; Ge, S. Sharp increase in central Oklahoma seismicity since 2008 induced by massive wastewater injection. Science 2014, 345 (6195), 448−451. (14) van der Elst, N. J.; Savage, H. M.; Keranen, K. M.; Abers, G. A. Enhanced Remote Earthquake Triggering at Fluid-Injection Sites in the Midwestern United States. Science 2013, 341 (6142), 164−167. (15) Ferrar, K. J.; Michanowicz, D. R.; Christen, C. L.; Mulcahy, N.; Malone, S. L.; Sharma, R. K. Assessment of effluent contaminants from three wastewater treatment plants discharging Marcellus Shale wastewater to surface waters in Pennsylvania. Environ. Sci. Technol. 2013, 47 (7), 3472−3481. (16) Getzinger, G. J.; O’Connor, M. P.; Hoelzer, K.; Drollette, B. D.; Karatum, O.; Deshusses, M. A.; Ferguson, P. L.; Elsner, M.; Plata, D. L. Natural Gas Residual Fluids: Sources, Endpoints, and Organic Chemical Composition after Centralized Waste Treatment in Pennsylvania. Environ. Sci. Technol. 2015, 49 (14), 8347−8355. (17) Hladik, M. L.; Focazio, M. J.; Engle, M. Discharges of produced waters from oil and gas extraction via wastewater treatment plants are sources of disinfection by-products to receiving streams. Sci. Total Environ. 2014, 466, 1085−1093. (18) Wilson, J. M.; VanBriesen, J. M. Oil and Gas Produced Water Management and Surface Drinking Water Sources in Pennsylvania. Environ. Pract. 2012, 14 (04), 288−300.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.est.7b01696. Additional information as noted in the text (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*Phone 814-863-0578; e-mail: [email protected]. ORCID

William D. Burgos: 0000-0003-3269-2921 Thomas Borch: 0000-0002-4251-1613 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This research was partially supported by the United State Geological Survey 104B Grant No. G11AP20102 to W.B. and P.D.; the Penn State Institutes of Energy and the Environment to N.W., P.D., and W.B.; and by Colorado State’s Water Center and School of Global Environmental Sustainability to J.B. and T.B. L.C. was supported by the Fulbright Commission of Colombia and Universidad Pontificia Bolivariana seccional Bucaramanga. We thank Rose Reilly, Carl Nim, and Mark Keppler of the U.S. Army Corps of Engineers, Pittsburgh District for logistical and scientific support during site selection and sampling, Katherine Van Sice for XRD interpretation and analyses, and Scott Hynek and the Penn State Laboratory for Isotopes and Metals in the Environment (LIME) for support during isotopic analysis.

■

REFERENCES

(1) Hydraulic Fracturing for Oil and Gas: Impacts from the Hydraulic Fracturing Water Cycle on Drinking Water Resources in the United States H

DOI: 10.1021/acs.est.7b01696 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Article

Environmental Science & Technology (19) Kondash, A.; Vengosh, A., Water Footprint of Hydraulic Fracturing. Environ. Sci. Technol. Lett. 2015.227610.1021/acs.estlett.5b00211 (20) PADEP PA DEP Oil and Gas Reporting Website: Statewide Data Downloads by Reporting Period. . https://www. paoilandgasreporting.state .pa.us/publicreports/Modules/ DataExports/ (January 2016),. (21) Warner, N. R.; Christie, C. A.; Jackson, R. B.; Vengosh, A. Impacts of Shale Gas Wastewater Disposal on Water Quality in Western Pennsylvania. Environ. Sci. Technol. 2013, 47 (20), 11849− 11857. (22) Warner, N. R.; Jackson, R. B.; Darrah, T. H.; Osborn, S. G.; Down, A.; Zhao, K. G.; White, A.; Vengosh, A. Geochemical evidence for possible natural migration of Marcellus Formation brine to shallow aquifers in Pennsylvania. Proc. Natl. Acad. Sci. U. S. A. 2012, 109 (30), 11961−11966. (23) Haluszczak, L. O.; Rose, A. W.; Kump, L. R. Geochemical evaluation of flowback brine from Marcellus gas wells in Pennsylvania, USA. Appl. Geochem. 2013, 28, 55−61. (24) Rowan, E.; Engle, M.; Kirby, C.; Kraemer, T. Radium Content of Oil- And Gas-Field Produced Waters in the Northern Appalachian Basin (USA)Summary and Discussion of Data: U.S. Geological Survey Scientific Investigations Report 2011−5135; U.S. Geological Survey 2011. (25) Shih, J.-S.; Saiers, J. E.; Anisfeld, S. C.; Chu, Z.; Muehlenbachs, L. A.; Olmstead, S. M. Characterization and Analysis of Liquid Waste from Marcellus Shale Gas Development. Environ. Sci. Technol. 2015, 49 (16), 9557−9565. (26) Vengosh, A.; Jackson, R. B.; Warner, N.; Darrah, T. H.; Kondash, A. A Critical Review of the Risks to Water Resources from Unconventional Shale Gas Development and Hydraulic Fracturing in the United States. Environ. Sci. Technol. 2014, 48 (15), 8334−8348. (27) Maloney, K. O.; Yoxtheimer, D. A. Production and disposal of waste materials from gas and oil extraction from the Marcellus Shale play in Pennsylvania. Environ. Pract. 2012, 14 (04), 278−287. (28) Ray, J.; Engelhardt, F. Produced Water − Technological/ Environmental Issues And Solutions; Plenum Press: New York & London, 1992. (29) Wilson, J.; Wang, Y.; VanBriesen, J., Sources of High Total Dissolved Solids to Drinking Water Supply in Southwestern Pennsylvania. J. Environ. Eng. 2013, 140, (5).B401400310.1061/ (ASCE)EE.1943-7870.0000733 (30) Parker, K. M.; Zeng, T.; Harkness, J.; Vengosh, A.; Mitch, W. A. Enhanced Formation of Disinfection Byproducts in Shale Gas Wastewater-Impacted Drinking Water Supplies. Environ. Sci. Technol. 2014, 48 (19), 11161−11169. (31) Kahrilas, G. A.; Blotevogel, J.; Corrin, E. R.; Borch, T. Downhole Transformation of the Hydraulic Fracturing Fluid Biocide Glutaraldehyde: Implications for Flowback and Produced Water Quality. Environ. Sci. Technol. 2016, 50 (20), 11414−11423. (32) Kahrilas, G. A.; Blotevogel, J.; Stewart, P. S.; Borch, T. Biocides in Hydraulic Fracturing Fluids: A Critical Review of Their Usage, Mobility, Degradation, and Toxicity. Environ. Sci. Technol. 2015, 49 (1), 16−32. (33) Swenberg, J. A.; Kerns, W. D.; Mitchell, R. I.; Gralla, E. J.; Pavkov, K. L. Induction of Squamous Cell Carcinomas of the Rat Nasal Cavity by Inhalation Exposure to Formaldehyde Vapor. Cancer Res. 1980, 40 (9), 3398. (34) Kassotis, C. D.; Tillitt, D. E.; Davis, J. W.; Hormann, A. M.; Nagel, S. C., Estrogen and Androgen Receptor Activities of Hydraulic Fracturing Chemicals and Surface and Ground Water in a DrillingDense Region. Endocrinology 2013.15589710.1210/en.2013-1697 (35) Kassotis, C. D.; Iwanowicz, L. R.; Akob, D. M.; Cozzarelli, I. M.; Mumford, A. C.; Orem, W. H.; Nagel, S. C. Endocrine disrupting activities of surface water associated with a West Virginia oil and gas industry wastewater disposal site. Sci. Total Environ. 2016, 557−558, 901−910.

(36) von Gunten, H. R.; Sturm, M.; Moser, R. N. 200-Year Record of Metals in Lake Sediments and Natural Background Concentrations. Environ. Sci. Technol. 1997, 31 (8), 2193−2197. (37) Ferguson, P. L.; Bopp, R. F.; Chillrud, S. N.; Aller, R. C.; Brownawell, B. J. Biogeochemistry of Nonylphenol Ethoxylates in Urban Estuarine Sediments. Environ. Sci. Technol. 2003, 37 (16), 3499−3506. (38) Iozza, S.; Müller, C. E.; Schmid, P.; Bogdal, C.; Oehme, M. Historical Profiles of Chlorinated Paraffins and Polychlorinated Biphenyls in a Dated Sediment Core from Lake Thun (Switzerland). Environ. Sci. Technol. 2008, 42 (4), 1045−1050. (39) Bonvin, F.; Rutler, R.; Chèvre, N.; Halder, J.; Kohn, T. Spatial and Temporal Presence of a Wastewater-Derived Micropollutant Plume in Lake Geneva. Environ. Sci. Technol. 2011, 45 (11), 4702− 4709. (40) PADEP, Oil and Gas Reporting − Electronic (OGRE) Public Reporting Data. In 2015. (41) Huang, J. V.; Greimann, B. SRH-1D User’s Manual (Sedimentation and River Hydraulics − One Dimension), version 4.0; U.S. Bureau of Reclamation: Denver, CO, USA, 2017. (42) Köhler, M.; Preuße, W.; Gleisberg, B.; Schäfer, I.; Heinrich, T.; Knobus, B. Comparison of methods for the analysis of 226Ra in water samples. Appl. Radiat. Isot. 2002, 56 (1−2), 387−392. (43) Landis, J. D.; Renshaw, C. E.; Kaste, J. M. Measurement of 7Be in soils and sediments by gamma spectroscopy. Chem. Geol. 2012, 291, 175−185. (44) Parsa, B.; Obed, R. N.; Nemeth, W. K.; Suozzo, G. Health Phys. 2004, 86 (2), 145−149. (45) Greeman, D. J.; Rose, A. W. Factors controlling the emanation of radon and thoron in soils of the eastern U.S.A. Chem. Geol. 1996, 129 (1−2), 1−14. (46) Landis, J. D.; Renshaw, C. E.; Kaste, J. M. Beryllium-7 and lead210 chronometry of modern soil processes: The Linked Radionuclide aCcumulation model, LRC. Geochim. Cosmochim. Acta 2016, 180, 109−125. (47) Appleby, P. G.; Oldfield, F. The calculation of lead-210 dates assuming a constant rate of supply of unsupported 210Pb to the sediment. Catena 1978, 5 (1), 1−8. (48) Abril, J.-M.; Gharbi, F. Radiometric dating of recent sediments: beyond the boundary conditions. Journal of Paleolimnology 2012, 48 (2), 449−460. (49) Wade, R.; Freeman, G. E.; Teeter, A. M.; Thomas, W. A. Conemaugh River Lake Sediment Removal Study. ; 1994. (50) Landis, J. D.; Renshaw, C. E.; Kaste, J. M. Quantitative Retention of Atmospherically Deposited Elements by Native Vegetation Is Traced by the Fallout Radionuclides 7Be and 210Pb. Environ. Sci. Technol. 2014, 48 (20), 12022−12030. (51) Zhang, T.; Gregory, K.; Hammack, R. W.; Vidic, R. D. Coprecipitation of Radium with Barium and Strontium Sulfate and Its Impact on the Fate of Radium during Treatment of Produced Water from Unconventional Gas Extraction. Environ. Sci. Technol. 2014, 48 (8), 4596−4603. (52) Kondash, A. J.; Warner, N. R.; Lahav, O.; Vengosh, A. Radium and Barium Removal through Blending Hydraulic Fracturing Fluids with Acid Mine Drainage. Environ. Sci. Technol. 2013, 48 (2), 1334− 1342. (53) USEPA Voluntary sampling Data from Oil and Gas Wastewater Treatment Facilities, available at http://www.epa.gov/region3/ marcellus_shale/#epawpadep. (54) Sources Contributing Inorganic Species to Drinking Water Intakes During Low Flow Conditions on the Allegheny River in Western Pennsylvania; USEPA: Washington DC, 2015. (55) Pennsylvania Department of Environmental Protection Technologically Enhanced Naturally Occuring Radioactive Materials (TENORM) Study Report 2015. (56) Hunt, R. J.; Bullen, T. D.; Krabbenhoft, D. P.; Kendall, C. Using Stable Isotopes of Water and Strontium to Investigate the Hydrology of a Natural and a Constructed Wetland. Groundwater 1998, 36 (3), 434−443. I

DOI: 10.1021/acs.est.7b01696 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Article

Environmental Science & Technology (57) Chapman, E. C.; Capo, R. C.; Stewart, B. W.; Kirby, C. S.; Hammack, R. W.; Schroeder, K. T.; Edenborn, H. M. Geochemical and Strontium Isotope Characterization of Produced Waters from Marcellus Shale Natural Gas Extraction. Environ. Sci. Technol. 2012, 46 (6), 3545−3553. (58) Stewart, B. W.; Chapman, E. C.; Capo, R. C.; Johnson, J. D.; Graney, J. R.; Kirby, C. S.; Schroeder, K. T. Origin of brines, salts and carbonate from shales of the Marcellus Formation: Evidence from geochemical and Sr isotope study of sequentially extracted fluids. Appl. Geochem. 2015, 60, 78−88. (59) Elsner, M.; Hoelzer, K. Quantitative Survey and Structural Classification of Hydraulic Fracturing Chemicals Reported in Unconventional Gas Production. Environ. Sci. Technol. 2016, 50 (7), 3290−3314. (60) Tasker, T. L.; Piotrowski, P. K.; Dorman, F. L.; Burgos, W. D. Metal Associations in Marcellus Shale and Fate of Synthetic Hydraulic Fracturing Fluids Reacted at High Pressure and Temperature. Environ. Eng. Sci. 2016, 33 (10), 753−765. (61) John, D. M.; House, W. A.; White, G. F. Environmental fate of nonylphenol ethoxylates: Differential adsorption of homologs to components of river sediment. Environ. Toxicol. Chem. 2000, 19 (2), 293−300. (62) Ejlertsson, J.; Nilsson, M.-L.; Kylin, H.; Bergman, Å.; Karlson, L.; Ö quist, M.; Svensson, B. H. Anaerobic Degradation of Nonylphenol Mono- and Diethoxylates in Digestor Sludge, Landfilled Municipal Solid Waste, and Landfilled Sludge. Environ. Sci. Technol. 1999, 33 (2), 301−306. (63) USEPA, Regional Screening Levels (RSLs) - Generic Tables (May 2016). https://www.epa.gov/risk/regional-screening-levels-rslsgeneric-tables-may-2016 (June 14),.

J

DOI: 10.1021/acs.est.7b01696 Environ. Sci. Technol. XXXX, XXX, XXX−XXX