Search

The production of fossil fuels, including oil, gas, and coal, retains a dominant share in US energy production and serves as a major anthropogenic source of methane, a greenhouse gas with a high warming potential. In addition to directly emitting methane into the air, fossil fuel production can release methane into groundwater, and that methane may eventually reach the atmosphere. In this study, we collected 311 water samples from an unconventional oil and gas (UOG) production region in Pennsylvania and an oil and gas (O&G) and coal production region across Ohio and West Virginia. Methane concentration was negatively correlated to distance to the nearest O&G well in the second region, but such a correlation was shown to be driven by topography as a confounding variable. Furthermore, sulfate concentration was negatively correlated with methane concentration and with distance to coal mining in the second region, and these correlations were robust even when considering topography. We hypothesized that coal mining enriched sulfate in groundwater, which in turn inhibited methanogenesis and enhanced microbial methane oxidation. Thus, this study highlights the complex interplay of multiple factors in shaping groundwater methane concentrations, including biogeochemical conversion, topography, and conventional fossil extraction.

Per- and polyfluoroalkyl substances (PFASs) are a class of toxic organic compounds that have been widely used in consumer applications and industrial activities, including oil and gas production. We measured PFAS concentrations in 45 private wells and 8 surface water sources in the oil and gas-producing Doddridge, Marshall, Ritchie, Tyler, and Wetzel Counties of northern West Virginia and investigated relationships between potential PFAS sources and drinking water receptors. All surface water samples and 60% of the water wells sampled contained quantifiable levels of at least one targeted PFAS compound, and four wells (8%) had concentrations above the proposed maximum contaminant level (MCL) for perfluorooctanoic acid (PFOA). Individual concentrations of PFOA and perfluorobutanesulfonic acid exceeded those measured in finished public water supplies. Total targeted PFAS concentrations ranged from nondetect to 36.8 ng/L, with surface water concentrations averaging 4-fold greater than groundwater. Semiquantitative, nontargeted analysis showed concentrations of emergent PFAS that were potentially higher than targeted PFAS. Results from a multivariate latent variable hierarchical Bayesian model were combined with insights from analyses of groundwater chemistry, topographic characteristics, and proximity to potential PFAS point sources to elucidate predictors of PFAS concentrations in private wells. Model results reveal (i) an increased vulnerability to contamination in upland recharge zones, (ii) geochemical controls on PFAS transport likely driven by adsorption, and (iii) possible influence from nearby point sources.

Concerns over unconventional oil and gas (UOG) development persist, especially in rural communities that rely on shallow groundwater for drinking and other domestic purposes. Given the continued expansion of the industry, regional (vs local scale) models are needed to characterize groundwater contamination risks faced by the increasing proportion of the population residing in areas that accommodate UOG extraction. In this paper, we evaluate groundwater vulnerability to contamination from surface spills and shallow subsurface leakage of UOG wells within a 104,000 km2region in the Appalachian Basin, northeastern USA. We test a computationally efficient ensemble approach for simulating groundwater flow and contaminant transport processes to quantify vulnerability with high resolution. We also examine metamodels, or machine learning models trained to emulate physically based models, and investigate their spatial transferability. We identify predictors describing proximity to UOG, hydrology, and topography that are important for metamodels to make accurate vulnerability predictions outside their training regions. Using our approach, we estimate that 21,000–30,000 individuals in our study area are dependent on domestic water wells that are vulnerable to contamination from UOG activities. Our novel modeling framework could be used to guide groundwater monitoring, provide information for public health studies, and assess environmental justice issues.

Health studies report associations between metrics of residential proximity to unconventional oil and gas (UOG) development and adverse health endpoints. We investigated whether exposure through household groundwater is captured by existing metrics and a newly developed metric incorporating groundwater flow paths. We compared metrics with detection frequencies/concentrations of 64 organic and inorganic UOG-related chemicals/groups in residential groundwater from 255 homes (Pennsylvania n= 94 and Ohio n= 161). Twenty-seven chemicals were detected in ≥20% of water samples at concentrations generally below U.S. Environmental Protection Agency standards. In Pennsylvania, two organic chemicals/groups had reduced odds of detection with increasing distance to the nearest well: 1,2-dichloroethene and benzene (Odds Ratio [OR]: 0.46, 95% confidence interval [CI]: 0.23–0.93) and m-and p-xylene (OR: 0.28, 95% CI: 0.10–0.80); results were consistent across metrics. In Ohio, the odds of detecting toluene increased with increasing distance to the nearest well (OR: 1.48, 95% CI: 1.12–1.95), also consistent across metrics. Correlations between inorganic chemicals and metrics were limited (all |ρ| ≤ 0.28). Limited associations between metrics and chemicals may indicate that UOG-related water contamination occurs rarely/episodically, more complex metrics may be needed to capture drinking water exposure, and/or spatial metrics in health studies may better reflect exposure to other stressors.

Conflicting evidence exists as to whether or not unconventional oil and gas (UOG) development has enhanced methane transport into groundwater aquifers over the past 15 years. In this study, recent groundwater samples were collected from 90 domestic wells and 4 springs in Northeastern Pennsylvania located above the Marcellus Shale after more than a decade of UOG development. No statistically significant correlations were observed between the groundwater methane level and various UOG geospatial metrics, including proximity to UOG wells and well violations, as well as the number of UOG wells and violations within particular radii. The δ13C and methane-to-higher chain hydrocarbon signatures suggested that the elevated methane levels were not attributable to UOG development nor could they be explained by using simple biogenic–thermogenic end-member mixing models. Instead, groundwater methane levels were significantly correlated with geochemical water type and topographical location. Comparing a subset of contemporary methane measurements to their co-located pre-drilling records (n= 64 at 49 distinct locations) did not indicate systematic increases in methane concentration but did reveal several cases of elevated concentration (n= 12) across a spectrum of topographies. Multiple lines of evidence suggested that the high-concentration groundwater methane could have originated from shallow thermogenic methane that migrated upward into groundwater aquifers with Appalachian Basin brine.

We characterize the appropriation of surface water for the extraction of natural gas from Pennsylvania's Marcellus Shale, and we examine the influences of these diversions on stream flows at 300 sites. Our analysis reveals that permitted withdrawals range from 50 m3/d to more than 18,000 m3/d and that water is taken from streams of all sizes, from headwater streams to eighth‐order rivers. Flow alteration varies inversely with watershed area and, for larger streams, is compounded by upstream withdrawals. The ratio of daily permitted withdrawal to median stream flow ranges from 0.0001 to unity, although low flows in most, but not all, smaller streams are protected by pass‐by flow requirements. Temporal changes in surface water withdrawals track gas well completion activity, rather than changes in operational strategies, and while reuse of wastewater has increased since 2009, freshwater accounted for 75% of water used in hydraulic fracturing through the peak in gas well completion activity. Stream flow alteration by frac water withdrawals is evaluated at 300 sites in PennsylvaniaOne fifth of withdrawals are on small streams with drainage areas under 100 square kilometersActual diversions are less than permitted limits but can approach 10% of median stream flows

Water‐extractable organic matter (WEOM) in soil affects contaminant mobility and toxicity, heterotrophic production, and nutrient cycling in terrestrial and aquatic ecosystems. This study focuses on the influences of land use history and agricultural management practices on the water extractability of organic matter and nutrients from soils. Water‐extractable organic matter was extracted from soils under different crop rotations (an annual rotation of wheat–pea/bean–wheat–flax or a perennial‐based rotation of wheat–alfalfa–alfalfa–flax) and management systems (organic or conventional) and examined for its concentration, composition, and biodegradability. The results show that crop rotations including perennial legumes increased the concentration of water‐extractable organic carbon (WEOC) and water‐extractable organic nitrogen (WEON) and the biodegradability of WEOC in soil but depleted the quantity of water‐extractable organic phosphorus (WEOP) and water‐extractable reactive phosphorus. The 30‐d incubation experiments showed that bioavailable WEOC varied from 12.5% in annual systems to 22% for perennial systems. The value of bioavailable WEOC was found to positively correlate with WEON concentrations and to negatively correlate with C:N ratio and the specific ultraviolet absorbance of WEOM. No significant treatment effect was present with the conventional and organic management practices, which suggested that WEOM, as the relatively labile pool in soil organic matter, is more responsive to the change in crop rotation than to mineral fertilizer application. Our results indicated that agricultural landscapes with contrasting crop rotations are likely to differentially affect rates of microbial cycling of organic matter leached to soil waters.

This study focuses on characterizing the contributions of key terrestrial pathways that deliver dissolved organic carbon (DOC) to streams during hydrological events and on elucidating factors governing variation in water and DOC fluxes from these pathways. We made high‐frequency measurements of discharge, specific conductance (SC), and fluorescent dissolved organic matter (FDOM) during 221 events recorded over 2 years within four Vermont (USA) watersheds that range in area from 0.4 to 139 km2. Using the SC measurements, together with statistical information on discharge, we separated the event hydrographs into contributions from three terrestrial pathways, which we refer to as riparian quickflow, subsurface quickflow, and slow‐flow groundwater. The pathway discharges were used as input to a mixing model that closely approximated sub‐hourly streamwater DOC concentrations as measured with the FDOM sensors. Subsurface quickflow, comprised of pre‐event water, was the leading contributor to streamwater DOC fluxes, while riparian quickflow, comprised of event water, was the second‐leading contributor to streamwater DOC fluxes, despite comprising the smallest proportion of streamflow yield among the three end‐member pathways. Fixed‐effects regression analysis revealed that the relationship between DOC fluxes from the end‐member pathways and event magnitude was consistent across the four watersheds. This analysis also showed that DOC fluxes from the quickflow pathways increased significantly with temperature and varied inversely, but weakly, with catchment antecedent wetness. We believe that our approach, which leverages in‐stream sensors that enable high‐frequency measurements over extended periods, may be applicable for evaluating controls on DOC export from other watersheds within and beyond our study region. Calculations of a mixing model reproduce sub‐hourly observations of streamwater dissolved organic carbon (DOC) concentrations in four watershedsStreamwater specific conductance can be used to quantify discharge from terrestrial flowpaths and constrain the DOC mixing modelDOC fluxes from principal flow paths vary discernibly with event magnitude, temperature, and antecedent wetness Calculations of a mixing model reproduce sub‐hourly observations of streamwater dissolved organic carbon (DOC) concentrations in four watersheds Streamwater specific conductance can be used to quantify discharge from terrestrial flowpaths and constrain the DOC mixing model DOC fluxes from principal flow paths vary discernibly with event magnitude, temperature, and antecedent wetness

•PA residents may file complaints to report hazards from oil/gas activity.•Counties with higher socio-economic status were more likely to file complaints.•# of complaints filed were similar in counties with different racial compositions.•Higher % minority racial group associated with fewer confirmed impairments.•Results highlight possible procedural inequities in the oil/gas complaint process.

The rapid expansion of unconventional oil and gas development (UD), made possible by horizontal drilling and hydraulic fracturing, has triggered concerns over groundwater contamination and public health risks. To improve our understanding of the risks posed by UD, we develop a physically based, spatially explicit framework for evaluating groundwater well vulnerability to aqueous phase contaminants released from surface spills and leaks at UD well pad locations. The proposed framework utilizes the concept of capture probability and incorporates decision‐relevant planning horizons and acceptable risks to support goal‐oriented modeling for groundwater protection. We illustrate the approach in northeastern Pennsylvania, where a high intensity of UD activity overlaps with local dependence on domestic groundwater wells. Using two alternative models of the bedrock aquifer and a precautionary paradigm to integrate their results, we found that most domestic wells in the domain had low vulnerability as the extent of their modeled probabilistic capture zones were smaller than distances to the nearest existing UD well pad. We also found that simulated capture probability and vulnerability were most sensitive to the model parameters of matrix hydraulic conductivity, porosity, pumping rate, and the ratio of fracture to matrix conductivity. Our analysis demonstrated the potential inadequacy of current state‐mandated setback distances that allow UD within the boundaries of delineated capture zones. The proposed framework, while limited to aqueous phase contamination, emphasizes the need to incorporate information on flow paths and transport timescales into policies aiming to protect groundwater from contamination by UD. Capture probability is used to assess domestic well vulnerability to aqueous phase contamination from unconventional oil and gas developmentMultiple groundwater model conceptualizations provide more conservative risk estimates than a single calibrated modelThe approach provides a scientifically defensible basis for groundwater and human health protection regulations such as setback distances

In the work reported here, I examine colloid‐facilitated solute transport through a natural porous medium that exhibits grain‐scale heterogeneity in chemical composition. I conducted six column experiments using the contaminant hydroxyatrazine (HA) and either homogeneous clay‐colloid suspensions or mixtures of clay colloids and dissolved organic matter. The transport of colloids composed of illite and montmorillonite, although not conservative, is only slightly affected by rate‐limited mass transfer reactions with the oxide‐coated aquifer material. Coinjection of HA with monomineralogic suspensions of illite colloids produces modest increases in effluent concentrations of HA relative to experiments performed with colloid‐free influent solutions. Montmorillonite colloids have a much higher capacity to bind HA than illite colloids and substantially increase the transport rate of HA in the column experiments. Suwannee River humic acid (SRHA) does not strongly complex aqueous phase HA, but its addition to column influent mixtures that contain HA and montmorillonite suppresses HA adsorption to the clay colloids and promotes a decline in HA mobility. A model for colloid‐facilitated transport, formulated to account for nonuniformity in the kinetics of colloid deposition and HA adsorption, closely describes HA transport in experiments performed with monomineralogic clay‐colloid suspensions, heterogeneous suspensions made of both illite and montmorillonite, and mixtures of montmorillonite and SRHA.