Open Woda Thesis .Pdf

The Pennsylvania State University

The Graduate School

Department of Geosciences

WORKING WITH CITIZEN SCIENTISTS AND HOMEOWNERS IN PENNSYLVANIA

TO UNDERSTAND HYDROCARBON-RELATED CONTAMINATION OF WATER

RESOURCES

A Thesis in

Geosciences

Joshua Craig Woda

Submitted in Partial Fulfillment of the Requirements for the Degree of

Master of Science

May 2019

The thesis of Joshua Craig Woda was reviewed and approved* by the following:

Susan L. Brantley Distinguished Professor of Geosciences Thesis Advisor

Katherine H. Freeman Evan Pugh University Professor

Demian Saffer Professor of Geosciences Department Head, Geosciences

Mark E. Patzkowsky Professor of Geosciences Associate Head for Graduate Programs and Research

*Signatures are on file in the Graduate School

iii

ABSTRACT

Extensive shale gas development around the world has generated concerns about environmental impacts such as migration of natural gas into surface and groundwater resources. However, in Pennsylvania, these concerns are only the most recent in the state’s long history of hydrocarbon development which spans back to the mid 1700’s. Both recent and legacy forms of hydrocarbon extraction have the potential to leak hydrocarbons and other contaminants into the subsurface and the atmosphere. Legacy hydrocarbon development includes old and abandoned oil, gas and coal extraction. Yet, a lack of understanding exists regarding how hydrocarbons behave in the subsurface during leakage and how they may change aquifer quality during prolonged leaks. There is also a knowledge gap regarding how legacy extraction techniques cause contamination years after their abandonment. This thesis evaluates hydrocarbon leakage and its associated consequences from unconventional oil and gas activity in a specific case study, and broader regional leakage trends from various forms of hydrocarbon extraction statewide. In Chapter 2, I studied high gas concentrations in surface and groundwater at a site near problematic Marcellus Shale gas wells to determine the geological explanations and geochemical implications. I found that prolonged methane migration had changed redox conditions in the aquifer, mobilizing metals and transforming sulfate to sulfide. I also explored geochemical tracers in the waters that may be useful diagnostic tools to find other areas with recent methane migration (i.e., possible leakage related to human activity) rather than methane that has been present for a long time (i.e. methane from natural sources). In addition, I identified geologically “risky” areas where drillers should consider taking extra precautions. In Chapter 3, the objective was to collect stream methane samples in a broad area of the Appalachian Basin focused on Pennsylvania to learn which hydrocarbon sources are leaking methane. The goal was to investigate background concentrations and leakage while helping local groups and organizations learn about and understand hydrocarbon leakage in their areas. Throughout this study I helped organize sampling campaigns in streams across Pennsylvania and into West Virginia using the help of citizen scientists. The nonscientists helped with sampling and finding problematic areas at the same time

that we emphasized communication between scientists and the community. Many citizen scientists brought with them local concerns (usually not previously known to scientists) about contaminant source locations within their respective regions. This helped immensely with scientists’ decisions about where to focus resources during sampling. Sampling campaigns targeted areas with different techniques of hydrocarbon extraction, ranging from old to modern. Elevated stream methane was mostly observed around conventional oil and gas and coal mining activity. Specifically, the data showed that more leakage may be occurring near the oldest subset of conventional oil and gas wells. The methane concentrations in streams near wells emplaced before 1955 were often highest. In streams near abandoned coal mines, it was commonly observed that methane dissolved in water is escaping from the mines. Such methane ultimately emits into the atmosphere across Pennsylvania from flooded coal mines at a rate of at least 150,000 kg/y. This value is much lower than the rate of emission from abandoned oil and gas wells, newly emplaced unconventional wells, or non-flooded coal mines. Finally, in many areas of suspected hydrocarbon leakage, I discovered orange- colored springs reminiscent of abandoned mine drainage (AMD) but which were not associated with known coal mining activity. Geochemical analysis of these springs suggests that this phenomenon is a previously-unreported byproduct of some cases of hydrocarbon migration. This new type of discharge is referred to here as Gas Leak Drainage (GLD). GLD differs from AMD in that it is characterized by low sulfate concentrations, relatively low specific conductivity, and high hydrocarbon concentrations. The findings presented in this thesis highlight the importance of monitoring and research associated with the environmental impacts of both current and legacy hydrocarbon extraction. The extent of these types of methane leakage and associated migration in the subsurface remains poorly characterized. Additionally, legacy environmental impacts can continue to grow and evolve as infrastructure ages, justifying the need for continued research. The impacts of legacy hydrocarbon development should also be taken into account when considering future development.

Table of Contents

LIST OF FIGURES ...... vi

LIST OF TABLES ...... xv

ACKNOWLEDGEMENTS ...... xvi

Chapter 1 Introduction to thesis ...... 1

1.1 Summary of Chapter 2 ...... 1 1.2 Summary of Chapter 3 ...... 2

Chapter 2 Sugar Run Case Study ...... 6

2.1 Introduction ...... 6 2.2 Methods ...... 9 2.3 Results ...... 11 2.4 Discussion ...... 15 2.5 Conclusion ...... 27 2.6 Figures ...... 29

Chapter 3 Stream Methane, Citizen Science and Metal Rich Springs ...... 35

3.1 Introduction ...... 35 3.2 Methods ...... 42 3.3 Results ...... 46 3.4 Discussion ...... 64 3.5 Conclusion ...... 76 References for Chapter 2 and 3 ...... 78

Appendix A Supplemental Information for Sugar Run Case Study ...... 86

Figures for Chapter 2...... 105

Appendix B Additional Figures for Chapter 2 and Chapter 3 ...... 121

Appendix C Meshoppen Creek Case Study ...... 136

Appendix D Ludlow Seepage Case Study ...... 140

Appendix E Tables For Chapter 2 and 3 ...... 144

LIST OF FIGURES

Figure 2.1: (A) Study area showing Lycoming County, Pennsylvania. (B) Expanded view showing all active unconventional gas wells that have (red) or have not (green) received well-integrity violations from the state regulator (21), PA DEP (also shown on (C)). (C) Expanded view of Sugar Run watershed showing sample locations in streams (triangles) and bubbling seeps (arrows). Outcrop locations where methane was detected in the air near fractures are depicted as orange circles. Average dissolved methane concentrations in stream sites are shown as grey triangles (intensity of grey is contoured with respect to concentration as shown in legend and Appendix A, Figure S4). (D) Cross section S-S’, defined on (C), roughly follows the plunge of the Nittany Anticlinorium to the east. On this cross section, well API # 081-20292 intersects the Marcellus Formation at 997 meters and the depth of 0 represents sea level. The well pad where well 081-20292 is drilled contains only that gas well. A lateral from that well follows the Marcellus roughly perpendicular to the cross section and is shown in (C) as a dashed line. Where new data are presented herein for seep, homeowner, and air samples, they were sampled within the dotted circle in Figure 1C...... 29

Figure 2.2: (A) Time series plots of dissolved methane concentrations from homeowner (HO) water wells 1 – 10 sampled in the study region near Sugar Run in Lycoming County. Plotted data for water well HO8 includes two water wells (A & B) sampled on the same property. Vertical solid lines indicate spud dates for unconventional gas wells cited for cementing/casing violations within a 5 km radius of Seep 1.5. In addition, the dashed and dotted lines indicate the spud date and hydraulic fracturing date for well API # 081-20109, respectively. Samples indicated by open symbols are not discussed further and were reported by the gas company as part of the initial (32, 33) or subsequent investigations (31). Concentrations plotted after 10/26/2016 were sampled and measured by our team (Table E6-E8). B) A time series plot of dissolved ethane concentrations for HO6, HO5, HO4, HO3, and HO2 for data reported online (33), by PA DEP (31), or in this study. Horizontal dashed lines represent median and maximum concentrations for dissolved methane (n = 967) or ethane (n = 897) reported for Lycoming county samples outside the study region between 1995 and 2014. These samples were analyzed from published reports (35), or collected as “pre-drill” data by companies before drilling gas wells and were released to PA DEP and shared with Penn State University (34). For ethane, only 10 of 897 samples contained detectable ethane: line labelled “Lycoming Maximum (uncensored)” on (B) summarizes the highest pre-drill concentration for samples with detected ethane. Ethane concentrations were not reported for samples collected and measured for ethane indicated by arrows labeled “no ethane detected in HO5 & HO6” on (B). (Data were censored but no reporting limits were indicated). Ethane was not analyzed in HO4 before drilling commenced. The maximum and median values for the

vii

Lycoming data plotted in (A) and (B) are for samples collected outside the dotted circle in Figure 1C...... 30

Figure 2.3: Color contours showing the mean curvature, as defined by published algorithms (62), calculated for the top of the Marcellus formation (smoothed using a 2 km moving average window to minimize artifacts from modeling). Warmer colors indicate where the mean curvature is greatest and likely to have caused the greatest density of vertical fractures that may allow upflow of gas from depth. Curvature was calculated using Move’s surface geometry analysis tool. Labels show the locations of Sugar Run and the API # 081- 20292 well. The Marcellus outcrops 3.3 km from the Sugar Run location point, i.e. where the white area cuts into the Sugar Run valley as shown in the figure...... 31

Figure 2.4: Iron concentrations in groundwaters from HO4, HO5, and HO6 plotted versus time. Concentrations were normalized to the maximum values (0.58 mg/L, 0.14 mg/L and 3.02 respectively) (33). The HO4 sample collected on 7/18/2012 was slightly offset to avoid overlap. Downward arrows represent the reporting limit for censored analyses. For HO5, no reporting limit was indicated, so it is shown as an arrow at the concentration equivalent to the smallest reported concentration in that report. Data were derived from reports (31–33), with one HO4 sample collected on 7/26/2017 from this study (see main text)...... 32

Figure 2.5: Graphs of groundwater chemistry from homeowner wells sampled in this study (symbols labeled HO), the Lycoming County groundwater dataset (Lyco GW, empty black circles), and presumably contaminated sites in neighboring counties (orange symbols), as described in text. A) Concentrations of sulfate plotted versus methane show that most waters are either high in sulfate or methane but not both, as expected based on thermodynamic equilibrium. Waters from the Sugar Run area that are presumed to be experiencing a new methane influx (HO5 and HO4) plot in the upper right quadrant along with water from wells from the presumably contaminated sites. After several months, sulfate is inferred to be reduced and waters no longer plot in the upper right quadrant (e.g., compare HO4 in 2017 to 2016; empty red inverted triangles). B) Concentrations of iron (Fe) plotted versus methane show that generally, Fe concentrations are low with elevated methane concentrations. Some wells from the Sugar Run study area and presumably contaminated sites contain high methane and iron concentrations. Lyco Gw refers to data provided by the PA DEP, and from published reports (35) as described in text. The PA DEP data are also published online (34). Data from homeowner wells were either sampled in this study (Table E6-E8) or are published (31–33)...... 33

Figure 2.6: Generalized schematic of an evolving methane plume released from a shale gas well. The inserted plot shows how water chemistry changes over time. The plot is patterned from data from homeowner well HO4 (Table E6-

viii

E8). Two separate “plumes” are drawn for early and later time periods. The outer edge of the plume (dark gray) represents the zone of methane oxidation coupled to metal reduction. The inner section of the plume (light gray) represents the zone of methane oxidation coupled to sulfate reduction (creation of H2S). Once electron acceptors are consumed, hydrocarbons flow unhindered from the subsurface to the surface resulting in even higher concentrations and unaltered isotopic values, if a leak continues. This diagram is not drawn to scale or to reflect the geology of Sugar Run...... 34

Figure 3.1: Map illustrating the spatial distribution of known conventional and unconventional oil wells, gas wells and coal mines in Pennsylvania...... 38

Figure 3.2: Locations of the 533 sites where methane concentrations were measured in streams: increasing circle sizes correspond with increasing site- aggregated mean methane concentrations (i.e. larger circles correspond with larger methane concentrations). The highest individual measured methane concentrations in the dataset for streams (i.e., not site aggregated means) equaled 68.5 µg/L in Meshoppen Creek (Susquehanna County see Appendix C) and 76.7 µg/L in Sugar Run (Lycoming County)...... 47

Figure 3.3: Histogram of the site-aggregated means for 533 stream sites including all data from this study, published work (2, 19, 24, 25) and the SRBC. Of the 533 sites, site-aggregated means of only 75 were above 4 µg/L...... 49

Figure 3.4: Mean (left) and median (right) methane concentrations for each snapshot day analyzed as a part of this study. Overall, the means and medians for methane concentrations were higher during the snapshot days in the Allegheny National Forest than elsewhere...... 51

Figure 3.5: Histogram of the 75 site-aggregated mean concentrations for stream sites with methane > 4 µg/L...... 52

Figure 3.6: Schematic showing how samples were binned into categories for samples with methane concentration > 4 µg/L. Samples were binned into 1 of 8 categories. The binning shown here was designed as a quick tool to categorize sites in a preliminary way...... 54

Figure 3.7: Histograms showing the distributions of site aggregated mean methane concentrations for all bins containing more than 5 samples. See text for criteria used in categories...... 56

Figure 3.8: Site-aggregated mean methane concentrations for all sites sampled in Pennsylvania where methane was detected >4 µg/L. Colors indicate the category of each high-methane sample. Circled areas of Pennsylvania highlight areas characterized by different predominant hydrocarbon extraction techniques: northwest is mostly conventional oil and gas, northeast is mostly unconventional gas, and southwest is characterized by conventional oil and gas, unconventional gas, and coal mining ...... 57

Figure 3.9: Histograms of site-aggregated means methane concentrations measured in discharges categorized as acid mine drainage (AMD) or gas leak drainage (GLD) or leaking abandoned wells (LAW) ...... 59

13 Figure 3.10: Methane concentration (µg/L) and δ C-CH4 vs distance downstream from a discharge of AMD (the Gladding discharge) (Table E14- E16). The decrease in methane downstream is attributed to degassing as 13 opposed to oxidation because of the relative constancy in δ C-CH4...... 60

Figure 3.11: (A) Plot of site-aggregated mean methane concentrations for stream sites in the Allegheny National Forest vs distance to the nearest conventional oil or gas well. To calculate distance to well, all oil/gas wells in the ANF in the four counties mentioned in main text were considered. (B) Same as (A) except that only conventional oil or gas wells spudded prior to 1955 were considered. (C) Same as (A) and (B) except only conventional oil or gas wells spudded between 2005 were included. Additional plots for different date ranges are shown in the appendix (Figure B-13) ...... 68

Figure 3.12: A plot of Ca/Na vs SO4 for different types of waters: abandoned mine drainage (AMD)(this study, Hedin et al. (95) Cravotta et al. (91)), produced waters from conventional (conventional brines) and unconventional wells (unconventional brines) (USGS Produced Waters Database), gas leak drainage (GLD), including waters from abandoned wells. Waters from leaking abandoned wells are lumped together here into the region of GLD. GLD appear chemically distinct from other sources, except for one anomalous sample. Waters sampled in casing of leaking abandoned wells contain brine salts similar to ABB and are thus similar to discharges receiving brine-rich formation waters such as the salt spring in Salt Spring State Park located in Montrose, PA. Suspected GLD samples represent the subset of samples located within 2 km of a known coal mine. One GLD sample (Sample SRS 1.5 located near Sugar Run, Lycoming) contains an anomalously low Ca/Na ratio. This same sample (SRS 1.5)was also anomalously elevated in pH (9.01), low in metals, and appeared visually distinct from other seepages (i.e. dark black sediment surrounding the bubbling area). The double-sided arrow labeled “Ca/Na shallow groundwater” represents the median range of Ca/Na ratios for groundwaters sampled in northeastern PA aquifers of the Catskill, Lock Haven and Battelle formations and reported in previous work (46)...... 75

Figure A1: Photograph of Sugar Run showing Seep 1.5 (left) and the stream (right) ...... 105

Figure A2: A) Map of the Sugar Run watershed (colored blue), sample locations (triangles), seepage locations (blue circles), and outcrops (orange circles). B) Map of Sugar Run sampling locations with seeps and the average stream methane concentrations at each site. Active bubbling was observed in the

stream and seeps along this stretch. The dashed line and red and green symbols in (A) are described in caption for Figure 1...... 106

Figure A3: (A) Map showing measurements of methane concentrations in air measured on eight different days in the study region. Distinct colors represent different sampling days while the sizes of the symbols indicate concentration. (B) Methane concentration map for air in the highest-concentration region. Over 24,000 measurements were collected over 8 days. Measurements were completed by walking around the study region and holding a surface bell probe within one meter of the ground surface. At outcrops, the surface bell probe was placed directly on fractures...... 107

Figure A4: Methane concentration in stream water plotted versus distance downstream along Sugar Run in Lycoming County, Pennsylvania. The zero position was defined as the most upstream stream site (SR 8) that was sampled. Triangles are samples collected and analyzed by Heilweil et al. (24) and circles represent samples from this study. Largest methane concentrations in the stream consistently were measured at sites SR 1.5 and SR 1.55 (near two methane-rich seeps). The horizontal section of well 081- 20292 crosses underneath Sugar Run at the line, and locations downstream of this location are structurally up dip...... 108

Figure A5: Block diagram of the Nittany Anticlinorium, showing locations of Sugar Run and gas well 081-20292. The surface impression of the Nittany Anticlinorium can be observed to the west in the figure...... 109

Figure A6: Isotopic values for methane from samples collected at sites in Figure S2 plotted versus 1/[CH4] where [CH4] refers to CH4 concentrations in µg/L. In the blue region, δ13C increases with decreasing methane concentration, as expected for fractionation during methane oxidation (9). A few samples more negative than -28.3‰ plot outside the blue area and approach an inferred biogenic endmember (9). Following previous literature (24), samples with the largest methane concentrations were used to estimate the δ13C signature of the thermogenic endmember (-28.3‰; red solid line). The initial concentration of thermogenic methane in water before oxidation likely varies temporally and spatially. The [CH4] concentration of 1000 µg/L (1/[CH4]=0.001) is chosen to calculate expected Rayleigh fractionation (blue dashed lines). Two scenarios are considered for two values of fractionation factors () of 1.013 and 1.025, derived from previous work (116)...... 110

Figure A7: Concentrations of CH4, Fe, and As in the most methane-rich seep (1.6) plotted versus time. For simplicity, concentrations were normalized by dividing by the maximum values: Fe (17.8 mg/L), As (0.0058 mg/L), CH4 (8.6 mg/L)...... 111

Figure A8: R/Ra and He/Ne ratios of all collected noble gas samples including water from the salt spring at Salt Spring State Park, Pennsylvania, and

homeowner wells in Sugar Run. Dashed curves represent mixing between noble gases from ASW at 10 C and radiogenic noble gases (including crustal and mantle sources). Three scenarios with varying contributions of mantle He (i.e., 0%, 25%, and 50% by mass) are indicated. All water and gas samples reported for the Sugar Run area are located on the curve representing mixing between ASW and pure crustal components. This precludes the presence of significant mantle noble gas in these samples...... 112

4 20 36 Figure A9: He/CH4 and Ne/ Ar ratios of homeowner well water samples (red triangles) and salt spring water (blue circle). Short vertical and horizontal 4 lines within sample marks represent corresponding error bars for He/CH4 and 20Ne/36Ar, respectively. Predicted values are also plotted for four scenarios: (1) slow upward advection of brine containing dissolved methane (black line with upward arrow); (2) diffusion of gas from depth in aqueous solution (green dashed curve); (3) upward advection of two phases (i.e., free gas phase and brine phase) (i.e., free gas phase and brine phase) from depth (red dashed curves); and (4) fast upward advection of free phase gas with minor mixing of microbial gas in the shallow aquifer (vertical line with downward arrow). Note, we assume that microbial gas contributes methane but not noble gases. In all of these scenarios, we assumed the starting point of noble gas fractionation was the composition of Marcellus Formation 4 production gas (green rectangle). The pink rectangle represents He/CH4 and 20Ne/36Ar of natural gas samples from Upper Devonian formations (15). 20Ne/36Ar ratios of homeowner wells HO2 and HO4 are consistent with scenario (4) (fast upward flow of Marcellus gas mixed with biogenic gas), and is not consistent with scenarios (1), (2), nor (3). We hypothesize that relatively quick migration of Marcellus gas along faults in a free gas phase best explains why the homeowner well waters preserve the original 20Ne/36Ar signature...... 113

Figure A10: Plot of ratios of Cl/Br concentrations (both as mg/L) versus chloride for samples collected from Seep 1.5, Seep 1.6, stream sample SR 1, and homeowner water wells HO1, HO2, HO3, and HO4. Most waters plot within or very near dilute groundwater except for Seep 1.6. The plot, adapted from previous work, shows generalized regions of different water types (11). The circle labeled “ABB Brine” represents chemistry of directly sampled ABB...... 114

Figure A11: 87Sr/86Sr vs Sr/Ca molar ratio in samples taken near Sugar Run (stream samples labelled SR, homeowner well waters labelled HO, groundwater under the stream sampled by piezometers labelled Piezo, and seeps) and brines collected from oil and gas wells (97). The plot shows the possibility that there are two different sources of brine salts that have impacted waters in the Sugar Run valley. The black box encompasses the variation in chemical fingerprint of brines originating from the Marcellus as defined by previous work (10). Samples collected at Seep 1.6, HO4, and SR 1.5 Piezo have the distinctively higher Sr/Ca and lower 87Sr/86Sr that are typical of Marcellus brines (10). In contrast, stream sites SR 1 and SR 1.5,

xii

and Seep 1.5 show chemical characteristics more consistent with radiogenic upper Devonian formations such as the Bradford Group and Venango Group. Some data values were reproduced here from previous works (19, 24)...... 115

Figure A12: A) Plot of chloride versus methane concentrations in groundwater samples as labelled (see text in SI and main body). Most high methane water in Lycoming county contains high concentrations of chloride. However, waters from almost all locations with presumable contamination, including Sugar Run, contain relatively low concentrations of chloride. B) Ca/Na (mass ratio) versus methane concentrations in groundwater samples as labelled. Most high methane samples contain Ca/Na < 0.52. However, a small subset of presumably contaminated samples contains Ca/Na > 0.52...... 116

Figure A13: A contour map of depth of the top of the Marcellus Shale (dashed lines document where erosion has removed the shale and negative labels on contours denote depth below sea level in meters). The surface was calculated from cross sections, gas well reports, and topographic data using ordinary kriging. Coordinates are in UTM, Zone 18 North...... 117

Figure A14: Normalized concentrations of methane, iron, and sulfate versus time for groundwater from homeowner well HO4. Concentrations are normalized to the largest concentration for each analyte: methane = 61.6 mg/L, sulfate = 18.7 mg/L, iron = 0.576 mg/L. Concentrations are only plotted where at least two of the three analytes were available on a given date (Table E6-E8)...... 118

Figure A15: The set-up for the collection of gas samples for noble gas analyses at bubbling seep sites...... 119

Figure A16: F(4He), F(22Ne), F(84Kr), and F(132Xe) value of collected water and gas samples. F(4He), F(22Ne), F(84Kr), and F(132Xe) are measured 4He/36Ar, 22Ne/36Ar, 84Kr/36Ar, 132Xe/36Ar ratios normalized to corresponding air values. F values of ASW (air saturated water) at the temperature of 10 C (light blue line) and air (horizontal black line) are shown for comparison (40). .. 120

Figure B1: Methane concentrations measured in water samples from one homeowner water well in the Sugar Run Valley, Lycoming PA. Methane increases inconsistently (roughly 10 mg/L per year) over a roughly 5-10 year period from pre-drill concentrations. Dates and methane concentrations are not provided to protect homeowner identity...... 121

Figure B2: Ethane concentration over time for one homeowner water well in the Sugar Run Valley, Lycoming PA. Ethane increased (roughly 0.5 mg/L per year) over a roughly 5-10 year period from pre-drill concentrations. Dates and ethane concentrations cannot be provided for privacy reasons...... 122

Figure B3: Propane concentration over time for one homeowner water well in the Sugar Run Valley, Lycoming PA. propane increased (roughly 0.06 mg/L per

xiii

year) over a roughly 5-10 year period from pre-drill concentrations. Dates and propane concentrations cannot be provided for privacy reasons...... 123

Figure B4: Iron concentration over time for one homeowner in the Sugar Run Valley, Lycoming PA...... 124

Figure B5: Sulfate concentration over time for one homeowner in the Sugar Run Valley, Lycoming PA...... 125

Figure B6: Alkalinity over time for one homeowner in the Sugar Run Valley, Lycoming PA...... 126

Figure B7: Pictures of leaking abandoned wells (ANF Leaking Well 1 right and ANF leaking well 2 left) in the Allegheny National Forest. Locations: 41.5095, -78.7926 (right) and 41.6121, -78.7913 (left) ...... 127

Figure B8: Map of sample locations leaking abandoned wells (yellow diamonds), gas leak drainage (red stars), and AMD discharges (red circles)...... 128

Figure B9: Map of sample locations in northwestern Pennsylvania. Concentrations are site-aggregated means where more than one sample was analyzed. See text for how the categories were defined...... 129

Figure B10: Picture of the Morrison seepage (left) where methane and ethane rich waters flow out of a depression in the ground. White strings are assumed to be bacteria. A smell of hydrogen sulfide is present at this seepage. (Right) is a close up of orange “slimy” bacteria at Seep 1.5 near Sugar Run, Lycoming. ... 130

Figure B11: Photo of the Honey Pot Outflow near Wilkes-Barre, PA where waters flow out of a mine opening...... 131

Figure B12: Picture of the Gladding AMD discharge near Pittsburg, PA (left) and where flow form this discharge tumbles off a waterfall (right) into Miller’s Run...... 132

Figure B13: Plots of methane concentration in stream waters vs distance to the nearest conventional oil or gas well spudded between different date ranges. See text for more information ...... 133

Figure B14: Map of sample locations in southwestern Pennsylvania. Concentrations are site-aggregated means where more than one sample was analyzed. See text for how the categories were defined...... 134

Figure B15: Map of sample locations in northeastern Pennsylvania. Concentrations are site-aggregated means where more than one sample was analyzed. See text for how the categories were defined...... 135

xiv

Figure C1: Map of the Meshoppen Creek watershed in Susquehanna County, PA. Sample locations (maroon circles) are noted where methane concentrations are indicated by circle size. A circle is drawn around Dimock, the site of known hydrocarbon leakage from shale gas development (Hammond et al., 2016). The sample location Meshoppen US 2 is within the circled region. Small open circles represent unconventional gas wells as denoted by the legend...... 137

Figure C2: Zoomed in map of Meshoppen Creek near Dimock, PA (left). Known wetland locations (plotted - left) validate physical observations (right) that the stream is extremely boggy. The photo on the right hand side was located near Meshoppen US 2, see table E17...... 138

Figure D1: Orange coloration associated with a spring emerging from the hillslope near US 6 (can be seen in the top right of the image). The white colored fibrous matting is tentatively attributed on the basis of visual observation alone to bacterial mats. Sampling at this site (Tables E11-E13) was conducted on three separate occasions...... 140

Figure D2: Orange sediment observed along US 6 near the Ludlow seepages. When water levels are higher, orange discoloration is much more obvious...... 141

Figure D3: Screenshot from the PA Oil and Gas mapping tool. The “X” marks the approximate location of the largest Ludlow seepage (41.712, -78.912), located along route 6. The seepages are located in an area of gas wells, many of which appear on a USGS base map (open circles) but do not appear in the PA DEP’s database (where the only wells are shown here as colored circles). ... 142

Figure D4: East Branch Field Well 1461P is located a couple hundred feet away from the start of the Ludlow iron seepages shown in Figure D-1. The dark color surrounding the base of the well is an unknown substance that may be oil related. According to the PA Oil and Gas mapping tool, the well is an “Multiple Well Borehole Type.” The spud date is not listed and the location given in the tool appears to be incorrect as the well is situated on the wrong side of US 6 on the map (Figure 1)...... 143

LIST OF TABLES

Table 1: List of organizations that contributed samples, information or help in Chapter 3. .... 5

Table 2: Talks and outreach provided to citizen scientists by Josh Woda...... 5

Table E1: Table E1. Generalized timeline of events in the study area ...... 144

3 Table E2: Table E2. Noble gas and methane concentrations (cm STP/gH2O) of water samplesa...... 147

Table E3: Noble gas volume fractions (cm3/cm3) of gas samplesa ...... 148

Table E4: Noble gas isotopic ratios of water and gas samples ...... 149

Table E5: Hydrocarbons – Chapter 2 ...... 151

Table E6: Cations – all mg/L unless otherwise stated – Chapter 2 ...... 163

Table E7: Anions, YSI measurements and Sr isotopes – Chapter 2 ...... 173

Table E8: Measured hydrocarbons in waters sampled at wellheads of leaking or abandoned wells cited in Chapter 3 ...... 184

Table E9: Cation concentrations in waters sampled at wellheads of leaking or abandoned wells cited in Chapter 3 ...... 185

Table E10: Anion concentrations and YSI measurements in waters sampled at leaking or abandoned wells cited in Chapter 3 ...... 186

Table E11: Measured hydrocarbons in GLD and Salt Springs as cited in Chapters 2 and 3 (not including LAWs) ...... 187

Table E12: Measured cations in GLD and Salt Springs as cited in Chapters 2 and 3 (not including LAWs)...... 191

Table E13: Measured anions and YSI parameters in GLD and Salt Springs as cited in Chapters 2 and 3 (not including LAWs)...... 195

Table E14: Measured hydrocarbon concentrations and isotopes in AMD discharge cited in Chapter 3 ...... 198

Table E15: Measured cation concentrations (mg/L) in AMD discharge cited in Chapter 3.... 200

Table E16: Measured anion concentrations and YSI parameters in AMD discharge cited in Chapter 3 ...... 201

Table E17: Stream methane samples from all datasets cited in and presented in Chapter 3 ... 202

Table E18: Sites with site-aggregated mean [CH4] > 4 µg/L and observations from follow up investigations ...... 244

xvi

ACKNOWLEDGEMENTS

Firstly, I would like to give a thank you to Susan Brantley and my committee members Kate Freeman and Demian Saffer. I would also like to thank John Hooker for additional comments and help during my Thesis defense.

I want to give a special thanks to Tao Wen for the extensive support in sampling, writing, thinking and most importantly being a friend. Tao provided senior guidance, support and friendship during this roller coaster ride. I will keep fond memories of our sampling trips and late-night phone calls.

I want to give thanks to everyone in the Brantley group and elsewhere including

Laura Liermanm, Matt Gonzales, Callum Wayman, Virginia Marcon, Xin Gu, Kalle Jahn,

Joanmarie Del Vecchio, Jennifer Williams, Todd Sowers, Lauren Dennis, Allison Herman,

Greg Mount and hundreds of citizen scientists who cannot all be named at this time for all of their help in the field and laboratory. I want to give a special thanks to Virginia Marcon for sharing a room in Hosler these last two years and providing friendship, guidance, color help and mostly someone to talk too when others were not readily and physically able too.

I acknowledge the National Science Foundation IIS Award 1639150 (to S.L.B.)

(Pennsylvania State University), the Pennsylvania State University for purchase of the Gas

Rover, Paul D. Krynine scholarship, Charles E. Knopf Sr. Memorial Scholarship and

Richard R. Parizek Graduate Fellowship and a gift to Pennsylvania State University for the

Pennsylvania State University General Electric Fund for the Center for Collaborative

Research on Intelligent Natural Gas Supply Systems for providing funding for this work.

xvii

I would like to thank the department of Geosciences for providing great collogues and a home to work and study during my time at Penn State

Finally, thank you to my parents, Adam and Erin for providing me an unbelievable support network throughout this entire process. Without support from my family none of this work would have been possible.

Research Disclaimers: This material is based upon work supported by the National

Science Foundation under Award No. 1639150.

Any opinions, findings, and conclusions or recommendations expressed in this publication are those of the author and do not necessarily reflect the views of the National

Science Foundation or Pennsylvania State University or any of the other funding entities.

Chapter 1

Introduction

Pennsylvania has a long history of hydrocarbon extraction which extends back to the 1700’s with the beginning of the colonial era of coal mining. Throughout the years, the state has undergone extensive coal mining and oil and gas development including the drilling of the world’s first commercial oil well in 1859. Unconventional shale gas development is the newest hydrocarbon extraction technique implemented in Pennsylvania. Recent drilling has reawakened concerns about the needs for monitoring and environmental impacts associated with all forms of hydrocarbon extraction. In each chapter of this thesis, I combine field studies with geochemical analyses to shed light on poorly understood effects associated with leakage from hydrocarbon extraction. Chapter 2 highlights a case study surrounding methane migration and associated water chemistry changes in a region with reportedly problematic unconventional (shale gas) wells. Chapter 3 focuses on a stream methane study which used the help of citizen scientists to locate and understand leaking hydrocarbon sources throughout the state of Pennsylvania. Sampling in Chapter 3 focused on regions developed by legacy extraction techniques (i.e. coal mining and conventional oil and gas development).

1.1 Summary of Chapter 2

In Chapter 2, I studied high gas concentrations in waters at a site near problematic Marcellus Shale gas wells to try to understand geologic causes and geochemical implications. The local geology may explain why methane has discharged for seven years into groundwater, a stream, and the atmosphere. Gas may migrate easily near the gas wells in this location where the Marcellus Shale dips significantly, is shallow (~ 1 km),

and is more fractured. Methane and ethane concentrations increased in local water wells after gas development compared to pre-drill concentrations reported in the region. Noble gas and isotopic evidence are consistent with upward migration of gas from the Marcellus in a free gas phase. This upflow results in microbially mediated oxidation near the surface. Iron concentrations also increased in domestic water wells following the increase of natural gas concentrations. After a several-month transient, both iron and sulfate concentrations dropped in domestic water wells. These observations are attributed to iron and sulfate reduction associated with newly elevated concentrations of methane. These temporal trends, as well as data from other areas with reported leaks, document a way to distinguish newly migrated methane from pre-existing sources of gas unrelated to the new development activity. This work thus documents both geologically risky areas and geochemical signatures of iron and sulfate that could distinguish newly leaked methane from older methane sources in aquifers. I accomplished this work with the help of many co-authors under a grant provided by the National Science Foundation to Susan L. Brantley. I collected field samples with help from co-authors Tao Wen, David Yoxtheimer and David Oakley. I interpreted the results with the additional help of emeritus professor Terry Engelder and University of Michigan professor Clara Castro. Tao Wen and David Oakley were responsible for implementation and writing of the research describing noble gas analysis and structural geologic analysis respectively. Brantley extensively edited this work before its eventual publication in the Journal Proceedings of the National Academy of Sciences (1) in collaboration with all aforementioned co-authors.

1.2 Summary of Chapter 3

In Chapter 3, I discuss my work sampling methane in streams across Pennsylvania and into West Virginia with the help of citizen scientists and Jacob Lemon of Trout Unlimited. The goal was to characterize methane concentrations in streams while locating and understanding potential contaminant sources. Stream water elevated

in methane (>4 µg/L) but not near a wetland was identified as possibly anomalous and more likely to have been contaminated by human activities based on previous work (2). In some cases, such waters were further studied with a more thorough investigation to help pinpoint and understand the possible sources of elevated methane. The largest methane concentrations were predominantly located near wetlands, abandoned coal mines, and in areas of old oil and gas development. In the summary that follows, we focus on investigating the following hypotheses for samples in the Appalachian Basin, with a focus on Pennsylvania: 1) older oil and gas wells (i.e. wells drilled before 1955) leak methane at a higher frequency than younger wells; 2) coal mines do not leak methane when they are flooded; 3) discharges of leaking hydrocarbons from oil and gas wells create a new type of discharge — gas leak drainage — that is chemically distinct from AMD and can sometimes be highly concentrated in ABB salts. Thus the scientist- citizen scientist partnership has resulted in the discovery of a new, easily identified contaminant-containing discharge, namely orange seepage that is elevated in hydrocarbon concentration and heavy metals such as iron, manganese, and arsenic. Completion of this chapter was only possible via collaboration with various organizations and volunteer groups including Trout Unlimited, The US Forest Service, Western Pennsylvania Conservancy, Pitt-Bradford University, local conservation districts, and the Susquehanna River Basin Commission (SRBC). I collected water samples and performed site investigations with Jacob Lemon from Trout Unlimited and TU was partially funded by the National Science Foundation as part of our grant. I also collected samples with Tao Wen from Penn State, Chuck Keeports from the US Forest Service, Fred Zelt (retired geoscientist), Luanne Steffy from the Susquehanna River Basin Commission, Greg Mount from the Indiana University of Pennsylvania, Shaun Donmeyer (undergraduate student at the Pennsylvania State University), and dozens of citizen scientists who cannot all be named at this time, but without whom this work would not have been possible. Some of the stream samples that are collated in Chapter 3 were previously reported in Wendt et al. (2018). Table 1 is a list of all the volunteer groups of citizen scientists that contributed samples, information, or help that made the data in Chapter 3 possible. I also received information and tips on possible sample

locations from citizen scientists as well as from Luke Bobner from the Western Pennsylvanian Conservancy. Todd Sowers provided analytical assistance and isotopic analysis. I interpreted and wrote this chapter in collaboration with my adviser Susan Brantley. Working with volunteers provided a unique and rewarding experience for data collection. Not only did volunteers physically collect samples, they also provided invaluable local knowledge about sites and known contaminant sources. Many of the volunteers in this study were avid hunters or fisherman. The latter was usually the case given the connection with Trout Unlimited. This background allowed many volunteers to readily observe and report features that stood out to them as abnormal. For example, metal seepages were found by volunteers on multiple occasions even without a follow up investigation by PSU researchers. Special attention was placed throughout the project to keep volunteers informed with the state of research. This involved ongoing conversations with volunteers who collected single samples or seasonal sampling. In addition, short informal updates were provided to volunteers during each of the 5 snapshot days performed during this study (see Table 2). Presentation of a poster on the final snapshot day emphasized the project and work in the ANF (the site of that snapshot day). Volunteers observed the poster throughout the day. In addition, a webinar is planned at the end of February to update all Trout Unlimited volunteers about research progress and key findings.

Table 1: List of organizations that contributed samples, information or help in Chapter 3

Organization name Trout Unlimited Pitt-Bradford University Susquehanna River Basin Commission McKean County Conservation district Elk County Conservation District Warren County Conservation District Chartiers Creek Watershed Association Centre County Senior Citizen Environmental Corp West Virginia Rivers Coalition Western Pennsylvania Conservancy The US Forest Service Chartiers Creek Watershed Association The Centre County Pennsylvania Senior Environmental Corps QV Creeks from Fern Hollow Nature Center Penn State TeenShale Network

Table 2: Talks and outreach provided to citizen scientists by Josh Woda

Date Outreach 10/1/2016 Informal talk to volunteers about stream methane during Monongahela, WV snapshot day 6/24/2017 Informal talk to volunteers about stream methane during Pine Creek, PA snapshot day 10/24/2017 Informal talk to volunteers about stream methane during ANF, PA snapshot day 1 4/10/2018 Informal talk to volunteers about stream methane during ANF, PA snapshot day 2 10/30/2018 Poster and talk given to volunteers about water contamination in the ANF during ANF, PA snapshot day 3 2/27/2019 Webinar – “Results of volunteer synoptic water quality monitoring in the Allegheny National Forest” 4/8/2019 Planned oral presentation for US Forest Service and volunteers at ANF service headquarters

Chapter 2

Sugar Run Case Study

[This chapter is reproduced entirely from the following publication J. Woda, T. Wen, D. Oakley, D. Yoxtheimer, T. Engelder, C.M. Castro, S. Brantley (2018). Proceedings of the National Academy of Sciences]

2.1 Introduction

Recent advances in horizontal drilling and high volume hydraulic fracturing have helped the U.S.A. produce significantly more natural gas over the last decade (3). At the same time, shale gas development has led to increased public concern over impacts on water resources in areas of gas production. As of the end of 2017, about 12,000 shale gas wells have been drilled in Pennsylvania (PA). The Marcellus is the most productive shale gas play in the world (3). The most commonly reported water quality impacts in PA have been cases of natural gas migrating into water supplies from gas wells with construction issues (4). However, leaks can be difficult to detect in PA because natural sources of methane (CH4), the predominant hydrocarbon in natural gas, are common (5–8). For example, methane concentrations are often elevated in the region’s groundwaters because methane is produced biologically (9). In addition, thermogenic methane – CH4 produced at depth at higher temperatures – can migrate into aquifers through natural mechanisms that might include transport as a dissolved solute in waters accompanied by salts from formation brine, or, perhaps, as a separate free phase (10–15). To add to the complexity of determining source of gas in aquifers in the Marcellus and other shale plays, methane can

travel for kilometers along boreholes, fractures, faults, bedding plane openings, and through porous sandstones (16–20).

These observations point to the need for investigations into the importance of local geological features in allowing, causing, or accelerating gas migration after drilling and completion of shale gas wells. We also need better methods to detect leaks when they occur. Understanding the causes of migration is important because the gas is an explosion hazard and it can eventually discharge from aquifers into the atmosphere where it is a greenhouse gas (21).

Given that both natural and anthropogenically-affected sources of methane occur in most gas-producing shale plays, definitive assessment of the methane source and its migration pathway can be difficult. Isotopic measurements have been used in many shale gas plays to determine if stray gas was produced by bacterial processes (biogenic) or by higher temperature processes (thermogenic) (8, 9), and to delineate the migration pathway

(16, 18). Water chemistry data have also been used to investigate the methane source (e.g.,

2, 11, 12, 20).

In this study we address three questions: 1) Can we identify geological conditions that exacerbate the potential for methane migration from shale gas wells? 2) What is the impact of newly elevated methane concentrations on aquifer and stream chemistry? 3)

What tracers distinguish human-induced migration of methane versus naturally migrating methane? To answer these questions, we revisit a field site in central PA where high methane concentrations have been highlighted in seeps and groundwater near a small first- order stream named Sugar Run in Lycoming County (Figure 2.1, 2.2). Near Sugar Run,

elevated methane concentrations have been reported since 2010 and several researchers and the state regulator, the Pennsylvania Department of Environmental Protection (PA

DEP), have suggested the methane is related to nearby gas wells (19, 23–25).

2.1.1 Sugar Run

The study region lies in an area near Hughesville, PA, where a high density of shale gas wells were drilled between 2008 and 2012 (Figure 2.1). In the region, a high percentage of unconventional gas wells in a 13 km x 13 km square centered around latitude 41.237783 and longitude -76.600508 have been cited for issues related to cementing and casing by PA

DEP (23). Specifically, 33.3% of the 101 spudded unconventional wells in this region have received one or more cementing- or casing-related violations. These frequencies are much higher than statewide estimates of violations (4, 26).

Sugar Run is a gaining stream -- i.e., groundwater recharges into the stream. We report new data for sites that were sampled 225 meters upstream and downstream of a location (Figure 2.1) where intermittent bubbling and upwelling seeps have been discussed in published literature since 2014 (19, 24, 25). Three of these seeps that upwell outside of the wetted stream channel during normal conditions (labeled based on their nearest stream sample location as Seeps 1.5, 1.55, 1.6), were sampled repeatedly (Figure 2.1; Figure A1,

A2; Table E6-E8). Additionally, water samples were collected from four private water wells within the dotted circle in Figure 2.1C and some measurements were made of methane concentrations in air (Appendix A, Figure A3). The water samples were compared

to samples from a natural methane seep in Salt Spring State Park in PA (see Appendix A and Table E11-E13).

Culpability for methane leakage into Sugar Run has not been established (27). One unconventional gas well (API # 081-20292) has been cited by PA DEP for contaminating five homeowner water wells with methane in the area (Figure 2.1) after drilling began on

February 12th, 2011 (see Table E1 and description in Appendix E). Within 4 km of API #

081-20292, 24 additional gas wells were drilled between 2008 and 2012, eight of which received violations related to casing and cementing. These nine wells with citations are located within 5 km of the seeps studied at Sugar Run. Two additional gas wells that are situated just outside the 5 km radius were cited by the PA DEP for methane migration into seven homeowner wells in 2011 (28).

Heilweil et al. (24, 25) and Grieve et al. (19) reported water and hydrocarbon chemistry from Sugar Run, concluding that some samples of gases observed in stream and shallow groundwater in that area are consistent with a Marcellus origin. In this paper, we report new data for inorganic solutes, hydrocarbons, isotopes, noble gases, and limited atmospheric measurements in the Sugar Run area, and discuss the new data in context with previous Sugar Run data, regional groundwater data, the local geology, and the record of shale gas development.

2.2 Methods

The geological setting of the 16.7 km2 watershed of Sugar Run (topographic slope of

10.4%) has been discussed previously along with stream measurements for campaigns in

May, June, and November 2013 (25). During those time periods, the stream was dominated by baseflow, and discharge varied from 0.05 to 7.2 m3/s. No additional discharge measurements are reported here; however, the stream conditions were generally very similar to the previous report. In that study, modelling showed that the stream returns to baseflow conditions within 1.5 days of a storm event, and that approximately 170 m3/d of methane-containing groundwater was entering the stream with a concentration of 3.2 mg/L in the segment above Site 1.5 (19, 24, 25).

For new samples reported here, the waters were analyzed for field parameters and different suites of inorganic ions, hydrocarbons (methane (C1) and in some cases ethane

13 87 86 (C2) and propane (C3)), isotopic signatures (including δ C in CH4 and Sr/ Sr), and noble gases (Appendix A). All noble gas samples discussed in this study were collected in copper tubes following a modified standard sampling protocol (29, 30) and these samples were analyzed at the Noble Gas Laboratory at University of Michigan (Appendix A).

We compared our data for Sugar Run to data available online (31), previously published data for the same area (19, 24, 25), and 41 groundwater analyses from 8 water wells sampled during the Sugar Run investigation from published reports (32, 33). These

Sugar Run data were compared to 892 analyses of pre-drill groundwater in Lycoming

County, mostly from private water wells sampled by consulting companies for the oil and gas companies and provided to us by the PA DEP (these values are published online (34)) and 75 groundwater analyses in Lycoming County sampled by the USGS (35): together these are referred to as Lycoming County groundwater data. These samples were analyzed at commercial laboratories or the US Geological Survey between 1995 and 2014 (34). No

noble gas data were reported in the compiled data set from USGS, PA DEP, or gas companies.

2.3 Results

In this section, we summarize new water chemistry observations for Sugar Run in the context of the geologic setting.

2.3.1 Geological Observations

Sugar Run is incising outcrops into bedrock of the Trimmers Rock formation

(orange dots in Figure 2.1C). The study area on the stream lies up dip from the 9 gas wells that have received integrity-related violations by the PA DEP (Figure 2.1). Both the sample sites in Sugar Run and gas well 081-20292 lie nearly on the axis of the Nittany

Anticlinorium, a large east/west-trending, convex-up fold that plunges to the east under

Sugar Run. The limbs of the anticline dip gently to the south and less gently to the north

(Figure 2.1; Appendix A, Figure A5). Given this location, well 081-20292 intersects the

Marcellus Formation at a shallower depth (1,000 meters deep) than most other Marcellus wells in state.

2.3.2 Field Observations.

Groundwater upwelling was identified by the presence of off-channel springs, orange sediments, an occasional rotten egg smell from hydrogen sulfide, or bubbling, all

of which were reported to be new by local residents after drilling (Appendix A, Figure A1).

Within 60 meters of Seep 1.55 and 5 meters from stream location SR 2, we detected methane emitting along bedding planes and joints of all orientations in outcrops of

Trimmers Rock formation (orange dots in Figure 2.2. 1C). Methane in air near the jointed outcrop nearby Seep 1.55 (>9 vol. % in air) was above the lower explosion limit (lowest concentration in air necessary for combustion) for methane gas (i.e., 5 vol. %) (36) on three occasions (Appendix A, Figure A3).

2.3.3 Dissolved Hydrocarbons.

Methane concentrations in stream samples fluctuated from 0.0003 mg/L to 0.0766 mg/L and were highest at locations SR 1.5 and SR 1.55 (Appendix A, Figure A4; Table

E6-E8) which are located near Seeps 1.5 and 1.55, respectively. Consistent with this observation, seep concentrations (0.0001 mg/L to 8.6 mg/L) were generally larger than in the stream. Methane was most concentrated in the seep that was most isolated from the stream channel (Seep 1.6): in contrast, Seeps 1.5 and 1.55 appeared to be more diluted by stream water mixing.

At their highest, methane concentrations in four local homeowner wells HO1-HO4

(2.1 to 31.5 mg/L) were higher than the maxima measured in streams and seeps. Given that hydrocarbon concentrations in groundwater are known to vary with sampling technique

(37), we emphasize the hydrocarbon analyses for our new samples that were collected using the inverted-bottle technique. This method has also been used by PA DEP and consultants hired by gas companies for collection of groundwater samples (5, 28).

Hydrocarbon analyses for samples collected with different techniques are summarized in the Table E6-E8.

Concentrations of dissolved ethane in seeps and homeowner water wells, measured intermittently in this study, ranged from 0.005 to 0.060 mg/L and 0.051 to 0.595 mg/L, respectively. The molar ratio of methane (C1) to ethane (C2) ranged from 77 to 610 for seeps and homeowner water wells where ethane was measured (Table E6-E8). Propane

(C3) was investigated in one sample from each of two water wells (HO2, HO4), and detected in one of them (HO4, 0.018 mg/L).

13 For all samples where dissolved gases were analyzed for δ C-CH4, values ranged

13 from -54.4‰ to +37.6‰. Higher δ C-CH4 values were measured in seeps with lower methane concentration (Appendix A, Figure A6).

2.3.4 Inorganic Solute Chemistry in the Context of Hydrocarbons.

No trend in concentrations of methane versus sulfate was observed for stream water, but methane was generally more concentrated in water well samples when sulfate was less concentrated. All well samples emitted the odor of H2S, and when a few of these were analyzed, dissolved H2S ranged from 0.16 mg/L to 3 mg/L (HO1, HO4). Similar to the domestic well waters, Seep 1.6 typically showed higher methane and lower sulfate concentrations. Such trends are expected based on thermodynamics alone since methane can be used as an electron donor by sulfate-reducing bacteria (38).

Unlike the domestic water wells, H2S was not detected in any seep in the Sugar Run area. In addition, the concentrations of iron (Fe), arsenic (As), and manganese (Mn) were

higher in Seep 1.6 than reported in the homeowner water wells. In contrast to these elements that tend to occur at higher concentrations in anoxic environments, solutes that are associated with oxygenated waters (e.g., nitrate and uranium) were lower in Seep 1.6 as compared to Seeps 1.5 and 1.55 (Table E6-E8). Furthermore, the concentrations of Fe,

As, and methane in Seep 1.6 vary together across seasons (Appendix A, Figure A7).

Concentrations of methane, Fe, Mn, and As in seeps 1.5 and 1.55 were also elevated and variable with time, but were generally more diluted by stream waters, especially during higher water stage (Table E6-E8). Although some of these dissolved species were elevated above US EPA drinking water limits in seeps, concentrations in stream water were never observed at levels of concern for humans or ecosystems (Table E6-E8). Lower stream concentrations are consistent with dilution with “unimpacted” waters from upstream.

Concentrations of strontium (Sr), barium (Ba), bromide (Br), and in some cases chloride (Cl) were also elevated in waters from the seeps when compared to nearby surface water (Table E6-E8). Like Fe and Mn, the highest Cl concentrations (range = 12.2 mg/L to

53.6 mg/L) were observed in Seep 1.6. Finally, the measured 87Sr/86Sr in Seep 1.5

(0.71417) was significantly higher than that in HO4 (0.71160), Seep 1.6 (0.71168) (Table

E6-E8), and in groundwater collected using a drive point piezometer near SR 1.5 (0.71141) that was reported previously (19).

2.3.5 Noble Gas Concentrations

Water and gas samples were also collected in Sugar Run and the salt spring at Salt

Springs State Park, Montrose, PA, for noble gas analysis. These samples were collected

with copper tubes using standard sampling techniques slightly modified as described in SI

(29, 30). Measured ratios of 3He/4He, R, in water and gas samples (gas was sampled only from Seep 1.55 and Salt Spring) are reported as R/Ra where Ra is the corresponding atmospheric value, i.e., Ra = 1.384 x 10-6 (Appendix A, Table E2-E4). R/Ra values for

Seep 1.55 and samples from Salt Spring park are very low (0.0114 ± 0.0005 to 0.0165 ±

0.0006). In contrast, R/Ra ratios in all water samples are slightly higher and within the range of typical crustal values (~0.02-0.05) (39). Isotopic ratios of almost all other noble gases are atmospheric within 2-sigma except for Ar ratios of some samples (Appendix A,

Table E4) (40).

2.4 Discussion

A major difficulty in identifying impacts on water quality in areas near shale gas development is distinguishing species that recently have contaminated waters from species that were present prior to development (4, 6). Such prior occurrence is common for methane and salt contaminants in the Appalachian Basin. For example, sodium (Na), chloride (Cl), bromide (Br), barium (Ba), and strontium (Sr) from deep brines and NaCl contamination from human sources are commonly observed in PA groundwaters along with naturally derived methane (e.g., 8, 11–13, 20). Part of the difficulty is the lack of adequate data documenting pre-development water quality (4). Here, we compare our new data as well as previously published data for the region near Sugar Run measured since

2010 to an estimate of background chemistry in the area that we refer to as the Lycoming

County groundwater data set. This latter dataset includes published data (35) and newly

available pre-drill data (34) from Lycoming County (Figs. 2.1, 2.2). “Pre-drill” data are water quality measurements made by commercial laboratories on samples collected by consultants hired by hydrocarbon-extraction companies prior to drilling oil or gas wells nearby (3, 18). Water quality data in Sugar Run is also compared to waters reported to have been contaminated in other areas of PA since shale gas drilling began in the state in 2004

(41). These latter sites, referred to throughout as “presumably contaminated sites,” were deemed contaminated by government agencies after the drilling of nearby gas wells

(Appendix A).

2.4.1 Methane and Ethane

First, evidence is summarized as to why Sugar Run appears to have been contaminated by recent shale gas development. Methane concentrations in Sugar Run waters (0.0004 mg/L to 0.0616 mg/L) reach levels that are significantly higher than background concentrations for streams in PA that do not align with geologic lineaments or do not have wetland inputs (2, 24, 25, 34). In addition, CH4 concentrations increased after drilling commenced in the area (Figure 2.2). For instance, methane (referred to below as C1) and ethane (C2) concentrations for groundwater from wells HO4, HO5 and

HO6 (Figure 1C within dotted circle) initially increased in 2011-2012 after drilling, and are larger than the maximum concentrations (Figure 2.2) reported for the 967 samples in the Lycoming County groundwater dataset that were collected between 1995 and 2014

(34). C1 and C2 concentrations for the three homeowner water wells have persisted well above pre-drill measurements for over seven years (Figure 2.2).In addition, several lines

of evidence are consistent with a thermogenic origin for the gas detected in Sugar Run after drilling. For example, ethane, often detectable in thermogenic gas but only rarely detected in biogenic gas (42), was observed to increase in water wells HO4, HO5, and

HO6 after gas-well drilling in the nearby region (Figure 2.2). Decreases in C1/C2 ratios for HO4, HO5, and HO6 accompanied this increase in ethane (Table E6-E8). Such decreases have been used to argue for gas from a more thermogenic source in some systems (41). Ethane was also detected in water wells HO2 and HO3 and in Seeps 1.5 and 1.6. (We did not try to measure ethane in stream water). In contrast, for the compilation of groundwater data from Lycoming County, only 10 of the 897 samples where ethane was analyzed showed detectable ethane (Figure 2.2). Differing detection limits for ethane could also play some part in this latter discrepancy (i.e. some

13 laboratories had very high detection limits). δ C-CH4 in all samples near Sugar Run where isotopes were measured ranged from -54.4‰ to 37.6‰. Plots such as Figure S6

(Appendix A) (9) indicate that one of the sources of the methane is a thermally mature thermogenic source that is influenced by seasonal mixing with biogenic methane.

13 Enriched δ C-CH4 values for stream samples have been previously attributed to methane oxidation at this site (19, 24, 25). Indeed, Rayleigh fractionation of methane during

13 oxidation is a reasonable explanation for the very high δ C-CH4 values because biologic processes preferentially oxidize the lighter isotope (Appendix A, Figure A6). The δ13C-

CH4 values of the samples with largest methane concentrations (measured in Seep 1.6) plotted on Figure S6 (Appendix A) are assumed to approach that of the original unaltered

13 endmember: -28.3‰. This value of δ C-CH4, as well as values measured by Isotech

Laboratories, Inc., for nearby homeowner wells HO2 (-29.81‰) and HO4 (-27.53‰), are similar to reported values for well API # 081-20292, i.e., -28‰ to -29.5‰ (33). (We have not found isotopic values for other gas wells in the area). Possible thermogenic sources consistent with these values include the Marcellus or Upper Devonian formations (43).

2.4.2 Noble Gases

Noble gases dissolved in groundwater derive from the atmosphere, crust, and mantle (40) and relative contributions from these sources can be calculated based on a few assumptions

(Appendix A). For example, R/Ra ratios for He isotopes in our samples are much lower than the typical Mid-Ocean Ridge Basalt (MORB) mantle value of ~8 (44), negating the presence of mantle components. A comparison of R/Ra and He/Ne (Appendix A, Figure A8) are also consistent with the absence of mantle He. These data thus differ from a previous study that reported minor mantle He for shallow groundwater in the Marcellus Shale footprint (15).

Likewise, crustal 40Ar (noted here as 40Ar*) was detected in waters collected from homeowner well HO4 and in the gas from Seep 1.55 (Table E4). In addition, calculated

4He/40Ar* ratios of samples from HO4 range from 7.49 to 9.01, within the previously reported range (6.2-13.7) for samples of natural gas from the Marcellus Formation (45), and over an order of magnitude lower than similar values from the shallower Upper

Devonian Canadaway Formation (214.6-285.4) (45). These observations are consistent with the Marcellus Formation (and not Upper Devonian formations) as the source of both crustal noble gases and thermogenic methane.

Noble gases also yield insight about the mechanism of transport. If Marcellus gas

4 20 36 migrates as solute in upwelling groundwater, He/CH4 and Ne/ Ar would fractionate and become altered (15, 16, 18) upon reaching the aquifer (Appendix A). However, these two ratios in HO2 and HO4 are similar to that of Marcellus gas (Appendix A, Figure A9), an observation consistent with advective migration of methane in a free gas phase.

Solubility and mass balance arguments previously reported for methane and chloride also lead to the conclusion that methane is moving upward as a free gas phase into Sugar Run

(19).

2.4.3 Brine Salts and Migration Pathways

Although we argue that the new influx of methane into Sugar Run occurs as free gas phase migration, gases might also be dissolved in waters that are moving upwards with salts from Appalachian Basin brines (ABB) because traces of salts have been detected in Sugar Run that are consistent with these deep brines. For example, on a plot of concentration ratios for Cl/Br versus concentration of Cl (Appendix A, Figure A10), waters from the stream, seeps, and homeowner wells mostly lie in the sub-field representative of diluted ABB. In addition, 87Sr/86Sr values measured for groundwater near SR 1.5 (19), water well HO4, and Seep 1.6 (Table E6-E8) are consistent with the isotopic signature published for Middle Devonian formations such as the Marcellus, with

87Sr/86Sr = 0.71000-0.71212 (10). In contrast, samples collected from the stream (19) and

Seep 1.5 (Table E6-E8) yield 87Sr/86Sr = 0.71342-0.71417, a signature more consistent with brines from formations above the Marcellus (Appendix A, Figure A11).

It is not uncommon to observe both brine salts (46) and methane in uncontaminated waters in PA because thermogenic methane dissolved in such slightly saline waters moves naturally into aquifers in parts of the state (11). Thus, in most uncontaminated groundwaters in Lycoming as well as elsewhere in the Appalachian

Basin (West Virginia, New York), higher methane concentrations generally are observed in the presence of higher Cl concentrations (Appendix A, Figure A12). In comparison, most of the methane-containing groundwater in water wells in the Sugar Run valley are low in Cl (all <14.4 mg/L). We infer that the relatively high-methane and high-Cl samples from Lycoming (PA), WV, and NY contain meteoric water mixed with naturally upflowing groundwaters that contain dissolved methane and brine salts from natural sources (13, 14), while the high-methane, lower-Cl groundwaters entering Sugar Run are affected by natural gas from free-phase upflow (Appendix A, Figure A12).

2.4.4 Structural Characteristics

In this section we explore whether Sugar Run may be particularly susceptible to gas migration because of fracture development.

Vertical fractures (joints) in this area have been observed in several orientations

(47, 48), including regionally extensive NW- and NNW-striking joint sets as well as local joint sets that strike parallel or perpendicular to the axis of the anticline. If the local joint sets form during folding, the intensity of this jointing usually correlates with fold curvature, i.e., higher along the axis of the fold (49–52). The distribution of fracture intensity can be estimated using curvature analysis in such cases. Curvature is usually

greatest along axes of anticlines and synclines (Figure 2.3), although the exact relationship between jointing and curvature is difficult to predict (50).

Joints may enable methane migration in this area because the Marcellus is very shallow (Figure 2.1). The Marcellus Formation is closer to the surface under the seeps

(0.6 km) than it is under any of the eight gas well pads located within 5 km to the east.

The depth to Marcellus is greater east from Sugar Run because the axis of the large fold comprising the Nittany Anticlinorium directly underlies Sugar Run (Figure 2.1C;

Appendix A, Figure A13) and because the Anticlinorium plunges to the east. In fact, the

Marcellus comes to the surface about 3400 m to the west of the study area along the anticline axis (shown as white area in Figure 3).

In addition to joints caused by folding in this area, joints also likely formed during unroofing. Such joints only form within 0.5 km of the surface as a rock unit is exhumed

(53). This depth is much shallower than most of the Marcellus gas wells in Pennsylvania but similar to the 0.6 km depth of the Marcellus under the seep locations. Therefore, unroofing joints might also enable migration of methane gas from the Marcellus to the seeps. These joints could help methane migrate naturally or facilitate migration of anthropogenically-sourced methane in the event of gas-well leakage (54).

In addition to vertical migration, gas could be migrating updip along bedding planes and staircasing upward through bedding planes and joints. Updip gas migration has been shown in PA to correlate with gas pressures above the saturation point, i.e., transport as a free gas phase (16). Of the units overlying the Marcellus, the Mahantango

Formation is probably the most likely to accommodate layer-parallel gas migration, as a

hydrogeologic study of the region has shown that it is more hydrologically productive than overlying units (55). The Mahantango Formation (Middle Devonian) lies 200 meters below Seep 1.5 (47, 55). Such a hypothetical path would move gas from the gas wells updip and along the axis of the anticline toward the seeps. Consistent with this, water wells in the study area with methane concentrations >0.11 mg/L are mostly west

(updip) or north-west of gas wells (e.g., API# 081-20292), while water wells to the south and east (down dip) are much lower in methane concentration (32).

Recently, other anticlines in PA have also been shown to be associated with methane-containing groundwaters. Specifically, inspection of groundwater data has revealed that CH4 concentrations increase slightly near the Towanda anticline to the northeast of Lycoming County in Bradford County PA (20). Several cementing/casing related violations were also issued to shale gas wells by the PA DEP along that anticline

(23). That anticline is also associated with several large faults (20).

2.4.5 Methane Impacts on Groundwater

The evidence summarized so far is consistent with methane migration beneath

Sugar Run since 2011. The most direct groundwater impacts are observed in HO4, HO5, and HO6. Here we focus on long term groundwater impacts.

After methane concentrations increased in these three water wells, Fe concentrations increased and then decreased (Figure 2.4). These observations are similar in some respects to observations published recently for a subsurface methane plume caused by a blowout at a gas well (56). Those authors argued that micro-organisms

catalyzed anaerobic oxidation of methane coupled with reduction of ferric oxides to produce soluble Fe(II) along the leading edge of the plume. A decrease in aqueous Fe, observed after the methane plume moved through, was attributed to depletion of solid- phase Fe oxide minerals. Other reducible oxides such as Mn were similarly solubilized for a transient period. Transient spikes in Fe concentrations have also been observed in other water wells presumably affected by methane release from oil or gas activity (41,

57–60). Given the similarity between our observations and those reported in other research, we attribute the spikes in Fe concentrations (Figure 2.4; Appendix A, Figure

A14) after initial increases in methane and ethane for HO4, HO5, and HO6 to reduction and mobilization of the metal because of anaerobic methane oxidation. Groundwater samples collected by PA DEP and consultants are generally acidified but not filtered, meaning the analysis includes dissolved and some suspended particulate iron if it is present (46). In contrast, all of our samples analyzed for iron were filtered before acidification (see Table E6-E8). Therefore, for comparison with the other data, the sample collected on 7/26/2017 and plotted on Figure 4 was not filtered before analysis.

The Fe concentration in that sample, reported in Table E6-E8, was within a factor of 2

(0.02 mg/L) of its filtered counterpart sampled at the same time (0.01 mg/L).

Based on our interpretation of Figure S14 (Appendix A) and observations from the literature (38, 59), we might also expect to see methane oxidation coupled to sulfate reduction to sulfide. Indeed, high natural concentrations of methane are often observed with low sulfate in water supplies across the United States (e.g., 20, 59). Consistent with this, H2S was smelled or detected at HO1, HO2, HO3, and HO4 (Table E6-E8). One

reason for the observed drop in Fe concentrations (Figure 2.4) might therefore also be that after the onset of sulfate reduction, Fe precipitated as one of several highly insoluble iron sulfide phases such as pyrite (38).

2.4.6 Distinguishing New Methane from Pre-existing Methane

These observations suggest that onset of new methane contamination can sometimes be identified by a transient period of higher iron and higher sulfate in groundwater. To test this, Figure 2.5 shows plots of water quality data from presumably uncontaminated, naturally equilibrated groundwaters and presumably recently- contaminated, non-equilibrated groundwaters. Specifically, plots of sulfate versus methane (Figure 2.5A) and iron versus methane (Figure 2.5B) are shown for waters from i) our study area; ii) the Lycoming County groundwater data set, and iii) published data from four separate presumably contaminated sites in northeastern PA (41). For the latter sites, locations were inferred using maps in the report. The methane-containing waters in the Lycoming County groundwater dataset were assumed to have long received influxes of naturally derived methane. The waters from incidents in northeastern PA were presumed to be contaminated by shale gas development as reviewed by the US

Environmental Protection Agency (41). These sites from incidents are shown on Figures

5 and S12 and labelled as follows: i) six wells along Paradise Road (labelled as GW 13,

GW 18, GW 19, GW 20, GW 37, GW 38 following the EPA report) in Bradford County,

PA (labelled Paradise in Figure 2.5 and Appendix A, Figure A12); ii) one well (GW 23 in

EPA report) near Dimock, PA (labelled Dimock in Figures 2.5 and Appendix A, Figure

A12); iii) two wells near Granville Road and near the axis of the Towanda anticline in

Bradford County, PA (labelled GW 01, GW 02); iv) and one well near Marshview Road in Bradford County, PA (labelled GW 06).

On Figure 2.5A, waters with the highest sulfate but low methane concentrations and the waters with the highest methane but low sulfate concentrations in the Lycoming

County groundwater dataset were used to plot lines to delineate what we infer to be natural, unperturbed waters. These data are assumed to represent groundwaters that are close to thermodynamic equilibrium because the methane concentrations in those samples are interpreted as long-standing, background concentrations. A similar procedure was followed for Figure 2.5B to indicate highest iron at low methane concentrations and highest methane at low iron concentrations in the Lycoming County groundwater dataset.

This approach identified data in the upper right quadrants of Figure 2.5A and B as possible indicators of transient sulfate and Fe concentrations, respectively.

Strikingly, Figure 2.5A shows that concentrations in Sugar Run water wells HO4 and HO5 shortly after the onset of drilling plot in the upper right quadrant along with data from the presumably contaminated water wells from other PA incidents (41). These high methane and high sulfate samples stand out against pre-drill groundwater from the rest of

Lycoming County. For iron, samples from water wells HO4 and HO6 also plot in the upper right quadrant of Figure 2.5B along with several of the presumably contaminated water wells. High sulfate (> 6 mg/L) and iron (> 0.3 mg/L) in waters with high methane concentrations may therefore be good indicators of recent contamination (Figure 2.5).

Figures 2.5A, 2.5B, and A14 (Appendix A) also give indications of the duration of the inferred transient spike in sulfate and iron. For example, concentrations in HO4 of sulfate (pre-drill concentration of 16.6 mg/L) sampled by PA DEP on 6/14/2016 (8.21 mg/L) decreased to 5 mg/L (this study) on 7/26/2017 while elevated methane persisted

(Figure 2.5A inverted triangles). If our interpretation is correct, the transient sulfate period lasted at least 7 months after the onset of methane migration (Appendix A, Figure

A14). Likewise, Fe (concentration not detected in pre-drill data) collected from HO4 and analyzed by PA DEP on 6/14/2016 (0.154 mg/L) and 7/26/2017 (0.01 mg/L – this study) illustrates that the Fe transient may also last at least 7 months when elevated methane persists (Appendix A, Figure A14). Based on our interpretation of Figures 2.5 and A12

(Appendix A), we suggest a possible protocol for quickly assessing groundwaters that may have been impacted by recent methane migration. If methane concentrations are greater than 10 mg/L and contain concentrations of Fe > 0.3 mg/L and sulfate > 6 mg/L, further investigations are warranted. Such further work would be especially warranted if chloride < 30 mg/L and Ca/Na > 0.52 (mass ratio), because the waters would not look like natural brine salt-affected waters (Appendix A, Figure A12). Research is needed to test this protocol more broadly in the northeastern U.S.A and elsewhere. Of course, geochemical characteristics that are consistent with one or more of these tests do not confirm that the water has been contaminated, and other measurements – for example isotopic studies – would be needed. Nevertheless, the protocol can identify sites where further testing or monitoring should be conducted.

In some waters, no sulfate is present in the aquifer before or after contamination, and such waters would plot as false negatives on Figure 2.5A. In addition, if the duration of time since the onset of gas leaking is large enough, the new methane might exhaust the sulfate or Fe, providing another reason why some contaminated samples might plot as false negatives.

2.5 Conclusions

These observations of air and water chemistry at Sugar Run are best explained by a methane gas plume moving from depth into the aquifer over the last 7 years. During this period, one seep and one homeowner well were measured to contain gas and strontium that are isotopically like Marcellus fluids. At the upper part of the plume is a seasonal zone of oxidation of methane coupled to oxygen reduction (Appendix A). At depth, methane oxidation is coupled to metal reduction or sulfate reduction (Figure 2.6).

The electron acceptors are likely used up in sequence from oxygen to metals to sulfate.

Once the oxidants are depleted, methane, ethane, and other hydrocarbons pass through the system with less oxidation, allowing their concentrations to persist or increase (Figure

2.2). The rate of hydrocarbon plume migration in the subsurface thus is affected by the availability of electron acceptors in the aquifer. With ongoing methane influx to the aquifer, some deleterious contaminants such as arsenic can be mobilized.

Although all water quality data are not released to the public in the Appalachian

Basin, the rate of incidence of problems such as described in this paper appears to be relatively low compared to the number of shale gas wells that have been drilled (4, 5, 20).

Nonetheless, such presumably rare incidents are important to study for at least two reasons. First, problems that are understood can lead to better decisions. For example, drilling into a shale formation at shallow depth along the axis of a large fold, such as described in this study area, may produce wells that intersect fractures that interconnect to form good pathways for upward migrating contaminants. In addition, future research is needed to determine why some anticlines are associated with higher methane concentrations (Nittany, Towanda) while others apparently are not (Rome and Wilmot anticlines in Bradford County) (20). Such research could provide maps of areas where drilling should be precluded if methane migration is to be eliminated entirely or could lead to better management practices for drilling such areas.

Second, migration of methane into aquifers can present explosion hazards – a well known phenomenon – but can also change the redox state of the aquifer – a phenomenon that has not been as well documented. Such changes can feed consortia of bacteria that mobilize species that can degrade the aquifer, including the possibility of mobilization of arsenic. We have shown that the presence or absence of redox-active species in water samples with high methane concentration can be used to document recent, rather than long-duration natural, contamination by methane. Specifically, the observation of high sulfate – high methane and/or high iron – high methane in groundwater may indicate a new source of methane has entered a groundwater system. Multiple lines of evidence are nonetheless necessary to make firm conclusions for any given site.

2.6 Figures

Figure 2.1. (A) Study area showing Lycoming County, Pennsylvania. (B) Expanded view showing all active unconventional gas wells that have (red) or have not (green) received well-integrity violations from the state regulator (21), PA DEP (also shown on (C)). (C) Expanded view of Sugar Run watershed showing sample locations in streams (triangles) and bubbling seeps (arrows). Outcrop locations where methane was detected during this study in the air near fractures are depicted as orange circles. Cementation was not observed in any fractures. Average dissolved methane concentrations in stream sites are shown as grey triangles (intensity of grey is contoured with respect to concentration as shown in legend and Appendix A, Figure S4). (D) Cross section S-S’, defined on (C), roughly follows the plunge of the Nittany Anticlinorium to the east. On this cross section, well API # 081-20292 intersects the Marcellus Formation at 997 meters and the depth of 0 represents sea level. The well pad where well 081-20292 is drilled contains only that gas well. A lateral from that well follows the Marcellus roughly perpendicular to the cross section and is shown in (C) as a dashed line. Where new data are presented herein for seep, homeowner, and air samples, they were sampled within the dotted circle in Figure 1C.

Figure 2.2. (A) Time series plots of dissolved methane concentrations from homeowner (HO) water wells 1 – 10 sampled in the study region near Sugar Run in Lycoming County. Plotted data for water well HO8 includes two water wells (A & B) sampled on the same property. Vertical solid lines indicate spud dates for unconventional gas wells cited for cementing/casing violations within a 5 km radius of Seep 1.5. In addition, the dashed and dotted lines indicate the spud date and hydraulic fracturing date for well API # 081-20109, respectively. Samples indicated by open symbols are not discussed further and were reported by the gas company as part of the initial (32, 33) or subsequent investigations (31). Concentrations plotted after 10/26/2016 were sampled and measured by our team (Table E6-E8). B) A time series plot of dissolved ethane concentrations for HO6, HO5, HO4, HO3, and HO2 for data reported online (33), by PA DEP (31), or in this study. Horizontal dashed lines represent median and maximum concentrations for dissolved methane (n = 967) or ethane (n = 897) reported for Lycoming county samples outside the study region between 1995 and 2014. These samples were analyzed from published reports (35), or collected as “pre-drill” data by companies before drilling gas wells and were released to PA DEP and shared with Penn State University (34). For ethane, only 10 of 897 samples contained detectable ethane: line labelled “Lycoming Maximum (uncensored)” on (B) summarizes the highest pre-drill concentration for samples with detected ethane. Ethane concentrations were not reported for samples collected and measured for ethane indicated by arrows labeled “no ethane detected in HO5 & HO6” on (B). (Data were censored but no reporting limits were indicated). Ethane was not analyzed in HO4 before drilling commenced. The maximum and median values for the Lycoming data plotted in (A) and (B) are for samples collected outside the dotted circle in Figure 1C.

Figure 2.3. Color contours showing the mean curvature, as defined by published algorithms (62), calculated for the top of the Marcellus formation (smoothed using a 2 km moving average window to minimize artifacts from modeling). Warmer colors indicate where the mean curvature is greatest and likely to have caused the greatest density of vertical fractures that may allow upflow of gas from depth. Curvature was calculated using Move’s surface geometry analysis tool. Labels show the locations of Sugar Run and the API # 081-20292 well. The Marcellus outcrops 3.3 km from the Sugar Run location point, i.e. where the white area cuts into the Sugar Run valley as shown in the figure.

Figure 2.4. Iron concentrations in groundwaters from HO4, HO5, and HO6 plotted versus time. Concentrations were normalized to the maximum values (0.58 mg/L, 0.14 mg/L and 3.02 respectively) (33). The HO4 sample collected on 7/18/2012 was slightly offset to avoid overlap. Downward arrows represent the reporting limit for censored analyses. For HO5, no reporting limit was indicated, so it is shown as an arrow at the concentration equivalent to the smallest reported concentration in that report. Data were derived from reports (31–33), with one HO4 sample collected on 7/26/2017 from this study (see main text).

Figure 2.5. Graphs of groundwater chemistry from homeowner wells sampled in this study (symbols labeled HO), the Lycoming County groundwater dataset (Lyco GW, empty black circles), and presumably contaminated sites in neighboring counties (orange symbols), as described in text. A) Concentrations of sulfate plotted versus methane show that most waters are either high in sulfate or methane but not both, as expected based on thermodynamic equilibrium. Waters from the Sugar Run area that are presumed to be experiencing a new methane influx (HO5 and HO4) plot in the upper right quadrant along with water from wells from the presumably contaminated sites. After several months, sulfate is inferred to be reduced and waters no longer plot in the upper right quadrant (e.g., compare HO4 in 2017 to 2016; empty red inverted triangles). B) Concentrations of iron (Fe) plotted versus methane show that generally, Fe concentrations are low with elevated methane concentrations. Some wells from the Sugar Run study area and presumably contaminated sites contain high methane and iron concentrations. Lyco Gw refers to data provided by the PA DEP, and from published reports (35) as described in text. The PA DEP data are also published online (34). Data from homeowner wells were either sampled in this study (Table E6-E8) or are published (31–33).

Figure 2.6. Generalized schematic of an evolving methane plume released from a shale gas well. The inserted plot shows how water chemistry changes over time. The plot is patterned from data from homeowner well HO4 (Table E6-E8). Two separate “plumes” are drawn for early and later time periods. The outer edge of the plume (dark gray) represents the zone of methane oxidation coupled to metal reduction. The inner section of the plume (light gray) represents the zone of methane oxidation coupled to sulfate reduction (creation of H2S). Once electron acceptors are consumed, hydrocarbons flow unhindered from the subsurface to the surface resulting in even higher concentrations and unaltered isotopic values, if a leak continues. This diagram is not drawn to scale or to reflect the geology of Sugar Run.

Chapter 3

Stream Methane, Citizen Science and Metal Rich Springs

3.1 Introduction

The rapid development of shale gas in the past decade has increased public concern about water quality and emission of greenhouse gases such as methane (CH4).

Groundwater contamination is of particular concern since methane is the most commonly cited water quality impact related to shale gas in Pennsylvania (PA) (4). Methane in groundwater is an explosion hazard when it degasses and accumulates at high concentrations (50,000 ppm in the air) (21) and has been observed to cause mobilization of metals into natural waters if methane was introduced recently (1). If methane escapes to the air, the warming potential as a greenhouse gas is 84 times that of carbon dioxide

(CO2) on a 20-year timescale (63).

Despite recent concerns surrounding shale gas development in Pennsylvania, this

“new” development (i.e. about 12,000 wells as of the end of 2017) represents only a fraction of the state’s known oil and gas infrastructure because of the large number of conventional oil/gas wells (n > 140,000 active, plugged, or regulatory inactive conventional oil and gas wells) (Figure 3.1) (http://www.depgis.state.pa.us/pa oilandgasmapping/). These wells date to 1859 with the drilling of the first commercial oil well in the world. Drilling, casing, cementing and plugging standards have changed drastically over the last two centuries (64, 65). For example, legislation passed in 1921 required wells drilled through coal seems to be plugged with “well-seasoned, round

wooden plugs” (66) and some research has suggested that wells plugged before 1957 are likely improperly plugged or abandoned by today’s standards (65). The legacy of oil and gas drilling has led to the estimated abandonment of 470,000 – 750,000 oil and gas wells with unknown locations (67), most of which are located in western Pennsylvania (65).

Records for only ~11,000 of these wells exist in the Department of Environmental

Protection (DEP’s) online database of oil and gas wells in the state (i.e. ~150,000 known conventional oil and gas wells) (http://www.depgis.state.pa.us/paoilandgasmapping/).

Improperly abandoned wells pose multiple environmental concerns. Most notably, they can leak hydrocarbons to the atmosphere years after abandonment (67–69). Methane leakage is also a concern with respect to wells from ongoing oil and gas development.

For example, some researchers have estimated that ~11% of total production for conventional oil and gas wells leaks to the atmosphere, due in part to old and aging infrastructure (70). This fraction is much larger than the estimates for unconventional wells (range: 0.36 – 0.86 % total gas production) (70, 71). However, unconventional wells typically produce much more total gas than their conventional counterparts.

In addition to oil and gas development, coal mining has a long history in PA, dating to the 1700’s. Similar to oil and gas development, centuries of mining have led to the subsequent abandonment of many coal mines, over 5,000 of which are reported by the

State regulator (www. depgis.state.pa.us/emappa). During and after mining, the disturbance of the subsurface and interaction of oxygen-rich waters with sulfide-bearing minerals such as pyrite (72) commonly creates contaminated discharge known as abandoned mine drainage (AMD). AMD-impacted surface or groundwater typically is

characterized by low pH (although some mine drainage is alkaline), high conductivity, and high concentrations of heavy metals and sulfate (73). AMD poses serious environmental concerns for surface and groundwater and can impact biota (73).

Coal, and in particular coal mines, are also a large source of methane. Coal mines are estimated to produce approximately 9% of all man-made emissions in the United

States (74). In PA, the state with 1/3 of the abandoned mine-related problems in the U.S.

(75), coal mining may be a much larger contributor to statewide emissions (71). Prior to mining, methane is stored in both micro- and macro-pores (fractures also known as cleats) in the coal, either as a free gas or adsorbed to the coal surface (76). A change in pressure (i.e the physical removal of overburden from coal mining) can cause methane to quickly be released from cleats into mine void space and or the atmosphere (76). This methane can be thermogenic or produced by microbial communities as observed in the

Illinois Basin (77). After abandonment, mines can produce methane for decades to centuries (76). However, methane emissions from abandoned mines are typically the largest directly after abandonment, slowing quickly in the first decade, and then decreasing more slowly over time (76). Emission estimates for abandoned mines typically take into consideration many parameters such as the year of mine closure or abandonment, emission rate at closure, and status of the mine (i.e. venting, flooded, sealed or unknown) (76). The status of the mine is especially important to emissions estimates since, for example, most estimates assume that flooded coal mines do not produce emissions (74, 76, 78).

Figure 3.1: Map illustrating the known spatial distribution of conventional and unconventional oil and gas wells and coal mines in Pennsylvania.

In addition to anthropogenic sources of methane, additional sources of methane occur naturally. For example, micro-organisms can produce methane under reducing conditions at ambient surface conditions in a process known as methanogenesis (9).

Production of thermogenic methane occurs at greater depth during the breakdown of kerogen. After its formation, methane can migrate to the surface along faults, fractures, and other natural pathways in a gas phase or dissolved in water (16). If methane migration co-occurs with deep groundwater in PA, this water is commonly salty because it can contain original formation (connate) waters (brine) from deeper formations (10–

14). Some studies have observed correlations between methane concentration and

chloride concentration in groundwater in the Pennsylvania/New York parts of the

Appalachian Basin (13, 14). Numerous studies have attempted to characterize these brines (10, 11), denoted here as Appalachian Basin Brine (ABB).

Understanding the effect and extent of methane leakage from any source, particularly in the subsurface, requires extensive measurements. Monitoring near oil and gas development in PA has typically involved sampling groundwater from local private water wells; indeed, most gas companies now sample waters within 2,500 feet of planned development before drilling (7). However, groundwater monitoring can be expensive, time intensive, and samples difficult to obtain. Additionally, there is documentation of methane migration from shale gas wells far beyond distances of 2,500 feet (1, 17, 43).

Furthermore, even when large geographically extensive datasets are available (20) they generally lack time series data for sites of interest.

The recent introduction of a new form of methane monitoring (79) involves sampling methane in stream water with the help of local volunteers (2). Stream methane is a useful indicator of subsurface contamination since many streams in Pennsylvania gain discharge from groundwater (i.e. gaining streams). This groundwater can then assimilate potential contaminants such as methane and associated pollutants along flow paths (24). When compared to groundwater monitoring, stream sampling allows relatively rapid, low-cost sampling over large geographic areas. Additionally, stream sampling can be extremely efficient as no homeowner permission is needed (on public land). This approach is summarized extensively in previous work (2) and is made possible by various partnerships with volunteer groups and organizations (Table 1).

Working with volunteer groups and local organizations lowers the sampling time for scientists, increases the spatial density of collected samples, and also provides valuable information for site selections, access, and possible contaminant sources.

In a global compilation, measured methane concentrations have been observed in streams or rivers to be as high as 6190 µg/L (80). Very high values are often observed in tropical regions with extensive wetlands (81). However, methane concentrations in headwater streams and cooler regions are typically lower (0.8 – 70 µg/L) (2, 80, 82, 83).

In Pennsylvania, methane concentrations in stream water are typically very low (i.e. at

131 sites the median was ~1 µg/L) (2). In that Pennsylvania compilation, stream methane concentrations varied significantly depending on geographic location, but concentrations were almost always higher than CH4 in equilibrium with the atmosphere (0.08 µg/L) (2).

For stream waters in PA, a value of 4 µg/L or higher (2) in non-wetland sites was proposed in that work as a threshold signal that the gas source may be anthropogenic in origin. However, those authors argued that further investigation was still necessary to determine if methane was from anthropogenic or natural sources. Further investigation to determine the origin of methane could include the analysis of isotopes in the methane

13 2 13 (δ C-CH4 and δ H-CH4). For example, δ C-CH4 < -60‰ has often been used to

13 2 distinguish biogenic gas from thermogenic gas (δ C-CH4 > -50‰) (84). δ H-CH4 was not measured in this study or in previous work on PA streams (2).

Stream sampling does have drawbacks (as mentioned in previous work (2)).

These drawbacks include: general dilution concerns (i.e. large streams are not great candidates for sampling because the greater discharge of water in these streams can cause

excess dilution), seasonal and weather-related concerns with respect to variability of flows, inability to sample on private land, methane degassing concerns, time sensitivity of collected samples, and safety concerns for samplers.

In this study, we worked with Trout Unlimited as well as other organizations such as the US Forest Service and Susquehanna River Basin Commission (SRBC) to monitor stream methane in areas near shale gas development, legacy oil and gas development, and legacy coal mining. The goals of this study were to learn which, if any, of these sources were leaking methane into stream water, note any new observations associated with that leakage, and help local groups learn and understand about hydrocarbon contamination in their areas. For example, Trout Unlimited (TU) was concerned about the health of wild trout populations so most sample prioritization for sampling with TU took place in streams with wild trout populations. Broad sample locations were prioritized around the three major hydrocarbon extraction techniques mentioned above: unconventional gas development, conventional oil/gas development and other kinds of development including coal mining. We placed emphasis on stream sites containing methane greater than the threshold proposed by Wendt et al. (2), >4 µg/L. Namely, for some of the more methane-rich sites we completed a small number of further investigations. Sites were especially highlighted for further investigation if methane concentrations were elevated >

7 µg/L, the site was accessible, and if oil, gas or coal contamination was suspected.

Additionally, during sampling, special attention was placed on locating and sampling orange colored discharges or discharges that had an odor (rotten egg) (1). In addition, special attention was placed on locating and sampling abandoned wells (i.e. metal casings

discharging fluids and hydrocarbons, see Figure B-7 in Appendix B) because of their known association with leaking hydrocarbons (68).

In the summary that follows, we focus on investigating the following hypotheses for samples in the Appalachian Basin, with a focus on Pennsylvania: 1) older oil and gas wells (i.e. wells drilled before 1955) leak methane at a higher frequency than younger wells; 2) coal mines do not leak methane when they are flooded; 3) discharges of leaking hydrocarbons from oil and gas wells create a new type of discharge — gas leak drainage — that is chemically distinct from AMD and can sometimes be highly concentrated in ABB salts.

3.2 Methods

3.2.1 Sampling Design

The sampling and analytical design in this study follows some aspects from previous work (2). In the sampling design implemented here, we focused on areas of intense development of shale gas, conventional oil and gas, or coal mining to help local groups understand contamination in their areas and make observations about any detected leakage. Selection of sites within the areas of interest was based on such information as local knowledge and concerns of local organizations and volunteer groups (i.e. The US

Forest Service and Trout Unlimited), site accessibility, location of possible hydrocarbon contaminants, or location of wild trout populations. In some cases, “pristine” watersheds were targeted to collect background data as a control (i.e. Monongahela snapshot day;

Table E17). In this case, “background” refers to areas with as little human impact as

possible (85) such as undeveloped forests. In addition, because methane is inherently low in most PA streams (2), sites were typically selected in headwater streams where flows were lower so that methane would not be diluted to great extent by runoff.

Targeted sampling was also conducted at three types of sites: AMD discharges, colored or H2S-containing seeps, and leaking oil or gas wells that are now abandoned or orphaned. These latter sites were identified where metal casings emerged from the ground with upward flowing water (Figure B-7). This sampling targeted at the three types of sites was designed to: 1) understand if AMD discharges could be a source of methane to the atmosphere; 2) understand what was causing metal-rich and sulfur-rich seepages in areas without known coal mining; and 3) understand the composition of fluids discharging from abandoned wells and how this could affect streams. These sampling locations (Figure B-8) were determined through local knowledge (conversation with local volunteers or organizations) and using online tools and publications (www. depgis.state.pa.us/emappa (86)).

3.2.2 Sample collection

Stream methane sampling is described in depth in previous works (1, 2, 19).

When possible, streams were measured mid-stream and approached from a downstream location. Bottles were rinsed three times, filled and capped underwater making sure that no bubbles were present (when possible) after collection. Biocide (potassium hydroxide,

KOH) was not used by volunteers for health and safety reasons and was not used by scientists to maintain consistency. Samples were returned as quickly as possible to the laboratory for analysis (always analyzed within 5 days of collection).

Following site selection, sampling occurred in one of three ways. In the first method, sample bottles were mailed via Fedex overnight shipping to volunteers who mailed samples back as soon as possible for analysis within 5 days of collection at Penn

State. Collection of samples at these sites took place once or multiple times in the event of elevated methane.

In a second method for sampling, samples were collected over an entire watershed or region by large groups of volunteers (20-40 volunteers) in events organized by Trout

Unlimited (Jacob Lemon) and called “snapshot days.” Five snapshot days were conducted in three locations during this study (Pine Creek watershed, PA (n=1);

Monongahela National Forest, WV (n=1); and the Allegheny National Forest, PA (n=3)).

Site selection for snapshot days followed the outline described above with an emphasis on site accessibility for volunteers.

Finally, the third method of sampling, usually for cases where we had previously measured elevated methane concentrations, sites were re-visited by Penn State scientists

(ideally accompanied by local volunteers) for further investigation (see section 3.2.3) and additional sampling as described below. Not all sites with elevated methane concentrations could be re-visited so priority was given to sites with methane > 7 µg/L and if hydrocarbon leakage was suspected.

During follow-up trips, researchers re-sampled stream water for hydrocarbons

(methane and in rare cases ethane which could be analyzed from the same sample bottle) and hydrocarbon isotopes and performed small scale observational site investigations. If metal springs or leaking abandoned wells were discovered, additional samples and

measurements were collected from these discharges to enable analysis of dissolved cations and anions and to make field measurements. When waters were collected for inorganic cation and anion analyses, samples were filtered using a pre-rinsed syringe- based 0.45 micron filter (changed at each sample location). Samples for cation analysis were acidified using trace metal grade pure nitric acid. Samples were stored in a cooler after sampling and refrigerated until analysis. When possible, field parameters were measured (pH, oxidation reduction potential (ORP), temperature, specific conductivity

(SPC), and dissolved oxygen (DO) using a YSI Professional Plus meter, calibrated before each trip as previously described (1). The description of how measurements were made are described previously (1).

Local knowledge and expertise from volunteers proved crucial during many site investigations and follow up trips. In many cases, locations of leaking abandoned wells or metal discharges were provided to scientist by volunteers or agencies. These discharges were sampled without prior knowledge of locations of elevated stream methane nearby.

For example, 21 springs were sampled and discovered to be hydrocarbon-rich. In addition, one volunteer’s discovery that methane-rich waters were emitting from abandoned coal mine discharge led to the targeted sampling and methane analysis for 14 separate coal mine discharges in PA.

3.2.4 Laboratory methods

Hydrocarbon analysis, hydrocarbon isotopic analysis, and cation and anion analysis were all described previously (1). ICP-MS analysis was performed on some samples for chromium (Cr), arsenic (As) and uranium (U).

3.3 Results

3.3.1 Stream methane

In this study, we present analyses from 358 new sites that were sampled in

Pennsylvania and West Virginia. All stream and discharge data are available in Appendix

E (Tables E8-E18).

These site analyses were compiled with data for 140 sites from PA and NY from previous work (2, 19), 11 sites from PA (24, 25), 20 sites from the Susquehanna River

Basin Commission (SRBC), for a total database of 1,081 samples at a total of 533 individual sites in PA, WV, and NY (Figure 3.2). Almost all of the samples were located in Pennsylvania. When multiple samples were collected at the same site at multiple times

(either on the same day or over multiple days), the site-aggregated mean was used to describe the methane concentration (Table E17). A few of the samples were collected with a slightly different sampling technique (19) or were measured at different laboratories. For example, the SRBC used a slightly different sampling and analytical methodology. Notably, the SRBC used a smaller amber sampling bottle, a peristaltic pump to take up stream water, and they used biocide (KOH). The use of a peristaltic pump and biocide were not observed to alter methane concentrations as compared to

hand-sampled waters or waters without biocide, as reported previously, as long as the methane was analyzed within 5 days (2, 87).

Figure 3.2: Locations of the 533 sites where methane concentrations were measured in streams: increasing circle sizes correspond with increasing site-aggregated mean methane concentrations (i.e. larger circles correspond with larger methane concentrations). The highest individual measured methane concentrations in the dataset for streams (i.e., not site aggregated means) equaled 68.5 µg/L in Meshoppen Creek (Susquehanna County see Appendix C) and 76.7 µg/L in Sugar Run (Lycoming County).

Methane concentrations for all 1,081 samples ranged from <0.06 µg/L to 76.6

µg/L. The detection limit in this study was 0.06 µg/L. The distribution of methane concentrations was skewed (Figure 3.3): the mean methane concentration for all 1,078 samples (3.4 µg/L) was higher than the median concentration (1.1 µg/L). Likewise, the

mean for the total dataset of site-aggregated means (the average of multiple samples at individual sites collected at different times) was higher (2.46 µg/L) than the median (0.8

µg/L).

13 Isotopes (δ C-CH4) were generally only analyzed for samples with CH4 > 4 µg/L except for samples collect by the SRBC where they were measured when methane was >

2 µg/L (analyzed at the University of Arkansas Stable Isotope Laboratory). Analysis for

13 δ C-CH4 for all stream methane samples analyzed in this study ranged from -11.4‰ to -

61.4‰.

3.3.15 Snapshot day stream methane

Many samples were collected during snapshot days (n=285 samples). Data from six snapshot days are included: five completed as part of this study and one completed in the Sinnemahoning watershed from previous work (2). The location of these snapshot days and the number of samples collected are as follows: 1) Sinnemahoning Creek watershed, PA on 9/26/2015 (n=26 samples collected at 26 sites, approximately 20-30

citizen scientist volunteers, and 1 watershed group represented (TU)) (2); 2)

Monongahela National Forest (Cheat and Greenbrier River watersheds), West Virginia

(WV) on 10/1/2016 (n=49 samples at 49 sites by approximately 20-30 citizen scientist volunteers, and 2 watershed groups represented (TU and WV Rivers Coalition)); 3) Pine

Creek watershed, PA on 6/24/2017 (n=61 samples at 61 sites by approximately 20-30 citizen scientist volunteers, and 1 watershed group represented (TU)); 4) 5) and 6) three snapshot days conducted in the Allegheny National Forest, PA on 10/24/2017, 4/10/2018 and 10/30/2018 (n= 36, 76, and 63 respectively) at 175 sites in total by approximately 20-

30 citizen scientist volunteers per event, and 4 watershed groups represented per event

(TU, McKean County Conservation District, Elk County Conservation District, Warren

County Conservation District).

The Monongahela National Forest is largely untouched from development of any kind. The Pine Creek watershed has experienced recent development from unconventional gas activity in the last decade. The Allegheny National Forest contains some of the oldest development for oil and gas in the country. Approximately 50,000

“known” conventional wells have been drilled in the forest based on PADEP estimates

(www. depgis.state.pa.us/emappa) (Figure B-9). The ANF also hosts many abandoned and orphaned oil and gas wells: some are in known locations (i.e., these locations represent roughly half of the 11,000 total known abandoned and orphaned wells in PA listed online by the PA DEP at (http://www.depgis.state.pa.us/PaOilAndGasMapping)).

Mean and median methane concentrations varied for each snapshot day (Figure 3.4).

Overall, snapshot days conducted in the Allegheny National Forest had samples with higher mean and median methane concentrations than any other snapshot day locations.

3.3.2 High methane samples

Following Wendt et al. (2), site-aggregated mean methane concentrations in streams that were < 4.0 µg/L were regarded as not warranting further investigation for potential contamination. In total, 75/533 (=14%) sites from 35 individual streams revealed site aggregated mean methane concentrations > 4 µg/L. Of these sites, 45 were newly sampled as a part of this study (25 streams), 28 sites (10 streams) were reported previously (2, 19, 24), and 2 (2 streams) were sampled solely by the SRBC. When possible, we collected repeat samples to reproduce the measured elevated methane concentrations. The distribution of site-aggregated mean concentrations for these high- methane samples is skewed (Figure 3.5).

Figure 3.5. Histogram of the 75 site-aggregated mean concentrations for stream sites with methane > 4 µg/L.

3.3.3 Categorization of high methane stream samples

Given that our efforts were mostly aimed at reconnaissance and resources were limited, extensive additional analysis such as C and H isotopic analysis for each stream site was not possible. Of the 1081 samples at 533 stream sites, only 96 samples were analyzed for C isotopes. We nonetheless preliminarily binned sites based on available data or observations of scientists and citizen scientists into categories for inferred sources: 1) biogenic; 2) coal; 3) oil + gas; 4) biogenic + oil + gas + coal; 5) biogenic + oil

+ gas; 6); biogenic + coal; 7) oil + gas + coal; 8) unknown. This binning was a preliminary attempt to understand the distribution of the sources of high methane

samples. We found it to be especially helpful to give feedback to watershed groups trying to understand contamination in their areas without undertaking an extensive and expensive environmental forensic study. In order to categorize all sites with methane concentrations > 4.0 µg/L (if multiple samples were taken at one site over multiple trips, the site-aggregated mean was used), several criteria were tested as described below and shown in Figure 3.6. Only stream methane sites were categorized in this manner.

For sites where we visually observed discharging AMD-like waters entering the stream during sampling, the sample was categorized as “coal”. For sites where we visually observed waters affected by a leaking abandoned well entering the stream, the site was categorized as “oil and gas.” For sites where we made isotopic measurements

13 13 (δ C-CH4) on the gas, a threshold of δ C-CH4 < -60‰ was used to infer that methane in a sample was biogenic in origin. In some cases, sites were categorized based on their known association with leaking hydrocarbons published in the literature (1, 88). If isotopic values were not available or no hydrocarbon discharges were observed (as was normally the case), the planar distance of the sample location to the nearest wetland, oil or gas well, or coal mine was calculated using a near function in ArcGIS® to determine if the sample location was within a pre-determined zone of influence for wetlands, oil and gas wells, or coal mines (i.e. 30 m, 760 m, and 300 m respectively). The zone of influence to define a sample as “biogenic” (30m) was defined from previous work (2) as the zone of influence for a wetland by the Fish and Wildlife Service (89). The zones of influence to define samples as being influenced by “coal” and “oil and gas” wells were assumed based on the pre-drill sampling suggestions for coal or oil and gas companies

before their respective mining or drilling (conversation with PA DEP) (7). For example, gas companies typically collect “baseline” samples from homeowner water wells within a

2,500 foot radius of a planned gas well (7).

Wetland spatial data were acquired from the National Wetlands Inventory for

Pennsylvania, a database available from the Pennsylvania Fish and Wildlife Service

(https://www.fws.gov/wetlands/). The database was first scrubbed of wetland types labeled “riverine.” Coal mine and oil and gas well locational data was acquired from available online sources (www. depgis.state.pa.us/emappa and http://www.depgis.state.pa.us/paoilandgasmapping/).

Figure 3.6. Schematic showing how samples were binned into categories for samples with methane concentration > 4 µg/L. Samples were binned into 1 of 8 categories. The binning shown here was designed as a quick tool to categorize sites in a preliminary way.

Sample locations containing high methane were binned into the categories as described above. Of the 75 sites, 11 were binned as “biogenic”, 6 as “coal”, 37 as “oil and gas”, 2 as “biogenic + oil + gas + coal”, 11 as “biogenic + oil + gas”, 0 as “biogenic

+ coal”, 1 as “oil + gas + coal” and 5 as “unknown” (Figure 3.7, 3.8). One of these samples was located near oil and gas development, wetlands, and a landfill as described previously (2). Landfills are known to produce biogenic methane (90), yet the measured

13 δ C-CH4 for the sample indicated a thermogenic signature (-43.5 ‰). The sample was categorized biogenic + oil + gas accordingly because of the multiple possible sources.

This was the only sample known to be located near a landfill.

Of note, bins in this study deviated in some respects from bins described in previous work (1). For example, previous work described a “thermogenic” and

“putatively anthropogenic” bin in order to differentiate sample locations that may contain natural thermogenic methane vs methane from potentially leaking shale gas wells. The latter was reserved for samples collected within 2 km of suspected leaking shale gas wells. The thermogenic category was reserved for two sites (Nine Partners Creek and

Tunkhannock Creek), located within 100 meters of natural geologic lineaments, that were consistent with inputs of biogenic and thermogenic methane and thought to be located far from any known leaking gas wells (2). However, during analysis in this study, a gas well

(API: 115-20457) located ~1 km from the sample locations was cited by the PADEP as leaking methane during the time of sampling

(http://www.depgis.state.pa.us/paoilandgasmapping/). This gas well was subsequently fined ~$300,000 (we assume for methane migration). Therefore, here we categorized one

sample with methane < 4 µg/L from Nine Partners Creek (binned as “unknown” in this study). A “natural thermogenic” bin was not created for samples within this study since most regions were targeted based on known hydrocarbon extraction techniques and leakage could not be discounted nor confirmed.

Figure 3.7: Histograms showing the distributions of site aggregated mean methane concentrations for all bins containing more than 5 samples. See text for criteria used in categories.

3.3.4 Results from groundwater discharges

Groundwater discharges (seeps, springs and leaking abandoned wells) were sampled in 43 locations for a total of 124 samples. Special care was taken throughout this study to locate orange colored discharges similar to those described in Chapter 2 and shown in Figure A1 and B10. These were targeted based on the observation at Sugar Run

in Lycoming County that such discharges could be characterized by elevated methane concentrations. Some of these were identified during the subsequent investigations of high stream methane sites and were based on locations provided by volunteers and in published research (86). For example, 43 discharges lined with orange sediments, smelling of hydrogen sulfide, or leaking directly from metal casings were sampled. It was observed that almost all contained elevated methane and sometimes ethane. Of these sample locations, 14 were deemed to have been impacted by AMD based on their proximity to coal mines or to observed mine infrastructure (i.e. mine shaft or opening

Figure B-11) or because they were listed as coal mine discharge in the literature (86, 91) or because they were listed on public databases (www. depgis.state.pa.us/emappa).

Details of some of these sites are discussed in section 3.3.4. Of the 43 discharges, 8 were flowing directly from abandoned oil or gas wells. These discharges are discussed in section 3.3.7 as leaking abandoned wells (LAWs).

3.3.5 Abandoned Mine Drainage

Methane measured in samples collected directly from discharges of AMD (n= 30 samples from 14 sites) ranged in methane concentration from 49.8 to 549 µg/L (Figure

3.9).

Figure 3.9. Histograms of site-aggregated means methane concentrations measured in discharges categorized as acid mine drainage (AMD) or gas leak drainage (GLD) or leaking abandoned wells (LAW).

13 δ C-CH4 values measured at two of these sites (known as the Gladding discharge

(http://www.chartiersgreenway.net) and Blythdale discharge (92)) were -31.0‰ and

-43.0‰ respectively (Table E14). Methane concentrations measured downstream from

13 the Gladding discharge (Figure B-12) decreased with distance as δ C-CH4 stayed relatively constant (Figure 3.10). Measured values of pH and SPC ranged from 3.4 to 6.3 and 1400 to 2610 (µS/cm) respectively. Iron (Fe) and manganese (Mn) measured in water from three AMD-affected sites ranged from 36.9 mg/L to 62.0 mg/L and 0.6 mg/L to 1.2 mg/L respectively. Chloride (Cl) and sulfate (SO4) measured in samples from 6

discharges ranged from 71 mg/L to 380 mg/L and 64 mg/L to 587 mg/L respectively.

Additional cations and anions are reported below (Table E15, E16). Trace metals were not analyzed for AMD-affected discharges.

250 -30

-31 )

200 -32 µg/L -33

150 -34 4

CH -

-35 C 13

100 -36 δ

Methane -37 50 -38 Isotope MethaneConcentration ( -39 0 -40 0 50 100 150 200 Distance Dowsntream (meters)

13 Figure 3.10: Methane concentration (µg/L) and δ C-CH4 vs distance downstream from a discharge of AMD (the Gladding discharge) (Table E14-E16). The decrease in methane downstream is attributed to degassing as opposed to oxidation because of the relative 13 constancy in δ C-CH4.

3.3.6 Gas leak drainage (GLD)

Of the 43 groundwater discharges that were sampled, 13 were characterized by orange colored sediment, oil, or rotten egg smell, but were not located near known coal mining.

An additional 8 samples were located within 1,200 and 2,000 m of a known coal mine but without any obvious or recorded discharge associated with those mines.

These discharges are included in the dataset discussed below (see also Table E11-

E13). Several of these were observed to have high methane and ethane concentrations

(typically higher than AMD samples as observed in Figure 3.9). In measurements collected from these discharges, the pH ranged from 5.77 to 7.07 (with one outlier with a pH of 9.07). SPC for these discharges ranged from 51 to 425 (µS/cm). Iron (Fe) and manganese (Mn), measured on water intermittently at sites, ranged from 0.009 mg/L to

17.8 mg/L and 0.01 mg/L to 2.7 mg/L respectively. Arsenic (As), also measured on a few samples, ranged from <1 µg/L to 20.1 µg/L. Fe, Mn, and As at their highest concentrations were above US Environmental Protection Agency (US EPA) drinking water standards at several sites. Chloride and sulfate concentrations ranged from 0.3 to

53.5 mg/L and 0.3 to 10.4 mg/L respectively. Additional cations and anions are reported in Table E12, E13.

Many of these sites resembled (both visually and chemically) the presumably impacted seeps observed in Sugar Run that appear to be contaminated by gas migration

(1). For example, many of the sites of these discharges were characterized by abundant white or orange slimy material that was presumed to be bacterial growth (Figure B-10).

In addition, methane measured in water from these springs (Table E11) was observed in some cases to be extremely high, ranging from 0.11 to 11,200 µg/L (Figure 3.9). Ethane, analyzed in some of these samples, ranged from 0.58 to 2,450 µg/L (Table E11). Molar ratios of methane (C1) / ethane (C2) for samples where both methane and ethane were

13 13 measured ranged from 3 to 578. The measurements of δ C-CH4 and δ C-C2H6 made for a few of these springs ranged from 37.6‰ to -65.7‰ and -28.2‰ to -35.7‰ respectively

(Table E11). In terms of location, 7 of these types of discharges were observed and sampled in the ANF and six of the 7 were located within 350 meters of conventional oil and gas wells spudded (drilled) prior to 1955. The chemistry of these springs are discussed in section 3.4.5 below and summarized in Table E11-E13.

These groundwater discharges in the field appeared like AMD but they are chemically distinct from AMD. In particular, the discharges are associated with low sulfate concentrations compared to abandoned mine drainage: the minimum, maximum and median sulfate concentrations in these samples (0.3 mg/L, 10.4 mg/L and 5.5 mg/L respectively) are considerably different from the minimum, maximum and median sulfate concentrations measured in water samples from discharges from the 144 abandoned mine drainage sites summarized for PA (34 mg/L, 2000 mg/L and 510 mg/L respectively) in published literature (91). Given all of these lines of evidence, we hypothesize that these types of discharge are a new type of contaminated water which we tentatively name as

“gas leak drainage” or GLD. The chemistry of GLD was also compared to the chemistry of waters sampled in leaking abandoned/orphaned oil and gas wells as described in the next section.

3.3.7 Leaking abandoned wells (LAW)

Samples were collected at 8 locations (Figure B-8) where fluids were discharging from metal casings that looked like abandoned wells (see Figure B-7 for photographs). At most of these sites, water pooled around the casings showed active bubbling. Like the

GLD samples, methane was elevated in these waters (309 to 23,200 µg/L)(Figure 3.9).

Waters sampled in the casing of abandoned wells differed from other springs sampled

(GLD and AMD) for multiple constituents: unlike AMD they were circumneutral in pH and sulfate-depleted, and unlike both AMD and GLD, they were metal-depleted and chloride-enriched (Table E8-E11). For example, measurements of pH in water at these sites were typically in the range of 5.0 to 9.1 (Table E10). These sites also typically contained lower concentrations of heavy metals such as Fe (<0.01 to 0.57 mg/L) and Mn

(0.03 to 0.15 mg/L) when compared to either GLD or AMD. Sulfate concentrations in these samples were typically low (0.37 to 9.25 mg/L), similar to GLD samples, while chloride concentrations were much more elevated as compared to GLD samples (14.1 to

3,300 mg/L). Of the 8 locations sampled from abandoned oil/gas wells, arsenic was elevated above EPA drinking water standards (Table E10) in two samples (10.5 and 19.2 ug/L). This was similar to observations for GLD samples.

3.4 Discussion

In this section we discuss which sources of methane are inferred to be leaking into streams and test the three hypotheses listed previously: 1) older oil and gas wells leak methane more than younger wells (section 3.4.1); 2) coal mines do not leak methane when they are flooded (section 3.4.2); 3) discharges of leaking hydrocarbons from oil and gas wells create a new type of discharge— gas leak drainage-- that is chemically distinct from AMD and can sometimes be highly concentrated in ABB salts (section

3.4.3).

3.4.1 The effect of spud date on methane release from oil/gas wells

The dataset of site-aggregated means for samples collected from 115 stream methane sites during the three snapshot days in the Allegheny National Forest (ANF) yielded the highest median methane concentrations of any snapshot day location (Figure

3.4). The ANF is also a site where drilling of oil and gas wells has occurred recently and long in the past and where very few unconventional gas wells have been drilled. At the same time, the predominantly forested region is mostly lacking in coal mining and other similar contaminant sources such as landfills. For all of these reasons, we chose the ANF as the optimal region to investigate leakage from old oil and gas development. The ANF is also the location of some of the oldest oil and gas wells in the U.S.A.

(http://www.depgis.state.pa.us/paoilandgasmapping/).

Twenty-one stream sites within the ANF showed site-aggregated mean methane concentrations > 4 µg/L. Of these, 12 were binned as “oil + gas” and 7 as “biogenic + oil

13 + gas” (Table E18). Of the 14 sites where δ C-CH4 were analyzed in samples, 6 had

13 δ C-CH4 values > -50‰ (Table E17), indicating that gas could be of thermogenic or mixed thermogenic biogenic origin (84). Three of these samples contained detectable

13 ethane (Table E17), a hydrocarbon that is rarely found in biogenic gas (42). δ C-CH4 measured in the rest of the samples (ranging between -50‰ and -60‰) is consistent with the interpretation that the samples contain biogenic or a mixture of biogenic and thermogenic gas. The abundance of high, potentially thermogenic, methane samples in the ANF is of interest given the large number of active and abandoned conventional oil and gas wells (Figure B-9). Many of these wells are old, aging and drilled at a time when casing, cementing or plugging standards were less stringent than they are today.

Using the data collected from the ANF, we test the hypothesis that the frequency of leakage from older wells is higher than younger wells. Here, we look at stream methane samples collected during snapshot days and follow up trips (n=189) and compare the methane concentrations in these samples to the distances to the nearest conventional gas or oil well within the counties that encompass the ANF (Elk, Warren,

Forest, McKean).

After comparisons of methane concentrations with respect to distance to the nearest well, we further subdivided and performed the same analysis on wells spudded

(drilled) during different timeframes chosen to investigate the effects of different regulations or the introduction of new casing or cementing standards (64, 65).

Specifically, we investigated 1) wells drilled prior to 1955, the date of passage of Act

225; 2) wells drilled after 1955 but before 1984 where 1984 marks the passage of the Oil

and Gas act 3) wells drilled after 1984 and before 2005 where 2005 marks when the first unconventional gas wells were drilled in Pennsylvania; and 4) wells drilled after 2005.

Act 22 created standards for drilling in areas where coal mining was active and ongoing and the Oil and Gas act replaced Act 225 with more stringent regulations for drilling, including safeguards for water and wetlands.

In a plot of stream methane vs distance to nearest gas well, the largest methane concentrations were observed for some samples collected closer to oil and gas wells

(Figure 3.11). This is not necessarily surprising given the high density of oil and gas wells within the Allegheny National Forest: in other words, if some other factor caused high methane concentrations randomly, several of the high values would be likely to be near a well in the ANF (Figure B-9). Strikingly however, when plotting methane concentrations for gas and oil wells spudded before 1955 and after 2005, the before-1955 figure shows the largest methane concentrations are found in waters sampled closest to the oldest gas and oil wells (Figure 3.11). This is not observed in the after-2005 figure

(Figure 3.11) or in any other subdivisions of the wells (Figure B-13).

At least two reasons could explain this observation. First, older oil and gas wells could be leaking methane into groundwater (that then enter streams) because of inadequacies in drilling at the time of well emplacement or inadequacies of plugging upon abandonment. Indeed, improper abandonment or plugging has been suspected for wells drilled before 1957 (65). Second, it could be that the oldest oil and gas wells were drilled in areas where hydrocarbons are shallow and easiest to obtain. Such hydrocarbons could naturally make their way to the surface regardless of the presence of an oil/gas well

(7, 93). It is important to note that although we observe the highest stream methane sites near the oldest subset of wells, there is no significant statistical correlation using common statistical methods (i.e. Pearson and Spearman) or when using different functions to fit data (i.e. linear, logarithmic, exponential and power functions). This lack of a trend highlights a major drawback in stream sampling, notably dilution. For example, because of dilution and assuming that methane input is the same, small streams will show higher stream methane concentrations than larger streams. In addition, methane contamination can be extremely localized, with concentrations quickly decreasing downstream from the input location as observed previously (1,2) and shown in Figure 3.10 for an AMD- affected location. Thus, finding strong correlations is unlikely in most reconnaissance scenarios.

Figure 3.11 – (A) Plot of site-aggregated mean methane concentrations for stream sites in the Allegheny National Forest vs distance to the nearest conventional oil or gas well. To calculate distance to well, all oil/gas wells in the ANF in the four counties mentioned in main text were considered. (B) Same as (A) except that only conventional oil or gas wells spudded prior to 1955 were considered. (C) Same as (A) and (B) except only conventional oil or gas wells spudded between 2005 were included. Additional plots for different date ranges are shown in the appendix (Figure B-13)

3.4.2 Methane release from flooded coal mines

In this section we test the assertion made by multiple sources that flooded coal mines do not release methane to the atmosphere (78) after 14-15 Years (76, 94). These assertions are based on methane fluxes measured from mine vents into the air and not for

methane released as a species dissolved in water. Specifically, we tested to see how much methane is released in discharge waters from abandoned coal mines. It is estimated that

20-48% of mines are flooded in the Appalachian Basin (76) and flooded coal mines have the potential to leak fluids to the surface if an outlet is available or if groundwater flow paths to the surface exist. Possible outlets include: natural groundwater flow paths, abandoned mine openings, fractures, abandoned wells (95) or boreholes installed for the purpose of draining AMD impacted groundwater (86).

Of the 12 stream sites with methane concentrations > 4 µg/L that we sampled in the coal mining region of southwestern Pennsylvania, 6 were binned as “coal”, 3 as “oil + gas”, 2 as “bio + oil + gas + coal”, and “1 as oil + gas + coal” (Figure B-14). In some cases, mine discharges were observed upstream of stream sites with elevated methane concentrations. For example, on 10/18/2016, a volunteer collected a sample with high methane (19.6 µg/L) in Coal Run near Pittsburgh PA. During a follow up trip, this same volunteer noted discharging seepages in the stream bed itself, identified by the “boiling” at the surface above cracks in the streambed. Methane samples collected at the discharge location were much higher in methane concentration (up to 149 µg/L) than other parts of the stream.

Based on the discovery at Coal Run, a small study was conducted at 14 discharges spanning the anthracite coal region of northeastern PA and bituminous coal region of western PA to observe if methane was present in various reported mine discharges.

Methane concentrations were observed to be elevated > 20 µg/L (and usually >100 µg/L) in all of these coal-related mine discharges measured, in both the anthracite region of

northeastern PA and bituminous coal region of western PA. The mean methane concentration for discharges where methane was analyzed was 141 µg/L. Methane sampling downstream from the discharge at one of these sites (see Figure B-12) showed that methane concentrations quickly decreased and ultimately dropped to <1 µg/L after the discharge tumbled off a large waterfall (Figure 3.10, B-12), documenting that methane readily degasses to the atmosphere from the mine waters.

The importance of methane discharge from flooded coal mines in terms of degassing to the atmosphere can be roughly estimated for PA if we assume that all methane dissolved in the discharges degasses. To approximate the flux of methane to the atmosphere from sampled AMD discharges, we multiplied the measured methane concentrations by the known discharge rates for the 10 discharge points where both methane and discharge measurements are available ((86); www. depgis.state.pa.us/emappa). A total methane flux of 16,000 kg of methane per year was calculated for these ten sites. The mean for the ten sites was 1,600 kg per year per discharge. This estimate per coal mine discharge is much higher than the means reported for individual abandoned oil and gas wells in PA: 98.55 kg year/well (68). However, there are likely fewer discharges from coal mines than methane-emitting abandoned wells in the state. For example, although the exact number is not known, the PA DEP estimates at least 5,000 abandoned, orphaned, or operating coal mines and roughly 11,000 abandoned, orphaned, or operating oil and gas wells (www. depgis.state.pa.us/emappa).

However, sources in the literature (67-69) estimate that there should be well over 300,000 abandoned, orphaned, or operating oil and gas wells indicating that location of the

majority of wells is unknown. This discrepancy could suggest that PA DEP’s estimates for abandoned coal mines and mine discharges are also underrepresented. Regardless, the

PA DEP has documented 323 discharges (www. depgis.state.pa.us/emappa) with a total flow rate of 525,078 GPM or 1*1012 L/yr. If we assume that 323 discharges are present in

PA with a mean methane concentration of 141 µg/L, and that all discharges released methane to the atmosphere, coal mine discharge would contribute more than 150,000 kg of CH4 per year to the atmosphere. Nonetheless, the rate of loss only represents about

0.02% of the statewide man-made emissions of methane to the atmosphere as cited by

Kang et al., (68). This author estimated statewide anthropogenic methane emissions using online data (https://www.wri.org/our-work/project/cait-climate-data-explorer). Our estimates for PA are also much lower than estimated emissions from all non-flooded coal mines (9% man made emissions in the United States) as reported by the EPA (EPA 2017)

8 or from estimates of leakage of unconventional gas wells in Pennsylvania 4 x 10 kg CH4 per year (70). Of course, this studies calculation does not represent the flux from all mine discharges (as many are likely unrecorded in available datasets) and is thus a minimum value. In fact, the median of stream methane concentrations for 62 sites in the heavily coal-mined southwest PA region (2.4 µg/L) is double the median for all sites sampled in this study across the Appalachian Basin in PA, WV, and NY (Figure 3.4). This could suggest that coal-related methane is also migrating into streams via groundwater seepage.

3.4.3 An unnamed type of contamination: gas leak drainage (GLD)

In this section we explore the hypothesis that gas leak drainage is a newly identified type of contamination in the Appalachian Basin. We infer that it may have a distinct water chemistry compared to AMD and that in some cases contain high salt concentrations in ratios similar to ABB. Specifically, almost all samples of GLD that we identified had high methane concentrations. When the waters were discovered relatively far from wells they were low in salt concentration but when waters of GLD were discovered to be emitting directly from casings of abandoned oil or gas wells (LAW) they were high in salts.

The discovery and identification of GLD began with the work at Sugar Run

(Chapter 2). In addition, throughout this study and in various locations (Figure B-8), volunteers and scientists observed springs with bright orange discoloration, oil or noxious rotten egg smell (Figure A-1, B-10, C-1), and in some cases, abundant evidence for bacterial growth (Figure B-10). These characteristics are similar to contaminated seeps impacted by hydrocarbon migration that were discussed in Sugar Run (1). In addition, many of these discharges were located near stream sample locations that contained elevated stream methane. These sites, which we are referring to here as gas leak drainage

(GLD), were unique in that they contained elevated hydrocarbons (methane and sometimes ethane). For some of these discharges, C1/C2 ratios were extremely low (i.e. below 10), consistent with an oil rather than gas source (96). In addition, these springs were typically characterized by elevated metals such as Fe, Mn and As, low SPC (range:

50-424 (µS/cm)), slightly acidic to neutral pH (5.8-7.08) except for one alkaline sample

(9.1 which was relatively metal-poor, sulfate poor and chloride intermediate), typically low ORP, low sulfate, and low chloride (Table E11-E13). Waters from these springs usually exhibited a high molar concentration ratio of Ca/Na, consistent with shallow recharge as observed in other Pennsylvania waters (22, 46). Most of these seepages were located far from known coal mining locations (> 2 km and typically much further) and thus were not AMD. Samples that were tentatively identified as GLD based on characteristics above but that were located between 1.2 and 2 km to a known coal mine are identified as “suspected GLD” on the basis that they could be GLD or AMD from their location (Figure 3.12).

We can also compare the waters identified as GLD to waters discharging directly from abandoned oil/gas wells (LAW) in order to determine which of the chemical characteristics are really diagnostic of gas leakage (Table E8-E13). These waters were sampled directly from flooded casings that were observed to be bubbling, and they typically showed high methane, ethane, and other hydrocarbons (when the higher chain compounds were analyzed) as well as circumneutral pH and low sulfate concentrations

(Table E8-E10). On the other hand, they also showed low metal concentrations and high chloride concentrations (Table E9-E10). If GLD is related to gas migration and if LAW is impacted by gas, then one inference is that when gas migrates meters to kilometers from an oil/gas well, the discharge chemistry can be distinct from when the gas emits directly up the flooded casing of the well and waters are sampled at that location.

In Figure 3.12, we compare 14 springs that we suspect to have been impacted by oil or gas migration (tentatively identified as GLD) to discharges affected by AMD and to

the waters sampled directly from abandoned well casings. Here, we compare suspected

GLD samples collected in this study (n=60 samples) and previous work (1) to AMD samples collected in this study (n=3 samples) or reported by Hedin et al. (95) (n=20) or

Cravotta et al. (91) (n = 140) and also to samples collected directly from the casings of abandoned oil and gas wells in this study (n=5). Because the latter samples are high in salt concentrations that are similar in cation ratio to Appalachian Basin Brines (ABB), brine samples reported in the USGS produced waters database for conventional and unconventional oil and gas wells in PA (n=226) (97) are also reported. These produced waters are considered to be ABB (11). Because of the similarities between GLD and

LAW, we have lumped them into the category of GLD on Figure 3.12.

Based on Figure 3.12, GLD appear mostly chemically distinct from AMD and

ABB, although some of the low Ca/Na waters overlap with ABB. In particular, LAW samples (lumped with GLD) overlap with the ABB. The figure shows that the high

Ca/Na values of most of the rest of the GLD samples are inconsistent with waters affected by brine-rich fluids. Also, as discussed in previous work (1) the relatively high metal concentrations (Fe, Mn, As) observed in GLD waters could be caused by oxidation of some of the hydrocarbons coupled to metal reduction, releasing more soluble reduced metals. The relatively low sulfate concentrations could be be related to sulfate reduction or precipitation of metals sulfides because such reduction in the presence of high metal concentrations is typically accompanied by precipitation of metal sulfides (38).

Figure 3.12. A plot of Ca/Na vs SO4 for different types of waters: abandoned mine drainage (AMD)(this study, Hedin et al. (95) Cravotta et al. (91)), produced waters from conventional (conventional brines) and unconventional wells (unconventional brines) (USGS Produced Waters Database), gas leak drainage (GLD), including waters from abandoned wells. Waters from leaking abandoned wells are lumped together here into the region of GLD. GLD appear chemically distinct from other sources, except for one anomalous sample. Waters sampled in casing of leaking abandoned wells contain brine salts similar to ABB and are thus similar to discharges receiving brine-rich formation waters such as the salt spring in Salt Spring State Park located in Montrose, PA. Suspected GLD samples represent the subset of samples located within 2 km of a known coal mine. One GLD sample (Sample SRS 1.5 located near Sugar Run, Lycoming) contains an anomalously low Ca/Na ratio. This same sample (SRS 1.5)was also anomalously elevated in pH (9.01), low in metals, and appeared visually distinct from other seepages (i.e. dark black sediment surrounding the bubbling area). The double- sided arrow labeled “Ca/Na shallow groundwater” represents the median range of Ca/Na ratios for groundwaters sampled in northeastern PA aquifers of the Catskill, Lock Haven and Battelle formations and reported in previous work (46).

3.5 Conclusion

Overall, the partnerships with citizen scientists and other organizations has been integral to identifying surface water contamination trends and to observing the legacy of hydrocarbon recovery in the Appalachian Basin. Notably, and in agreement with our hypothesis, oil and gas development may have caused leakage into the subsurface particularly around the oldest oil and gas wells. This has been inferred from the location of stream samples with elevated methane concentrations within the ANF and through visual observation of leaking abandoned wells. Further analysis of springs that are apparently impacted by oil and gas wells has led to the identification of a previously unnamed form of contaminated discharge that appears somewhat like abandoned mine drainage but with a distinct geochemical signature (also in agreement with our hypothesis)—here called “gas leak drainage” (GLD). Springs impacted by GLD, observed in various locations in northern Pennsylvania, present a concern to local ecosystems because they are sometimes elevated in concentrations of metals (Fe, Mn, As) and hydrogen sulfide. GLD was also sampled directly from the casings of leaking abandoned wells that also may present a concern for local ecoystems and stream health.

GLD sampled in the casings were observed to often be higher in concentration of ABB salts and higher-chain hydrocarbons. In addition, observations of elevated methane in streams and discharges around coal mining led to the conclusion that although it is well known that flooded coal mines commonly leak abandoned mine drainage, they also leak methane dissolved in the AMD waters that then degasses to the atmosphere. We roughly

estimate that this total emission to the atmosphere from dissolved methane in coal-mining discharges across PA > 150,000 kg/yr and is likely higher.

Finally, water sampling in regions with unconventional shale gas activity, but without known problematic wells, generally revealed streams with lower methane concentrations. As noted by many others (1, 20) this could suggest that leakage from unconventional gas wells is rare compared to the number of shale gas wells. Thus, observations from this work suggest that legacy hydrocarbon extraction results in more widespread, systematic contamination of waters where such legacy exists, than recent and highly publicized unconventional oil and gas development. The lack of attention to contamination from legacy hydrocarbon extraction techniques should provide a strong lesson that monitoring of legacy and shale gas development will be needed long into the future.

References for Chapter 2 and 3

1. Woda J, et al. (2018) Detecting and explaining why aquifers occasionally become degraded near hydraulically fractured shale gas wells. Proc Natl Acad Sci. doi:10.1073/pnas.1809013115. 2. Wendt AK, et al. (2018) Scientist-nonscientist teams explore methane sources in streams near oil/gas development. J Contemp Water Res Educ 164(1):80–111. 3. US Energy Information Administration (2018) Annual Energy Outlook, Energy Information Administration. Available at: https://www.eia.gov/outlooks/aeo/ [Accessed April 3, 2018]. 4. Brantley SL, et al. (2014) Water resource impacts during unconventional shale gas development: The Pennsylvania experience. Int J Coal Geol 126:140–156. 5. Li Z, et al. (2016) Searching for anomalous methane in shallow groundwater near shale gas wells. J Contam Hydrol 195:23–30. 6. Vidic RD, Brantley SL, Vandenbossche JM, Yoxtheimer D, Abad JD (2013) Impact of shale gas development on regional water quality. Science 340(May):1235009. 7. Siegel DI, Azzolina NA, Smith BJ, Perry AE, Bothun RL (2015) Methane concentrations in water wells unrelated to proximity to existing oil and gas wells in northeastern Pennsylvania. Environ Sci Technol 49(7):4106–4112. 8. Baldassare FJ, McCaffrey MA, Harper JA (2014) A geochemical context for stray gas investigations in the northern Appalachian Basin: Implications of analyses of natural gases from Neogene-through Devonian-age strata. Am Assoc Pet Geol Bull 98(2):341–372. 9. Whiticar MJ (1999) Carbon and hydrogen isotope systematics of bacterial formation and oxidation of methane. Chem Geol 161(1–3):291–314. 10. Warner NR, et al. (2012) Geochemical evidence for possible natural migration of Marcellus Formation brine to shallow aquifers in Pennsylvania. Proc Natl Acad Sci 109(30):11961–11966. 11. Llewellyn GT (2014) Evidence and mechanisms for Appalachian Basin brine migration into shallow aquifers in NE Pennsylvania, USA. Hydrogeol J 22(5):1055–1066. 12. Christian K, et al. (2015) Methane occurrence is associated with sodium-rich valley waters in domestic wells overlying the Marcellus shale in New York State. Water Resour Res 52:206–226. 13. Harkness JS, et al. (2017) The geochemistry of naturally occurring methane and saline groundwater in an area of unconventional shale gas development. Geochim Cosmochim Acta 208:302–334. 14. Kreuzer RL, et al. (2018) Structural and hydrogeological controls on hydrocarbon and brine migration into drinking water aquifers in southern New York. Groundwater 56(2):225–244. 15. Darrah TH, et al. (2015) The evolution of Devonian hydrocarbon gases in shallow aquifers of the northern Appalachian Basin: Insights from integrating noble gas and hydrocarbon geochemistry. Geochim Cosmochim Acta 170:321–355.

16. Darrah TH, Vengosh A, Jackson RB, Warner NR, Poreda RJ (2014) Noble gases identify the mechanisms of fugitive gas contamination in drinking-water wells overlying the Marcellus and Barnett Shales. Proc Natl Acad Sci 111(39):14076– 14081. 17. Llewellyn GT, et al. (2015) Evaluating a groundwater supply contamination incident attributed to Marcellus Shale gas development. Proc Natl Acad Sci 112(20):6325–6330. 18. Wen T, et al. (2017) Characterizing the noble gas isotopic composition of the Barnett Shale and Strawn Group and constraining the source of stray gas in the Trinity Aquifer, north-central Texas. Environ Sci Technol 51(11):6533–6541. 19. Grieve PL, et al. (2018) Using environmental tracers and modelling to identify natural and gas well-induced emissions of methane into streams. Appl Geochemistry 91:107–121. 20. Wen T, et al. (2018) Big groundwater datasets reveal possible rare contamination amid otherwise improved water quality for some analytes in a region of Marcellus Shale development. Environ Sci Technol 52:7149–7159. 21. Gorody AW (2012) Factors affecting the variability of stray gas concentration and composition in groundwater. Environ Geosci 19(1):17–31. 22. Molofsky LJ, et al. (2016) Environmental Factors Associated With Natural Methane Occurrence in the Appalachian Basin. Groundwater 54(5):656–668. 23. PA DEP OG Compliance - Report Viewer. Available at: http://www.depreportingservices.state.pa.us/ReportServer/Pages/ReportViewer.asp x?/Oil_Gas/OG_Compliance [Accessed January 1, 2018]. 24. Heilweil VM, et al. (2015) Stream measurements locate thermogenic methane fluxes in groundwater discharge in an area of shale-gas development. Environ Sci Technol 49(7):4057–4065. 25. Heilweil V, Risser D, Conger R, Grieve P, Hynek S (2014) Estimation of Methane Concentrations and Loads in Groundwater Discharge to Sugar Run, Lycoming County, Pennsylvania. US Geol Surv Open-File Rep 2014-1126. 26. Ingraffea AR, Wells MT, Santoro RL, Shonkoff SBC (2014) Assessment and risk analysis of casing and cement impairment in oil and gas wells in Pennsylvania, 2002-2012. Proc Natl Acad Sci 111(30):10955–10960. 27. The Pennsylvania Environmental Hearing Board (2017) Commonwealth of Pennsylvania Department of Environmental Protection vs Range Resources Available at: http://ehb.courtapps.com/public/document_shower_pub.php?csNameID=5093 [Accessed May 1, 2018]. 28. PA DEP (2015) DEP Reaches Penalty Agreements with Three Natural Gas Exploration Companies in the Northern Tier. Available at: http://www.media.pa.gov/Pages/DEP_details.aspx?newsid=487. [Accessed March 1, 2018]. 29. Wen T, et al. (2016) Methane sources and migration mechanisms in shallow groundwaters in Parker and Hood Counties, Texas - A heavy noble gas analysis. Environ Sci Technol 50(21):12012–12021.

30. Wen T, Castro MC, Hall CM, Pinti DL, Lohmann KC (2015) Constraining groundwater flow in the glacial drift and Saginaw aquifers in the Michigan Basin through helium concentrations and isotopic ratios. Geofluids 16(1):3–25. 31. PA DEP eMapPA. Available at: http://www.depgis.state.pa.us/emappa/ [Accessed January 1, 2017]. 32. Range Resources - Appalachia, LLC, Green Valley Rd Water Well Investigation Moreland TWP, Lycoming Co., PA, (2012) Available at: http://www.rangeresources.com/docs/default-source/lycoming/appendix-iv-range- presentation-to-pa-dep-10-1-12.pdf?sfvrsn=2. 33. Range Resources - Appalachia, LLC, Green Valley Rd Water Well Investigation Moreland TWP, Lycoming Co., PA (2013) Available at: http://www.rangeresources.com/docs/default-source/lycoming/range-final-report- 4-8-12.pdf?sfvrsn=2. 34. Shale Network (2015) doi:10.4211/his-data-shalenetwork. 35. Gross E, Cravotta III C (2016) Groundwater Quality for 75 Domestic Wells in Lycoming County, Pennsylvania, 2014. 36. Gas L (2013) Lower and Upper Explosive Limits for Flammable Gases and Vapors (LEL/UEL) Available at: http://www.chrysalisscientific.com/pg443- Lower-LEL-Upper-UEL-Explosive-Limits.pdf. 37. Molofsky LJ, et al. (2016) Effect of different sampling methodologies on measured methane concentrations in groundwater samples. Groundwater 54(5):669–680. 38. Wolfe AL, Wilkin RT (2017) Evidence of Sulfate-Dependent Anaerobic Methane Oxidation within an Area Impacted by Coalbed Methane-Related Gas Migration. Environ Sci Technol 51(3):1901–1909. 39. Oxburgh E., O’Nions R., Hill R. (1986) Helium isotopes in sedimentary basins. Nature 324:632–635. 40. Ozima M, Podosek F (2002) Noble Gas Geochemistry (Cambridge University Press, New York). 41. US Environmental Protection Agency (2015) Retrospective Case Study in Northeastern Pennsylvania Study of the Potential Impacts of Hydraulic Fracturing on Drinking Water Resources. 42. Schoell M (1980) The hydrogen and carbon isotopic composition of methane from natural gases of various origins. Geochim Cosmochim Acta 44(5):649–661. 43. Reese S, Neboga V V, Pelepko S, Kosmer WJ, Beattie S (2014) Groundwater and Petroleum Resources of Sullivan County, Pennsylvania. 44. Graham DW (2002) Noble Gas Isotope Geochemistry of Mid-Ocean Ridge and Ocean Island Basalts: Characterization of Mantle Source Reservoirs. Rev Mineral Geochemistry 47(1):247–317. 45. Hunt AG, Darrah TH, Poreda RJ (2012) Determining the source and genetic fingerprint of natural gases using noble gas geochemistry: A northern Appalachian Basin case study. Am Assoc Pet Geol Bull 96(10):1785–1811. 46. Siegel DI, Smith B, Perry E, Bothun R, Hollingsworth M (2015) Pre-drilling water-quality data of groundwater prior to shale gas drilling in the Appalachian

Basin: Analysis of the Chesapeake Energy Corporation dataset. Appl Geochemistry 63(June):37–57. 47. Faill R (1979) Geology and Mineral Resources of the Montoursville South and Muncy Quadrangles and Part of the Hughesville Quadrangle, Lycoming, Northumberland, and Montour Counties, Pennsylvania (Pennsylvania Geological Survey Atlas). 48. Engelder T, Lash GG, Uzcátegui RS (2009) Joint sets that enhance production from Middle and Upper Devonian gas shales of the Appalachian Basin. Am Assoc Pet Geol Bull 93(7):857–889. 49. Lisle R (1994) Detection of zones of abnormal strains in structures using gaussian curvature analysis. Am Assoc Pet Geol Bull 78(12):1811–1819. 50. Schultz-Ela D, Yeh J (1992) Predicting fracture permeability from bed curvature. Rock Mech 33:579–590. 51. Fischer MP, Wilkerson MS (2000) Predicting the orientation of joints from fold shape: Results of pseudo-three-dimensional modeling and curvature analysis. Geology 28(1):15–18. 52. Hennings P, Olson JE, Hennings PH, Olson JE, Thompson LB (2000) Combining Outcrop Data and Three- Dimensional Structural Models to Characterize Fractured Reservoirs : An Example from Wyoming. Am Assoc Pet Geol Bull 84(6):830–849. 53. Hancock PL, Engelder T (1989) Neotectonic joints. Geol Soc Am Bull 101(10):1197–1208. 54. Bair ES, Freeman DC, Senko JM (2010) Subsurface Gas Invasion Bainbridge Township, Geauga County, Ohio Available at: http://oilandgas.ohiodnr.gov/resources/investigations-reports-violations- reforms#THR. 55. Lloyd O., Carswell L. (1981) Groundwater resources of the Williamsport region, Lycoming County, Pennsylvania (Pennsylvania Geological Survey). 56. Schout G, Hartog N, Hassanizadeh SM, Griffioen J (2018) Impact of an historic underground gas well blowout on the current methane chemistry in a shallow groundwater system. Proc Natl Acad Sci 115(2):296–301. 57. Alawattegama SK, et al. (2015) Well water contamination in a rural community in southwestern Pennsylvania near unconventional shale gas extraction. J Environ Sci Heal Part A 50(5):516–528. 58. Amos RT, et al. (2012) Evidence for iron-mediated anaerobic methane oxidation in a crude oil-contaminated aquifer. Geobiology 10(6):506–517. 59. Van Stempvoort D, Maathuis H, Jaworski E, Mayer B, Rich K (2005) Oxidation of fugitive methane in ground water linked to bacterial sulfate reduction. Ground Water 43(2):187–199. 60. Kelly WR, Matisoff G, Fisher JB (1985) The effects of a gas well blow out on groundwater chemistry. Environ Geol Water Sci 7(4):205–213. 61. McMahon PB, Belitz K, Barlow JRB, Jurgens BC (2017) Methane in aquifers used for public supply in the United States. Appl Geochemistry 84:337–347. 62. Bergbauer S, Pollard DD (2002) How to calculate normal curvatures of sampled geological surfaces. J Struct Geol 25(2):277–289.

63. Myhre G, et al. (2013) Anthropogenic and natural radiative forcing. Climatic Change, pp 659–740. 64. Carter KM, Harper JA, Schmid KW, Kostelnik J (2011) Unconventional natural gas resources in Pennsylvania: The backstory of the modern Marcellus Shale play. Environ Geosci 18(4):217–257. 65. Dilmore RM, Sams JI, Glosser D, Carter KM, Bain DJ (2015) Spatial and Temporal Characteristics of Historical Oil and Gas Wells in Pennsylvania: Implications for New Shale Gas Resources. Environ Sci Technol 49(20):12015– 12023. 66. Oil and Gas Wells, Drilling Regulating Act of May 17, 1921 (1921) Available at: http://www.legis.state.pa.us/WU01/LI/LI/US/PDF/ 1921/0/0322. 67. Kang M, et al. (2016) Identification and characterization of high methane-emitting abandoned oil and gas wells. Proc Natl Acad Sci:201605913. 68. Kang M, et al. (2014) Direct measurements of methane emissions from abandoned oil and gas wells in Pennsylvania. Proc Natl Acad Sci U S A 111(51):18173–7. 69. Amy Townsend-Small, Thomas W. Fferrara (2016) Emissions of coalbed and natural gas methane from abandoned oil and gas well in United States. AGU Publ 17(4):21–24. 70. Omara M, et al. (2016) Methane Emissions from Conventional and Unconventional Natural Gas Production Sites in the Marcellus Shale Basin. Environ Sci Technol 50(4):2099–2107. 71. Barkley ZR, et al. (2017) Quantifying methane emissions from natural gas production in north-eastern Pennsylvania. Atmos Chem Phys 17(22):13941–13966. 72. Rose A, Cravotta C (1998) Coal Mine Drainage Prediction and Pollution Preventiona in Pennsylvania Available at: https://www.legis.state.pa.us/WU01/LI/LI/US/PDF/1921/0/0322..PDF. 73. Akcil A, Koldas S (2006) Acid Mine Drainage (AMD): causes, treatment and case studies. J Clean Prod 14(12–13 SPEC. ISS.):1139–1145. 74. Agency USEP (2017) Abandoned Coal Mine Methane Opportunities Database. (July). 75. 2002 Pennsylvania water quality assessment 305(b) report: Harrisburg, PA, Pennsylvania Department of Environmental Protection doi:3800-BK-DEP2530. 76. Cote M, Collings R, Pilcher RC, Talkington C, Franklin P (2004) Methane Emissions from Abandoned Coal Mines in the United States: Emissino Inventory Methodology and 1990 - 2002 Emissions Estimates. (April):90. 77. Stra D, et al. (2008) Methane-Producing Microbial Community in a Coal Bed of the Illinois Basin ᰔ. 74(8):2424–2432. 78. Coal Mine Methane in Colorado Market Research Report (2016) (March). 79. Heilweil VM, et al. (2013) A stream-based methane monitoring approach for evaluating groundwater impacts associated with unconventional gas development. Groundwater 51(4):511–524. 80. Stanley E, et al. (2016) The ecology of methane in streams and rivers: patterns, controls, and global significance. 86(2):146–171. 81. Bartlett KB, Crill PM, Bonassi JA, Richey JE, Harriss RC (1990) Methane flux

from the amazon river floodplain emissions during rising water. 95(D10):773–788. 82. Jones JB, Mulholland PJ (1998) Methane input and evasion in a hardwood forest stream: Effects of subsurface flow from shallow and deep flowpaths. Limnol Oceanogr 43(6):1243–1250. 83. Angelis M de A de, Lilley MDM (1987) Methane in surface waters of Oregon estuaries and rivers. Limnol Oceanogr 32(3):716–722. 84. Révész KM, Breen KJ, Baldassare AJ, Burruss RC (2012) Carbon and hydrogen isotopic evidence for the origin of combustible gases in water-supply wells in north-central Pennsylvania. Appl Geochemistry 27(1):361–375. 85. Edmunds WM, Shand P, Hart P, Ward RS (2003) The natural (baseline) quality of groundwater: A UK pilot study. Sci Total Environ 310(1–3):25–35. 86. Burrows JE, Peters SC, Cravotta CA (2015) Temporal geochemical variations in above- and below-drainage coal mine discharge. Appl Geochemistry 62:84–95. 87. Grieve PL (2014) Measuring concentrations of dissolved natural gas in three streams in Pennsylvania to estimate methane fluxes from the subsurface. Dissertation (The Pennsylvania State University). 88. Hammond PA (2016) The relationship between methane migration and shale-gas well operations near Dimock , Pennsylvania , USA. 2012:503–519. 89. Castelle AJ, Johnson AW, Conolly C (1994) Wetland and Stream Buffer Size Requirements—A Review. J Environ Qual 23(5):878. 90. Bogner JE, Burton EA (1997) Kinetics of Methane Oxidation in a Landfill Cover Soil: Temporal Variations, a Whole-Landfill Oxidation Experiment, and Modeling of Net CH 4 Emissions. 31(9):2504–2514. 91. Iii CAC (2007) Dissolved metals and associated constituents in abandoned coal- mine discharges , Pennsylvania , USA . Part 2 : Geochemical controls on constituent concentrations. Appl Geochemistry 23(2):203–226. 92. Vesper DJ, Moore JE, Adams JP (2016) Inorganic carbon dynamics and CO2 flux associated with coal-mine drainage sites in Blythedale PA and Lambert WV, USA. Environ Earth Sci 75(4):1–14. 93. Schimmelmann A, et al. (2018) Natural geological seepage of hydrocarbon gas in the Appalachian Basin and Midwest USA in relation to shale tectonic fracturing and past industrial hydrocarbon production. Sci Total Environ 644:982–993. 94. Krause E, Pokryszka Z (2013) Investigations on Methane Emission from Flooded Workings of Closed Coal Mines. J Sustain Min 12(2):40–45. 95. Hedin RS, Stafford SL, Weaver TJ (2005) Acid Mine Drainage Flowing from Abandoned Gas Wells. 104–106. 96. Jones V, Drozd R (1982) Prediction of Oil or Gas Potential by Near-Surface Geochemistry. Am Assoc Pet Geol Bull 67(6):932–952. 97. USGS Produced Waters. Available at: https://energy.usgs.gov/EnvironmentalAspects/EnvironmentalAspectsofEnergyPro ductionandUse/ProducedWaters.aspx#3822349-data [Accessed January 2, 2018]. 98. Edwin Exploration and Development Well Information Network Available at: https://edwin.onbaseonline.com/1500AppNet/Login.aspx [Accessed July 27, 2018].

99. Rivard C, Bordeleau G, Lavoie D, Lefebvre R, Malet X (2018) Can groundwater sampling techniques used in monitoring wells influence methane concentrations and isotopes? Environ Monit Assess 190(4). doi:10.1007/s10661-018-6532-7. 100. Jackson RE, Heagle DJ (2016) Sampling domestic/farm wells for baseline groundwater quality and fugitive gas. Hydrogeol J 24(2):269–272. 101. Smith B, Becker M, Siegel D (2016) Temporal variability of methane in domestic groundwater wells, northeastern Pennsylvania. Environ Geosci 23(1):49–80. 102. Isotech Laboratories (2011) Collection of Ground Water Samples from Domestic and Municipal Water Wells for Dissolved Gas Analysis Using IsoFlasks. (877). Available at: http://www.isotechlabs.com/products/isoflask/ISOFLASK_SAMPLING.pdf. 103. Kampbell DH, Vandegrift S a (1998) Analysis of dissolved methane, ethane, and ethylene in ground water by a standard gas chromatographic technique. J Chromatogr Sci 36(5):253–256. 104. Weiss RF (1968) Piggyback sampler for dissolved gas studies on sealed water samples. Deep Res Oceanogr Abstr 15(6):695–699. 105. Castro MC, Warrier RB, Hall CM, Lohmann KC (2012) A late Pleistocene-Mid- Holocene noble gas and stable isotope climate and subglacial record in southern Michigan. Geophys Res Lett 39(19):1–6. 106. PA Shale Viewer Available at: https://www.fractracker.org/map/us/pennsylvania/pa-shale-viewer/ [Accessed February 2, 2018]. 107. PA Oil and Gas Mapping Available at: http://www.depgis.state.pa.us/PaOilAndGasMapping/OilGasWellsStrayGasMap.ht ml? [Accessed February 2, 2018]. 108. Wells R, Bucek M (1980) Geology and mineral resources of the Montoursville North and Huntersville quadrangles, Lycoming County, Pennsylvania. (Pennsylvania Geological Survey Atlas). 109. Bedrock Geology of Pennsylvania. 1:250000 scale Available at: http://www.dcnr.pa.gov/Geology/PublicationsAnddata/Pages/default.aspx [Accessed July 1, 2017]. 110. Li H, Carlson KH (2014) Distribution and Origin of Groundwater Methane in the Wattenberg Oil and Gas Field of Northern Colorado. Environ Sci Technol 48(3):1484–1491. 111. Molofsky LJ, Connor JA, Wylie AS, Wagner T, Farhat SK (2013) Evaluation of Methane Sources in Groundwater in Northeastern Pennsylvania. GroundWater 51(3):333–349. 112. Feng G, Wu L, Letey J (2002) Evaluating Aeration Criteria By Simultaneous Measurement of Oxygen Diffusion Rate And Soil-Water Regime. Soil Sci 167(8):495–503. 113. Campbell KM, Malasarn D, Saltikov CW, Newman DK, Hering JG (2006) Simultaneous microbial reduction of iron(III) and arsenic(V) in suspensions of hydrous ferric oxide. Environ Sci Technol 40(19):5950–5955. 114. Meliker J, Slotnick M, AvRuskin G, Schottenfeld D, Jacque (2010) Lifetime

exposure to arsenic in drinking water and bladder cancer: a population-based case– control study in Michigan, USA. Natl Institues Heath 21(5):745–757. 115. Harkness JS, et al. (2015) Iodide, bromide, and ammonium in hydraulic fracturing and oil and gas wastewaters: Environmental implications. Environ Sci Technol 49(3):1955–1963. 116. Coleman D, Risatti B, Schoell M (1981) Fractionation of carbon and hydrogen isotopes by methane-oxidizing bacteria. Geochemica 45:1033–1037.

Appendix A

Supplemental Information for Sugar Run Case Study

SI History of Gas Development in Sugar Run area Additional Background. Summarizing the entire history of gas well development in the

Sugar Run area is beyond the scope of this contribution. We therefore emphasize the gas well closest to our study area (Table S1). This condensed summary derives from information online (27, 33, 98). Well API #081-20292 was hydraulically fractured between

June 28th, 2011 and July 1st, 2011. On January 9th, 2012 the PA DEP was notified of discolored water in a private water well nearby, and since that complaint, the PA DEP determined that at least five private water supplies were impacted (27). During testing in

January and February 2012, a shut-in pressure of 325 psi was measured between casing strings in the cemented annulus of 081-20292. Since April 13th, 2015, PA DEP investigated complaints of stray gas as far as 9,850 feet (3,000 meters) from the gas well and 5,200 feet (1,600 meters) from the originally affected water supplies. In June, 2015 PA

DEP levied a multi-million dollar fine against the drilling company but it was rescinded in

May, 2015.

SI Materials and Methods

Water Studies. Field measurements. In general in the Sugar Run area, stream flows are highest in late fall and spring after snow melt, and lowest in the summer during the hottest and driest months when evapotranspiration is highest. Multiple measurements of stream

base flow were completed between SR 1 and SR 2 on Sugar Run in 5/21/2013 to

11/12/2013 (24). Discharge measured at SR 1 varied from 0.105 m3/s in May to 0.037 m3/s in June. During this study, sites in Sugar Run were sampled intermittently for a year and a half and at increased frequency during summer (e.g., Fig. S1). When possible, temperature, pressure, dissolved oxygen (DO), specific conductivity (SPC), pH, and oxidation and reduction potential (ORP) were measured in the field using a YSI

Professional Plus meter. The YSI was calibrated before each trip using pH 4 and 7 buffers,

Zobel’s solution, and atmospheric oxygen. Hydrogen sulfide was measured using a Hach

Hydrogen Sulfide Test Kit (Model HS-WR) in accordance with Hach protocol for a few samples.

For dissolved methane and ethane in surface waters, samples were collected using

VWR polycarbonate bottles. We observed these bottles showed a memory effect for samples with high concentrations (>5 mg/L); therefore, VWR bottles containing dissolved methane >0.5 mg/L were discarded after use. The stream was sampled in the middle and approached from downstream to prevent contamination from disturbing the riverbed.

Bottles were opened underwater and rinsed three times before filling and capping underwater (when water depth made this possible). If bubbles were present, the bottle was emptied and refilled. Samples for methane and ethane analyses were stored in coolers, returned to the laboratory, and prepared the night of their return for analysis (see below).

Biocides were not used in any sample of groundwater or surface water collected using

VWR bottles, based on tests on surface water and groundwater that showed that biocide had no effect on methane concentrations in waters in capped bottles without head space as

long as the time between sampling and analysis was kept less than 5 days and, for seeps,

ORP was relatively low. Hydrocarbon concentrations were always measured within two days of collection, and most within one day.

Over about one year, groundwater samples were collected by the authors from four homeowner wells within 1000 meters of the seep locations in Sugar Run. In every case, wells were purged for at least 15 minutes, making sure that pH and water temperature were steady for at least 5 minutes before sampling. Several authors have documented that the reproducibility of high-methane concentrations in effervescing groundwater samples is low, and much of this lack of reproducibility may be related to sampling techniques (37,

99–101). Therefore, several sampling methods were tried for comparison and reproducibility among methods was indeed sometimes observed to be low (data from two sampling methods summarized in Dataset S1). Most samples were taken with the “inverted bottle” method (slightly modified from published procedures in that we used VWR polycarbonate bottles without biocide as described above (37)). In addition, a few samples were collected from two homeowner water wells using an Isoflask® in accordance with

Isotech protocol (102). Although this collection method is now suggested to be the best method for effervescing samples (37), it was only introduced relatively recently and therefore is not as useful in comparisons with samples collected years ago. In the early years, the inverted bottle method was commonly used (e.g., (7)). We therefore emphasize samples taken with inverted bottles in discussion in main text. Hydrogen sulfide was measured using a Hach Hydrogen Sulfide Test Kit (Model HS-WR) in accordance with

Hach protocol for a few samples.

Occasionally at surface and at all groundwater sites, additional samples of water were taken and then filtered with a 0.45-micron pore-size filter to be analyzed for anions and cations. Cation samples were acidified with nitric acid in the field. Samples were collected in the same way at select sites (HO4, Seep 1.5, Seep 1.6) for strontium isotope analysis. All samples were returned to the laboratory and refrigerated within 8 hours of collection.

Atmospheric methane was analyzed in the field on occasion using a Bascom-Turner

Gas Rover (Fig. S3). Measurements, logged every 1-3 seconds, included location

(latitude/longitude) and methane concentration in ppm (parts per million by volume). This device was calibrated using a 2.5% methane gas standard from Bascom-Turner and their internal autocalibration program.

Laboratory analyses. Cations were measured using a Perkin-Elmer Optima 5300 inductively coupled plasma atomic emission spectrophotometer (ICP-AES). Select samples were also measured using a Thermo Fisher Scientific X Series 2 Inductively

Coupled Plasma Mass Spectrometer (ICP-MS) with Collision Cell Technology (Dataset

S1). When the same sample was analyzed by both instruments, ICP-MS results were compared to the ICP-AES results for Fe, Al, Ba, Mn, and Sr. If concentrations were less than 0.1 mg/L and discrepant by more than 10%, ICP-MS values were reported because of their higher accuracy for low concentrations (Dataset S1). Arsenic and uranium were only measured using ICP-MS. Anion samples were analyzed using a Dionex ICS 2500 ion chromatograph (IC). Strontium isotopes (87Sr/86Sr) were measured on a Thermo Scientific

Triton Plus Series Multicollector Thermal Ionization Mass Spectrometer (TIMS). Prior to

analysis, strontium was purified from water samples using the Elemental Scientific prepFAST-MC automated chromatography system. The values of 87Sr/86Sr for standards run at the time of analysis were as follows: NIST SRM987 = 0.710257, BCR-1 =

0.705027, and IAPSO = 0.709194.

Hydrocarbon samples were analyzed using previously published methods at

Pennsylvania State University unless otherwise noted (2, 19). First, a helium headspace was introduced to each bottle (103). These bottles were then placed on a shaker for at least

12 hours to allow dissolved gases to equilibrate with headspace. Hydrocarbons were then analyzed using a HP 5890 Series II Gas Chromatograph with a flame ionization detector

13 and custom vacuum inlet system. δ C-CH4 values were analyzed on some samples within seven days of concentration measurements according to methods outlined in previous research (2).

Noble gas sampling and analytical methods. For noble gas analyses, five water samples were collected from two domestic water wells (two duplicates from each well) in the Sugar

Run area and one natural spring in Salt Spring State Park in Bradford County, PA (July

2017). These were analyzed for concentrations and isotopic ratios of the complete suite of stable noble gases (He, Ne, Ar, Kr, and Xe). Two gas samples were also collected, one from the salt spring site, the other from a seep adjacent to the Sugar Run stream (Seep 1.55) in July 2017. These two sites were chosen for gas analysis based on their high rate of

bubbling. Both gas samples were analyzed for noble gas volume fractions and isotopic ratios.

For noble gas samples, groundwater was flushed through standard refrigeration grade 3/8” copper (Cu) tubes for approximately 10 min. Once temperature, pH and specific conductivity reached constant values in the outflow, Cu tubes were sealed by steel pinch- off clamps (29, 104). Additional details on groundwater sampling for noble gas analysis can be found elsewhere (30, 105). In the salt spring, a peristaltic pump was used to pump the water through the Cu tube.

To collect gas samples from the salt spring and the stream seep, a customized funnel was assembled with plastic tubes and a copper tube as shown in Figure S15 and then pre- filled with water prior to moving the funnel over the gas outlet. The funnel was kept under water continuously during sampling to allow the gas to flow through the Cu tube and to displace the water. Between 15-30 minutes were required for the Cu tube to be entirely filled with gas. The plastic tube was checked for residual water prior to closing the clamps at both ends of the copper tube.

Noble gas measurements of all water and gas samples were carried out in the Noble

Gas Laboratory at the University of Michigan. He and Ne were analyzed in a Thermo

Scientific Helix SFT mass spectrometer while Ar, Kr, and Xe were sequentially allowed into an ARGUS VI mass spectrometer using a computer-controlled double-head cryo- separator. Extraction, purification, and analysis procedures are described elsewhere (18,

29).

Methods for Development of Maps. Groundwater data and violations map. Several sets of groundwater data were compiled, mapped, or compared to new analyses. Most of these data were considered to represent background groundwater in the area. The largest dataset is referred to here as “pre-drill data” because it includes analytical data from commercial laboratories of samples collected by consultants for oil and gas companies (5). The companies sample groundwater within a certain distance of planned well drilling and give the data to the PA DEP to safeguard against future liability. Discussions of strengths and limitations of these types of datasets are discussed below and have been summarized elsewhere (5, 7, 20). These data were shared with the authors by the PA DEP and are published in the Shale Network database (34). In addition to this pre-drill dataset, data were summarized from other published groundwater analyses from the Lycoming county area

(35) and this entire dataset is referred to throughout as the Lycoming groundwater dataset.

In addition, available data from within the Sugar Run study area from the PA DEP were recovered using the PA DEP eMap tool (31). Another set of data – for sites in PA referred to throughout as “presumably contaminated” -- were also summarized for waters that were presumed by various government agencies to have been contaminated by oil and gas development activity (33, 41). For example, data from the US Environmental

Protection Agency (US EPA) were summarized here for groundwater from water wells that were presumed to be affected by nearby shale-gas development activity (41) based on comparisons of pre-drill to post-drill hydrocarbon concentrations, hydrocarbon ratios, isotopic compositions, metal concentrations, and water type. One of these presumably impacted water wells was sampled near the alleged gas migration incident in Dimock,

Pennsylvania. Six other presumably impacted water wells were sampled near an alleged gas migration event along Paradise Road in Bradford County that has also been described elsewhere (17). The final three presumably impacted water wells (GW 06, GW 01, GW

02) were reported for two other alleged gas migration incidents (41). The locations for these presumably contaminated sites were not revealed in the EPA report but locations were verified from Figures 3, 4, 33, 34, 35, and 37 of the report (41).

For these data (pre-drill or presumably contaminated samples), the commercial or government laboratories that were used for analyses are not always mentioned; however, commercial laboratories that were named include ALS Environmental, Environmental

Service Laboratories, Benchmark Analytics, ESL, Fairway, Geochemical Testing,

Groundwater and Environmental Services, Inc, Hess and Fisher, Lancaster, Seewald

Laboratories, and Test America. The pre-drill data were originally provided by commercial analytical laboratories as paper copies or pdfs, and were transcribed into spreadsheets either by U.S. Geological Survey workers or Pennsylvania State University workers.

Data used to create Figure 1B were gathered on 8/15/2017 with the Pennsylvania

Shale Viewer tool (106) and Oil and Gas Mapping Tool (107). Wells with violations were first identified using the Shale Viewer Tool. If a casing-related violation was identified, the violation was confirmed using the PA DEP Oil and Gas Mapping Tool. No discrepancies between the two tools were observed. Violations binned as casing- or cementing-related were based on guidelines outlined in previous research (4) (see also Table S1). A few additional casing-related violations were included that were not identified in the previous paper (4): 78.81(a) (Casing and Cementing - Operator conducted casing and cementing

activities that failed to prevent migration of gas or other fluids into sources of fresh groundwater); 78.81(a)3 (Casing and Cementing - Operator conducted casing and cementing activities that failed to prevent pollution or diminution of fresh groundwater);

OGA 3217(B) (Failure to prevent migration of gas or fluids into sources of fresh water causing pollution or diminution, failure to properly case and cement well through a fresh water-bearing strata in regulated manner or depth).

Construction of structural geology model. A three-dimensional model was constructed for the eastern end of the Nittany Anticlinorium based on well and map data (Fig. S5). Well data consisted of formation logs from 22 gas wells in the area (98). These mostly include the Marcellus and units stratigraphically above it, although a few extend deeper (47, 108,

109). Data were digitized and analyzed with Midland Valley’s MoveTM software. Cross sections derived from two datasets (47, 108) were also digitized and added to the model.

These sections were extended and additional sections were created based on map and well data to represent the three-dimensional structure of the anticline more completely. The cross sections and the well data were then used to create three-dimensional surfaces by ordinary kriging.

A block diagram (Fig. S5) and a cross section (Fig. 1D) from the closest gas well

081-20292 to Sugar Run were created from the surfaces. The block diagram was slightly edited manually in areas where the kriging was poorly constrained by data in order to remove geologically unrealistic artifacts or to smooth horizons. The surface representing the top of the Marcellus Formation was contoured to produce a structural contour map of this horizon (Fig. S13).

SI Results and Discussion

Field Measurements. The pH of seeps (5.9 to 7.0) was lower than observed in Sugar Run

(7.1 to 8.2) which was generally lower than observed in nearby homeowner water wells

(7.6 to 8.9). In contrast, oxidation reduction potential (ORP) decreased from the stream

(177 to 241 mV) to the seeps (-90 to 169 mV) to the homeowners’ water wells (-277 to -

146 mV). Late in our investigation we discovered another seep that had the highest pH

(9.1) of any samples and had dark black sediment. This bubbling seep (SRS 1.5, Fig. 1) was in a channel of a small intermittent tributary where we had limited permission for sampling.

Comparison of Groundwater Hydrocarbon Samples. As discussed in SI Materials and

Methods, the groundwater samples we report in Figure 2 were limited to those collected with the inverted bottle technique so as to compare concentrations over a longer period of time for samples collected in the same manner.

Three (HO2, HO3, HO4) out of the four homeowner water wells that we sampled in this study were observed to effervesce. As discussed in SI Materials and Methods, large variability was observed in these high-methane samples for methane concentrations

(concentrations were generally >20 mg/L) when collected with different methods (two are summarized in Dataset S1). For example, samples collected from HO2 and HO4 using the

Isoflask® method contained CH4 concentrations of 28.0 mg/L and 49.0 mg/L, respectively.

In comparison, methane concentrations of HO2 and HO4 water samples collected at the same time using the inverted bottle method were analyzed to contain 26.4 mg/L and 29.1 mg/L respectively (Dataset S1). Groundwater at both HO2 and HO4 were observed to effervesce.

Ethane concentrations for samples collected using the inverted bottle method (594

µg/L) and using the Isoflask® (1,200 µg/L) were also observed to differ for water well

HO4. Propane was

detected in one Isoflask sample (HO4) by Isotech laboratories (Dataset S1).

Noble Gas Data. Noble gas data overview. Noble gas concentrations (for water samples), volume fractions (for gas samples), and isotopic ratios (for both water and gas samples) are listed in Table S2-S4, respectively. Dissolved methane concentrations of water samples are also included in Table S2. Measured 3He/4He (R) ratios of water and gas samples are normalized to the corresponding atmospheric value (Ra = 1.384x10-6) and shown as R/Ra in Table S4. R/Ra values of all gas samples and the salt spring water sample are particularly low (0.0114 ± 0.0005 to 0.0165 ± 0.0006), pointing to the presence of a highly pristine crustal signature while well water samples are higher and fall within the range of typical crustal R/Ra ratios (~0.02-0.05) (39). These R/Ra values are much lower than the atmospheric value (R/Ra = 1), strongly suggesting the presence of a largely dominant contribution of crustal helium. This also shows that our sampling technique successfully preserves pristine noble gas signatures. Isotopic ratios of all other gases are atmospheric within a 2σ error with only a few exceptions (e.g., 40Ar/36Ar values of HO4). For example,

the isotopic ratios for Ne, Kr, and Xe are mostly consistent with atmospheric origin. These gases are incorporated into the subsurface by recharge water in equilibrium with the atmosphere (i.e., air-saturated water or ASW). Figure S16 shows 4He/36Ar, 22Ne/36Ar,

84Kr/36Ar, and 132Xe/36Ar ratios normalized to corresponding air values and plotted as F(i) where F denotes the ratio of isotope i to 36Ar in air. From Figure S16, it is apparent that

F(4He) values for all samples are far higher (>1 to 3 orders of magnitude) than both air and

ASW values at 10 C, pointing to an almost entire dominance of crustal He. A temperature of 10C for the ASW component was chosen to represent recent recharge conditions in northeastern PA.

F(22Ne) ratios of gas samples are higher than that of ASW while their F(84Kr) and

F(132Xe) ratios are lower than corresponding ASW values. However, F(22Ne), F(84Kr), and

F(132Xe) values of water samples generally display exactly the opposite pattern from the gas samples (except for F(22Ne) ratios of water samples from homeowner wells). Such patterns of noble gas elemental ratios were also observed in methane-rich groundwater within the Barnett Shale footprint and corresponding Barnett and Strawn shale gas (18, 29).

Because the solubility of noble gases increases with atomic weight, light noble gases (i.e.,

22Ne) will preferentially partition into the gas phase when groundwater degasses. A single- stage water degassing model can therefore explain the relative depletion of light 22Ne and relative enrichment of heavy 84Kr and 132Xe observed in salt spring water. The F(22Ne) values of well water samples mimic ASW values (Figure S16), suggesting limited noble gas fractionation. This might be due to a short contact time between the gas and liquid phases.

Source of helium and methane: shallower vs. deeper formations. As shown above, crustal

He largely dominates the total measured He. To evaluate whether crustal He is produced in-situ in the shallower formations (i.e., Upper Devonian strata such as the Lock Haven

Formation) (24) or has an external origin (i.e., deeper formations such as the Marcellus

Formation), 4He ages were calculated for all water samples assuming that all crustal 4He is produced within the Lock Haven Formation as follows:

A = P( iHe)´ r ´ L´((1-w) /w)cm3STPg-1 yr-1 i He r H 2O (1)

4 -13 -14 3 -1 -1 P( He) =1.207´10 [U]+2.867´10 [Th]cm STPgrock yr (2)

3 Here, ρr is the density of the rock in g/cm , ω is the porosity of the reservoir rock, Λ is the transfer efficiency of He from the rock matrix to the water (assumed to be 1), U and Th represent uranium and thorium concentrations (in mg/kg) respectively in the host rock. The rock density is assumed to be 2.5 g/cm3 while the porosity is set to 0.2. Average U and Th contents for the Lock Haven Formation were set to 2.6 and 9.7 mg/kg, respectively, following a recent publication (15).

Calculated 4He ages are listed in Table S2 and range from 0.37 Ma to 0.85 Ma.

These ages are much older than the previously reported, mostly modern ages (19). These

4He ages calculated here are much larger than reasonable residence times for shallow groundwater, indicating that in-situ production is not responsible for most of the crustal

4He present in this groundwater. An external source of 4He must be introduced to account for the majority of measured 4He concentrations in these groundwater samples. This external source might also bring thermogenic methane into the shallow aquifer as

thermogenic methane contents usually correlate positively with crustal He content in groundwater (15, 29).

The presence of crustal 40Ar (40Ar*) was detected in water samples collected from a homeowner well (HO4) and in a gas sample from Seep 1.55 (Table S4). Calculated

4He/40Ar* ratios of sample HO4 range from 7.49 to 9.01, values within the previously reported range (6.2-13.7) of natural gas samples from the Marcellus Formation (16) but over an order of magnitude lower than reported natural gas sample values (214.6-285.4) from the Upper Devonian formations (i.e., Canadaway Formation) (16). This leads to the inference that the Marcellus Formation is likely to be the external source of both crustal noble gas and thermogenic methane in the samples of HO4. Noble gas fractionation due to water degassing, which might alter the original 4He/40Ar*, is not considered possible here for HO4 because the fractionation is limited for water samples collected from homeowner wells (Figure S16). However, the gas sample from Seep 1.55 displays 4He/40Ar* values of

82.57-98.29, which might be consistent with some mixing of Marcellus gas and Upper

Devonian gas. Alternatively, degassing of the water could preferentially have released some of the light 4He into the gas phase but have retained the heavier 40Ar in the water phase. This mechanism is a possible explanation for the increased 4He/40Ar* observed in

Seep 1.55.

Mechanism of methane migration: free gas vs. dissolved phase. Natural gas can be transported as a free gas phase or as a dissolved species in groundwater and such transport might happen naturally, perhaps along with brine salts, or as a result of hydrocarbon production activities, e.g., drilling or hydraulic fracturing (16–18, 110, 111).

100

To further determine how the external Marcellus gas migrates into the shallow

4 20 36 aquifer in the Sugar Run area, we plotted He/CH4 as a function of Ne/ Ar for well water

4 20 36 samples (Figure S9). Predicted He/CH4 and Ne/ Ar ratios were calculated for four scenarios (15): (1) slow upward advection of water containing brine salts and dissolved methane; (2) diffusion of gas from depth through aqueous solution; (3) combined upward advection of two phases (i.e., free gas phase and brine salt-containing water phase); and

(4) fast upward advection of free phase gas with minor mixing of microbial gas in the shallow aquifer. Note, we assumed that microbial gas contributes methane but not noble gases. For all scenarios, we assumed the starting point was the composition of Marcellus production gas (green square in Figure S9). Since the chloride concentrations in water collected from HO2 and HO4 are very low (< 10 mg/L), we can exclude scenario (1). From

4 20 36 Figure S9, He/CH4 and Ne/ Ar ratios of water samples HO2 and HO4 are distinct from predictions for scenarios (2) and (3), and thus argue against gas migration coupled with aqueous solution as a two-phase system. In contrast, short-timescale migration of natural gas in a free gas phase (scenario (4), e.g., gas leaking from a faulty gas well that then migrates along faults and fractures) could maintain the original noble gas and hydrocarbon

4 20 36 ratios with minimal fractionation. The difference in He/CH4 and Ne/ Ar values between

Marcellus production gas and water samples is attributed either to variability of noble gas composition of Marcellus production gas or to the input of a small amount of microbial methane (as shown and labelled along line (4)).

In summary, while other interpretations are possible – i.e., multiple sources of gas and multiple migration and oxidation steps -- these noble gas data are consistent with the

101

interpretation that thermogenic methane detected in water wells HO2 and HO4 within the

Sugar Run area is from the Marcellus Formation. These Marcellus gases most likely migrate into the shallow aquifer in a free gas phase along faults, fractures, and porous formations.

Seasonal Changes. Seasonal changes in methane concentrations for the most methane- rich seep, Seep 1.6 (Fig. S7), were not simply caused by dilution: methane was typically most concentrated during the winter and early spring when water levels were at their highest, and smallest during the summer months when water levels were lowest. This observation argues against methane migration as a dissolved solute; seasonally high methane concentrations in the wetter months are therefore attributed largely to free-phase methane moving upward, sometimes entrained in groundwaters as they move within the shallow surface. Free-phase gas migration may also be consistent with the observations of methane emitting from fractures in outcrops near the study site (Fig. 1C). This explanation for the mechanism of migration (free-phase gas) is also consistent with noble gas data discussed above and with published arguments for the Sugar Run site (19).

The similarity in seasonal changes in methane, iron, and arsenic concentrations for

Seep 1.6 are consistent with the explanation that methane concentrations drive iron and arsenic variability (Fig. S7). Specifically, we argue that iron is observed at higher concentrations at Seep 1.6 when subsurface bacteria couple methane oxidation to iron reduction. (Ferric iron minerals are much less soluble than ferrous iron minerals at circumneutral pH). It is well known that oxygen does not diffuse as easily through water-

102

filled as compared to air-filled pores (112). Consequently, the enhanced water saturation and lower subsurface oxygen concentrations during the wet season at Sugar Run likely accelerate anaerobic oxidation of methane coupled to reduction of ferric iron oxide minerals, releasing ferrous iron to solution.

In the summer when water saturation of porous subsurface material is lower, less methane likely reaches the surface because it is not entrained as much in upflowing groundwater. In addition, oxygen is more abundant in the less water-saturated subsurface and can be used by bacteria as the electron acceptor for aerobic methane oxidation instead of ferric iron. This draws down oxygen but does not release ferrous iron to solution.

According to this explanation, low methane concentrations in the summer are caused both by low upflow rates of groundwater with entrained free-phase methane in the shallow subsurface and by high rates of methane oxidation by oxygen, the thermodynamically preferred oxidant.

The release of arsenic to solution seasonally documented in Figure S7 is consistent with anaerobic methane oxidation that reduces ferric iron oxides to release iron and adsorbed arsenic into solution (113). This is especially notable because arsenic is a known carcinogen and thus could be of concern in drinking water systems (114). In Seep 1.6, concentrations above the EPA drinking water standard were never measured (Dataset S1).

Using Salt Tracers to Identify Groundwater That Warrants Further Investigation.

Chloride concentrations. In this section we explore whether concentrations of Cl, Na, and

Ca could be useful tools in identifying water supplies that contain methane derived from

103

non-natural sources. These tools may be specific to the northeastern USA where natural dissolved thermogenic methane is often detected with fluids that contain salts from

Appalachian Basin brine (ABB). Specifically, in the Appalachian Basin, naturally high methane concentrations correlate with high chloride (11, 14, 115). Some have even suggested that non-impacted groundwaters with naturally high chloride concentrations in

PA may document connections to deeper fracture networks that could be locations for easy gas migration if gas wells allow leakage nearby (10).

To investigate such correlations, we plotted the Lycoming groundwater dataset

(34), assumed to represent natural background, with groundwaters from the Sugar Run area and from the presumably impacted water wells in Fig. S12 (31–33, 41) to see if any waters are chemically distinct. As shown in Figure S12A, almost all high-methane samples from the Lycoming groundwater dataset contained high chloride (>30 mg/L) whereas almost all the presumably impacted samples -- and Sugar Run samples -- contained very low chloride

(<30 mg/L).

Figure S12A documents that samples near Sugar Run are anomalous in that they are high in methane and low in chloride. Such samples could be explained by at least two scenarios. First, methane could be produced biogenically in low chloride waters: such a mechanism can explain high methane in swamps, for example. Second, methane could migrate in a free gas phase independent of ABB-containing waters (19). This second scenario is more likely for Sugar Run waters and the presumably affected waters since many of those waters were shown to contain thermogenic methane, as determined from isotopic analysis and C1/C2 ratios (41). In addition, migration of free gas for Sugar Run

104

was consistent with noble gas analysis (Fig. S9) and other considerations (19). Wetlands were also not observed in the bedrock-lined channel of Sugar Run.

Na-rich versus Ca-rich waters. In the Appalachian Basin, higher methane concentrations have also been associated with Na-rich as opposed to Ca-rich waters (22). Ca-rich waters are believed to represent near-surface recharge or shallow groundwater (35, 46). In this light, we hypothesized that Ca-rich high-methane waters are less likely to contain thermogenic methane from natural sources.

We tested datasets to see if such inferences would be useful in highlighting anomalous methane concentrations in groundwaters, i.e., methane that warrants further investigation because it might be related to anthropogenic activities. On a plot showing the ratio of Ca to Na concentrations for the Lycoming data and the presumably impacted waters (Figure

S12B), fewer outliers are detected in the blue quadrant compared to outliers on Figure

S12A. Most of the samples, regardless of the source of data, have Ca/Na <0.52. However, some samples from Sugar Run and the presumably impacted water wells have Ca/Na

>0.52. This shows that relatively Ca-rich high-methane water samples are another water type that warrant further investigation to detect a recent invasion of methane gas.

105

Appendix A Figures for Chapter 2

Figure A1. Photograph of Sugar Run showing Seep 1.5 (left) and the stream (right).

106

Figure A2. A) Map of the Sugar Run watershed (colored blue), sample locations (triangles), seepage locations (blue circles), and outcrops (orange circles). B) Map of Sugar Run sampling locations with seeps and the average stream methane concentrations at each site. Active bubbling was observed in the stream and seeps along this stretch. The dashed line and red and green symbols in (A) are described in caption for Figure 1.

107

Figure A3. (A) Map showing measurements of methane concentrations in air measured on eight different days in the study region. Distinct colors represent different sampling days while the sizes of the symbols indicate concentration. (B) Methane concentration map for air in the highest-concentration region. Over 24,000 measurements were collected over 8 days. Measurements were completed by walking around the study region and holding a surface bell probe within one meter of the ground surface. At outcrops, the surface bell probe was placed directly on fractures.

108

Figure A4. Methane concentration in stream water plotted versus distance downstream along Sugar Run in Lycoming County, Pennsylvania. The zero position was defined as the most upstream stream site (SR 8) that was sampled. Triangles are samples collected and analyzed by Heilweil et al. (24) and circles represent samples from this study. Largest methane concentrations in the stream consistently were measured at sites SR 1.5 and SR 1.55 (near two methane-rich seeps). The horizontal section of well 081-20292 crosses underneath Sugar Run at the line, and locations downstream of this location are structurally up dip.

109

Figure A5. Block diagram of the Nittany Anticlinorium, showing locations of Sugar Run and gas well 081-20292. The surface impression of the Nittany Anticlinorium can be observed to the west in the figure.

110

Figure A6. Isotopic values for methane from samples collected at sites in Figure S2 plotted 13 versus 1/[CH4] where [CH4] refers to CH4 concentrations in µg/L. In the blue region, δ C increases with decreasing methane concentration, as expected for fractionation during methane oxidation (9). A few samples more negative than -28.3‰ plot outside the blue area and approach an inferred biogenic endmember (9). Following previous literature (24), samples with the largest methane concentrations were used to estimate the δ13C signature of the thermogenic endmember (-28.3‰; red solid line). The initial concentration of thermogenic methane in water before oxidation likely varies temporally and spatially. The [CH4] concentration of 1000 µg/L (1/[CH4]=0.001) is chosen to calculate expected Rayleigh fractionation (blue dashed lines). Two scenarios are considered for two values of fractionation factors () of 1.013 and 1.025, derived from previous work (116).

111

Figure A7. Concentrations of CH4, Fe, and As in the most methane-rich seep (1.6) plotted versus time. For simplicity, concentrations were normalized by dividing by the maximum values: Fe (17.8 mg/L), As (0.0058 mg/L), CH4 (8.6 mg/L).

112

Figure A8. R/Ra and He/Ne ratios of all collected noble gas samples including water from the salt spring at Salt Spring State Park, Pennsylvania, and homeowner wells in Sugar Run. Dashed curves represent mixing between noble gases from ASW at 10 C and radiogenic noble gases (including crustal and mantle sources). Three scenarios with varying contributions of mantle He (i.e., 0%, 25%, and 50% by mass) are indicated. All water and gas samples reported for the Sugar Run area are located on the curve representing mixing between ASW and pure crustal components. This precludes the presence of significant mantle noble gas in these samples.

113

4 20 36 Figure A9. He/CH4 and Ne/ Ar ratios of homeowner well water samples (red triangles) and salt spring water (blue circle). Short vertical and horizontal lines within sample marks 4 20 36 represent corresponding error bars for He/CH4 and Ne/ Ar, respectively. Predicted values are also plotted for four scenarios: (1) slow upward advection of brine containing dissolved methane (black line with upward arrow); (2) diffusion of gas from depth in aqueous solution (green dashed curve); (3) upward advection of two phases (i.e., free gas phase and brine phase) (i.e., free gas phase and brine phase) from depth (red dashed curves); and (4) fast upward advection of free phase gas with minor mixing of microbial gas in the shallow aquifer (vertical line with downward arrow). Note, we assume that microbial gas contributes methane but not noble gases. In all of these scenarios, we assumed the starting point of noble gas fractionation was the composition of Marcellus 4 Formation production gas (green rectangle). The pink rectangle represents He/CH4 and 20Ne/36Ar of natural gas samples from Upper Devonian formations (15). 20Ne/36Ar ratios of homeowner wells HO2 and HO4 are consistent with scenario (4) (fast upward flow of Marcellus gas mixed with biogenic gas), and is not consistent with scenarios (1), (2), nor (3). We hypothesize that relatively quick migration of Marcellus gas along faults in a free gas phase best explains why the homeowner well waters preserve the original 20Ne/36Ar signature.

114

Figure A10. Plot of ratios of Cl/Br concentrations (both as mg/L) versus chloride for samples collected from Seep 1.5, Seep 1.6, stream sample SR 1, and homeowner water wells HO1, HO2, HO3, and HO4. Most waters plot within or very near dilute groundwater except for Seep 1.6. The plot, adapted from previous work, shows generalized regions of different water types (11). The circle labeled “ABB Brine” represents chemistry of directly sampled ABB.

115

Figure A11. 87Sr/86Sr vs Sr/Ca molar ratio in samples taken near Sugar Run (stream samples labelled SR, homeowner well waters labelled HO, groundwater under the stream sampled by piezometers labelled Piezo, and seeps) and brines collected from oil and gas wells (97). The plot shows the possibility that there are two different sources of brine salts that have impacted waters in the Sugar Run valley. The black box encompasses the variation in chemical fingerprint of brines originating from the Marcellus as defined by previous work (10). Samples collected at Seep 1.6, HO4, and SR 1.5 Piezo have the distinctively higher Sr/Ca and lower 87Sr/86Sr that are typical of Marcellus brines (10). In contrast, stream sites SR 1 and SR 1.5, and Seep 1.5 show chemical characteristics more consistent with radiogenic upper Devonian formations such as the Bradford Group and Venango Group. Some data values were reproduced here from previous works (19, 24).

116

Figure A12. A) Plot of chloride versus methane concentrations in groundwater samples as labelled (see text in SI and main body). Most high methane water in Lycoming county contains high concentrations of chloride. However, waters from almost all locations with presumable contamination, including Sugar Run, contain relatively low concentrations of chloride. B) Ca/Na (mass ratio) versus methane concentrations in groundwater samples as labelled. Most high methane samples contain Ca/Na < 0.52. However, a small subset of presumably contaminated samples contains Ca/Na > 0.52.

117

Figure A13. A contour map of depth of the top of the Marcellus Shale (dashed lines document where erosion has removed the shale and negative labels on contours denote depth below sea level in meters). The surface was calculated from cross sections, gas well reports, and topographic data using ordinary kriging. Coordinates are in UTM, Zone 18 North.

118

Figure A14. Normalized concentrations of methane, iron, and sulfate versus time for groundwater from homeowner well HO4. Concentrations are normalized to the largest concentration for each analyte: methane = 61.6 mg/L, sulfate = 18.7 mg/L, iron = 0.576 mg/L. Concentrations are only plotted where at least two of the three analytes were available on a given date (Table E6-E8).

119

Figure A15. The set-up for the collection of gas samples for noble gas analyses at bubbling seep sites.

120

Figure A16. F(4He), F(22Ne), F(84Kr), and F(132Xe) value of collected water and gas samples. F(4He), F(22Ne), F(84Kr), and F(132Xe) are measured 4He/36Ar, 22Ne/36Ar, 84Kr/36Ar, 132Xe/36Ar ratios normalized to corresponding air values. F values of ASW (air saturated water) at the temperature of 10 C (light blue line) and air (horizontal black line) are shown for comparison (40).

121

Appendix B

Additional Figures for Chapter 2 and Chapter 3

Figure B-1: Methane concentrations measured in water samples from one homeowner water well in the Sugar Run Valley, Lycoming PA. Methane increases inconsistently (roughly 10 mg/L per year) over a roughly 5-10 year period from pre-drill concentrations. Dates and methane concentrations are not provided to protect homeowner identity.

122

Figure B-2: Ethane concentration over time for one homeowner water well in the Sugar Run Valley, Lycoming PA. Ethane increased (roughly 0.5 mg/L per year) over a roughly 5-10 year period from pre-drill concentrations. Dates and ethane concentrations cannot be provided for privacy reasons.

123

Figure B-3: Propane concentration over time for one homeowner water well in the Sugar Run Valley, Lycoming PA. propane increased (roughly 0.06 mg/L per year) over a roughly 5-10 year period from pre-drill concentrations. Dates and propane concentrations cannot be provided for privacy reasons.

124

Figure B-4: Iron concentration over time for one homeowner water well in the Sugar Run Valley, Lycoming PA.

125

Figure B-5: Sulfate concentration over time for one homeowner water well in the Sugar Run Valley, Lycoming PA.

126

Figure B-6: Alkalinity over time for one homeowner water well in the Sugar Run Valley, Lycoming PA.

127

Figure B-7: Pictures of leaking abandoned wells (ANF Leaking Well 1 right and ANF leaking well 2 left) in the Allegheny National Forest. Locations: 41.5095, -78.7926 (right) and 41.6121, -78.7913 (left).

128

Figure B-8: Map of sample locations for leaking abandoned wells (yellow diamonds), gas leak drainage (red stars), and AMD discharges (red circles).

129

Figure B-9: Map of sample locations in northwestern Pennsylvania. Concentrations are site-aggregated means where more than one sample was analyzed. See text for how the categories were defined.

130

Figure B-10: Picture of the Morrison seepage (left) where methane and ethane rich waters flow out of a depression in the ground. White strings are assumed to be bacteria. A smell of hydrogen sulfide is present at this seepage. (Right) is a close up of orange “slimy” bacteria at Seep 1.5 near Sugar Run, Lycoming.

131

Figure B-11: Photo of the Honey Pot Outflow near Wilkes-Barre, PA where waters flow out of a mine opening.

132

Figure B-12: Picture of the Gladding AMD discharge near Pittsburg, PA (left) and where flow form this discharge tumbles off a waterfall (right) into Miller’s Run.

133

Figure B-13: Plots of methane concentration in stream waters vs distance to the nearest conventional oil or gas well spudded between different date ranges. See text for more information.

134

Figure B-14: Map of sample locations in southwestern Pennsylvania. Concentrations are site-aggregated means where more than one sample was analyzed. See text for how the categories were defined.

135

Figure B-15: Map of sample locations in northeastern Pennsylvania. Concentrations are site-aggregated means where more than one sample was analyzed. See text for how the categories were defined.

136

Appendix C

Meshoppen Creek Case Study

Using stream methane data collected during this study, previous work (Heilweil et al., 2013), and in collaboration with the Susquehanna River Basin Commission, an investigation was conducted into elevated methane in Meshoppen Creek - Susquehanna

County, PA. Stream methane samples reported originally by Heilweil et al. (2013) revealed some of the highest individual measurements of stream methane concentrations in this study (~70 µg/L – Meshoppen US 2 (Table E17)). Samples collected upstream and downstream of this location (Figure C-1) were also elevated in methane with 6 sites containing methane >4 µg/L and 4 sites >10 µg/L.

137

Figure C-1: Map of the Meshoppen Creek watershed in Susquehanna County, PA. Sample locations (maroon circles) are noted where methane concentrations are indicated by circle size. A circle is drawn around Dimock, the site of known hydrocarbon leakage from shale gas development (Hammond et al., 2016). The sample location Meshoppen US 2 is within the circled region. Small open circles represent unconventional gas wells as denoted by the legend.

Meshoppen Creek is of interest for multiple reasons. Most notably, site

Meshoppen US 2 is located nearby Dimock Pennsylvania. Dimock was the location of an early and controversial gas migration incident associated with shale gas development

(Hammond et al., 2016). This author has suggested that some gas wells in the region may have been improperly remediated and may still be leaking methane.

However, analysis of methane in Meshoppen Creek itself is ambiguous given that

Meshoppen Creek largely flows through wetlands and bogs (Figure C-2). Isotopic δ13C-

CH4 measured at multiple sites including Meshoppen US 2 is also ambiguous ranging

138

13 from methane that appears to have undergone oxidation (-13.7‰), δ C-CH4 values more consistent with biogenic gas (-58.2‰) and values in between these which could be

13 interpreted as thermogenic gas. Explanations for the dramatic shifts in δ C-CH4 values include seasonal mixing of biogenic and thermogenic gas and oxidation of methane by methanotrophic bacteria.

Figure C-2: Zoomed in map of Meshoppen Creek near Dimock, PA (left). Known wetland locations (plotted - left) validate physical observations (right) that the stream is extremely boggy. The photo on the right hand side was located near Meshoppen US 2, see table E17.

Regardless, observed methane concentrations in Meshoppen are some of the highest in streams sampled in the state. Similarly high methane concentrations have been observed at Sugar Run in Lycoming County near suspected leakage from unconventional gas activity. Given the elevated methane concentrations, the presence of thermogenic

13 δ C-CH4 values, and the known problematic gas wells in the direct vicinity, we suggest

139

further investigation is warranted in the Meshoppen area, ideally involving groundwater sampling in nearby homeowner water wells. The ambiguity involved in isotopic and methane concentrations at this site further reinforces that stream methane can be used as a monitoring tool, but further investigation is necessary for more firm conclusions.

140

Appendix D

Ludlow Seepage Case Study

[The following was submitted to the US Forest Service for review based on Forest Service concerns about Orange Seepage along US Route 6 in the ANF]

We sampled an iron seepage (iron-rich springs) near Ludlow, PA occasionally between 2017-2018 (See Tables E11-E13). The largest seepage is pictured below but multiple seepages are present (Figure 1). Sampling was only conducted at the largest seepage).

Figure D-1: Orange coloration associated with a spring emerging from the hillslope near US 6 (can be seen in the top right of the image). The white colored fibrous matting is tentatively attributed on the basis of visual observation alone to bacterial mats. Sampling at this site (Tables E11-E13) was conducted on three separate occasions.

The seepage emerges at various points along a roughly 2,000 ft stretch of US 6 where water flows along the side of the road during times of high water. There is a strong smell of hydrogen sulfide throughout the area. When water is low, we have observed a ferric iron-colored coating along the side of the road (Figure D-2). When water is high, water can be seen flowing along the side of the road and is associated with much more

141

obvious indications of iron coatings. Local residents have communicated that the springs, odor of H2S (rotten egg smell), and orange coatings are relatively new. Outside of this general area, springs and their associated sediments were observed to be clear in color with no hydrogen sulfide smell.

Figure D-2: Orange sediment observed along US 6 near the Ludlow seepages. When water levels are higher, orange discoloration is much more obvious.

Concentrations of dissolved hydrocarbons (methane (C1) and ethane (C2)) as well as cations and anions measured for this iron seepage reported in tables E11-E13. Methane concentrations in the largest seepage, Ludlow Seepage (Figure D-1) are elevated compared to typical springs in Pennsylvania (8-11 mg/L). Initial analysis of hydrocarbons dissolved in water are consistent with a wet gas source (96) ([C1]/[C2] <

16, where [C1] and [C2] refers to concentration of methane and ethane in mg/L). Carbon isotopic ratios (δ13C) of methane (-33.5‰ to -45.4‰) and ethane (-28.2‰ to -31‰) are consistent with a thermogenic gas that may be experiencing some oxidation by bacteria that can alter the isotopic ratio in the positive direction. Metals such as iron, manganese,

142

and arsenic are elevated above EPA drinking water standards. However, chloride (0.3 to

0.6 mg/L) is low and consistent with the interpretation that gas is not traveling with a brine-rich fluid.

The iron-rich springs are located in an area of historic gas wells that are not marked on the PA DEP oil and gas mapping tool but appear on a USGS topographic map

(these wells are shown on Figure D-3 as small open circles).

Figure D-3: Screenshot from the PA Oil and Gas mapping tool. The “X” marks the approximate location of the largest Ludlow seepage (41.712, -78.912), located along route 6. The seepages are located in an area of gas wells, many of which appear on a USGS base map (open circles) but do not appear in the PA DEP’s database (where the only wells are shown here as colored circles).

A conventional well (East Branch Field Well 1461P) is located a couple hundred feet to the north of the source of the iron seepage (Figure D-4).

143

Figure D-4: East Branch Field Well 1461P is located a couple hundred feet away from the start of the Ludlow iron seepages shown in Figure D-1. The dark color surrounding the base of the well is an unknown substance that may be oil related. According to the PA Oil and Gas mapping tool, the well is an “Multiple Well Borehole Type.” The spud date is not listed and the location given in the tool appears to be incorrect as the well is situated on the wrong side of US 6 on the map (Figure 1).

Further investigation is warranted because: i) the location of the iron seepage is visible; ii) local residents believe that the iron seepages are relatively new; ii) isotopic compositions and C1/C2 ratios are consistent with thermogenic wet gas in the seepage; iii) based on our inferences from Chapter 3, low chloride concentrations are consistent with gas traveling independently of brines (possibly as a free gas phase); IV) the seepages are located in an area with known gas wells and some newer development.

144

1 Appendix E Tables for Chapter 2 and Chapter 3 2 3 Table E1. Generalized timeline of events in the study area - This table appears almost identical to that which was published in SI for Woda 4 et al. (2018) Date (m/d/y) Events: Associated with problematic wells within 5 km of well 081-20292 9/14/2008 Well API # 081-20109 spudded 11/4/2008 Well API # 081-20119 spudded 10/31/2009 Well API # 081-20205 spudded 11/27/2009 Well API # 081-20209 spudded 7/9/2009 Well API # 081-20144 spudded 12/15/2009 Well API # 081-20210 spudded 6/1/2010 Well API # 081-20296 spudded 8/9/2010 Well API # 081-20348 spudded 8/19/2010 Well API # 081-20275 spudded 10/19/2010 Well API # 081-20287 spudded 19 private water wells located along Green Valley Road in Moreland Township were sampled within 2,500 feet of 081-20292. 11/1/2010 Methane was detected in four samples. 12/19/2010 Well API # 081-20371 spudded 12/20/2010 Well API # 081-20275 received violation 78.85; DEP observed gas in the 2" vent 12/20/2010 Well API # 081-20348 received two casing-related violations (78.85 & 78.86); DEP observed gas in 2" vent pipe Well API # 081-20209 receives two casing-related violations (78.85 & 78.86); DEP observed 18 psi pressure at the 5 ½" x 9 1/9/2011 5/8" pipe. 1/9/2011 Well API # 081-20210 received two casing-related violations (78.85 & 78.86); DEP observed 580psi on the 5 1/2" x 9 5/8". 1/9/2011 Well API # 081-20205 received two casing-related violations (78.85 & 78.86); DEP observed 685psi on the 5 1/2" x 9 5/8". 1/9/2011 Well API # 081-20119 received two casing-related violations (78.85 & 78.86); DEP observed 250psi on the 5 1/2" x 9 5/8". 1/9/2011 Well API # 081-20109 received two casing-related violations (78.85 & 78.86); DEP observed 150psi on the 5 1/2" x 9 5/8". 1/26/2011 Well API # 081-20144 received casing-related violations 78.86 & 78.85 2/1/2011 Well API # 081-20241 spudded 2/12/2011 Drilling began for well API # 081-20292 3/9/2011 Well API # 081- 20532 spudded

145

Date (m/d/y) Events: Associated with problematic wells within 5 km of well 081-20292 3/17/2011 Drilling completed for well API # 081-20292 3/25/2011 Company received three violations from DEP related to spillage of drilling mud 4/27/2011 Well API # 081-20496 spudded End 6/2011 Horizontal portion of API # 081-20292 hydraulically fractured 7/6/2011 Well API # 081-20296 received violation 78.86, defective cement; DEP observed gas in the annulus of the 9 5/8" x 5 1/2" casing. 7/10/2011 Well API # 081-20532 received violation 78.86, defective cement; DEP observed gas in the annulus of the 9 5/8" x 5 1/2" casing. 7/10/2011 Well API # 081-20287 received violation 78.86; defective cement; DEP observed gas in the annulus of the 9 5/8" x 5 1/2" casing. 7/10/2011 Well API # 081-20496 received violation 78.86, defective cement; DEP observed gas in the annulus of the 9 5/8 x 5 1/2 casing. 1/9/2012 DEP was given notice of discolored water in water supply near well API # 081-20292 1/18/2012 Water samples were collected from private water supplies for analysis 2/7/2012 DEP inspection found defective cement in annulus of well API # 081-20292 based on shut-in tests 12-Feb Waters from well API # 081-20292 (three casing strings) and homeowner wells were sampled for isotopic tests 5/14/2012 Gas detected outside surface casing in well API # 081-20292 4/10/2013 Well API # 081-20371 received casing-related violation 78.86; DEP discovered on 4/9/13 that the operator found combustible gas in the 9 5/8" x 5 1/2" annulus on 8/6/12 and that no action had been taken to correct the defect. DEP issued a notice of violation (NOV) to well API # 081-20292 for 1) failing to prevent migration of gas or other fluids into 9/20/2013 ground water (78.81(a)2 and (3)); 2) an unpermitted discharge of gas into surface or groundwaters (violation of section 401 of the Clean Stream Law; 3) defective casing or cementing (violation of section 78.86) 12/13/2013 DEP issued a NOV indicating failure to plug an abandoned well (78.91(a)), referring to well API # 081-20292

146

Date (m/d/y) Events: Associated with problematic wells within 5 km of well 081-20292 Since this date, DEP has investigated potential complaints involving stray gas migration in an area approx. 9,850 feet from 4/13/2015 well 081-20292 and 5,200 feet from originally affected water supplies noted above. Soil gases measured in a farm field with dead vegetation near Greg’s Run contained as much as 100% methane. Combustible gas was also identified in Greg’s Run. 5/11/2015 DEP issued order to remediate well API # 081-20292 6/5/2015 Company appealed order to remediate well API # 081-20292 Early 8/2015 DEP and company began making tests and field observations of the water and soil resources 10/22/2015 Company collected logging data from well API # 081-20292 16-Dec Drilling commenced through a cement plug the in 5.5" production casing in order to prepare well API # 081-20292 for flaring. This attempt to re-enter the lower portion of the casing at approximately 2,540 feet was unsuccessful. ?? Homeowners in the area were offered water treatment systems voluntarily by the company. 3/9/2016 Well API # 081-20241 received violation 78.86; DEP observed 4% methane in 13 3/8" x 95/8" annular vent 6/9/2016 Well API # 081-20371 received violation for 78.86 (pressure still noted in annulus). Well is shut in. 6/14/2016 Waters in wells near Greg’s Run (southwest of well API # 081-20292) observed to contain elevated methane and metals. 12/7/2016 Violations at Well API # 081-20296 still outstanding 3/27/2017 Violations at Well API # 081-20241 still outstanding; DEP observed 5% methane on the 13 3/8" x 9 5/8" annulus. 7/31/2018 Eight violations Well API # 081-20292 still outstanding: 78.73(a), 78.81(a)2, 78.81(a)3, 78.86, CSL 301, CSL 307(a), CSL 401, CSL 402(b) 5 Information in table was summarized and paraphrased from online and published reports (23, 27, 31, 107)

147

3 a 6 Table E2. Noble gas and methane concentrations (cm STP/gH2O) of water samples . - This table appears almost identical to that which 7 was published in SI for Woda et al. (2018) Sampling Sample CH He Ne Ar Kr Xe Date 4 HO2-A 7/18/2017 0.0232 4.6091E-06 1.4615E-07 2.6860E-04 7.5827E-08 1.4747E-08 HO2-B 7/18/2017 0.0232 5.0329E-06 1.5090E-07 2.7387E-04 8.0181E-08 1.4405E-08 HO4-A 7/26/2017 0.0686 2.2174E-06 8.2216E-08 1.4475E-04 3.5555E-08 5.4094E-09 HO4-B 7/26/2017 0.0686 2.2490E-06 8.3953E-08 1.4827E-04 3.6238E-08 5.6055E-09 Salt Spring 7/17/2017 0.0235 4.4078E-06 6.0571E-08 1.9561E-04 5.8446E-08 8.0483E-09 Water ASWb - - 4.4814E-08 1.9457E-07 3.7237E-04 8.7948E-08 1.2779E-08 8 Sample 4He 22Ne 36Ar 84Kr 132Xe 40Ar* +/- 4He Age (Ma) HO2-A 4.6091E-06 1.3480E-08 9.0355E-07 4.3221E-08 3.9653E-09 - - 0.78 HO2-B 5.0329E-06 1.3906E-08 9.2129E-07 4.5703E-08 3.8735E-09 - - 0.85 HO4-A 2.2174E-06 7.5831E-09 4.8694E-07 2.0266E-08 1.4546E-09 2.9598E-07 9.2239E-08 0.37 HO4-B 2.2490E-06 7.7427E-09 4.9878E-07 2.0656E-08 1.5073E-09 2.4959E-07 1.4356E-07 0.38 Salt Spring 4.4078E-06 5.5972E-09 6.5804E-07 3.3314E-08 2.1642E-09 - - 0.74 Water ASWb 4.4814E-08 1.7968E-08 1.2527E-06 5.0130E-08 3.4364E-09 - - - 9 a Errors of He, Ne, Ar, Kr, and Xe concentrations are 1.5%, 1.3%, 1.3%, 1.5%, and 2.2%, respectively 10 b (40)

148

Table E3. Noble gas volume fractions (cm3/cm3) of gas samplesa - This table appears almost identical to that which was published in SI for Woda et al. (2018) Sample Sampling Date He Ne Ar Kr Xe Seep 1.55-1 7/10/2017 2.4923E-04 5.8148E-07 3.8034E-04 5.5354E-08 7.1257E-09 Seep 1.55-2 7/10/2017 2.2308E-04 5.0435E-07 3.3451E-04 4.4825E-08 6.1898E-09 Salt Spring-1 7/17/2017 4.2829E-04 1.7102E-06 1.7291E-03 2.7790E-07 2.3181E-08 Salt Spring-2 7/17/2017 4.2654E-04 1.7786E-06 1.7729E-03 2.7652E-07 2.4059E-08 Air b - 5.2400E-06 1.8180E-05 9.3400E-03 1.1400E-06 8.7000E-08

Sample 4He 22Ne 36Ar 84Kr 132Xe 40Ar* +/- Seep 1.55-1 2.4923E-04 5.3395E-08 1.2795E-06 3.1552E-08 1.9161E-09 2.5356E-06 2.4439E-07 Seep 1.55-2 2.2308E-04 4.6267E-08 1.1253E-06 2.5550E-08 1.6644E-09 2.7019E-06 3.257E-07 Salt Spring-1 4.2829E-04 1.5755E-07 5.8167E-06 1.5840E-07 6.2335E-09 - - Salt Spring-2 4.2654E-04 1.6381E-07 5.9640E-06 1.5762E-07 6.4695E-09 - - Air b 5.2400E-06 1.6780E-06 3.1420E-05 6.4980E-07 2.3394E-08 - - a Errors of He, Ne, Ar, Kr, and Xe concentrations are 1.5%, 1.3%, 1.3%, 1.5%, and 2.2%, respectively b (40)

149

Table E4. Noble gas isotopic ratios of water and gas samples - This table appears almost identical to that which was published in SI for Woda et al. (2018) Sample 20Ne/22Ne +/- 21Ne/22Ne +/- 38Ar/36Ar +/- 40Ar/36Ar +/- 80Kr/84Kr +/- Water Samples HO2-A 9.8121 0.0018 0.02902 0.00001 0.1878 0.0001 294.9504 0.1935 0.0395 0.0001 HO2-B 9.8204 0.0013 0.02904 0.00001 0.1879 0.0001 295.3313 0.1677 0.0396 0.0001 HO4-A 9.8119 0.0027 0.02901 0.00002 0.1883 0.0001 296.1078 0.1926 0.0395 0.0001 HO4-B 9.8128 0.0026 0.02908 0.00002 0.1880 0.0001 296.0004 0.1748 0.0394 0.0001 Salt Spring Water 9.7935 0.0019 0.02909 0.00003 0.1878 0.0001 295.3836 0.1713 0.0396 0.0001 Gas Samples Seep 1.55-1 9.8555 0.0022 0.02928 0.00002 0.1880 0.0001 297.4818 0.1709 0.0393 0.0001 Seep 1.55-2 9.8654 0.0021 0.02924 0.00003 0.1882 0.0001 297.9010 0.1924 0.0394 0.0001 Salt Spring-1 9.8234 0.0016 0.02905 0.00001 0.1878 0.0001 294.9606 0.1635 0.0400 0.0001 Salt Spring-2 9.8262 0.0017 0.02906 0.00002 0.1880 0.0001 295.2091 0.1709 0.0399 0.0001 Air a 9.8000 0.02900 0.1880 295.5000 0.0396

Sample 82Kr/84Kr +/- 83Kr/84Kr +/- 86Kr/84Kr +/- 128Xe/130Xe +/- 129Xe/130Xe +/- Water Samples HO2-A 0.2023 0.0003 0.2015 0.0002 0.3060 0.0004 0.4663 0.0009 6.565 0.013 HO2-B 0.2022 0.0003 0.2012 0.0002 0.3055 0.0004 0.4674 0.0010 6.599 0.012 HO4-A 0.2021 0.0002 0.2013 0.0002 0.3052 0.0003 0.4699 0.0008 6.533 0.012 HO4-B 0.2020 0.0003 0.2011 0.0003 0.3055 0.0004 0.4716 0.0008 6.515 0.013 Salt Spring Water 0.2024 0.0003 0.2013 0.0002 0.3052 0.0004 0.4699 0.0008 6.536 0.012 Gas Samples Seep 1.55-1 0.2019 0.0002 0.2014 0.0002 0.3058 0.0004 0.4724 0.0010 6.493 0.012 Seep 1.55-2 0.2018 0.0003 0.2009 0.0002 0.3053 0.0004 0.4708 0.0010 6.497 0.013 Salt Spring-1 0.2021 0.0002 0.2013 0.0002 0.3056 0.0004 0.4693 0.0009 6.533 0.012 Salt Spring-2 0.2021 0.0002 0.2014 0.0002 0.3056 0.0004 0.4702 0.0010 6.519 0.012 Air a 0.2022 0.2014 0.3052 0.4715 6.496

150

Sample 131Xe/130Xe +/- 132Xe/130Xe +/- 134Xe/130Xe +/- 136Xe/130Xe +/- Water Samples HO2-A 5.251 0.009 6.680 0.007 2.587 0.002 2.198 0.001 HO2-B 5.266 0.009 6.694 0.007 2.593 0.002 2.197 0.002 HO4-A 5.231 0.008 6.646 0.005 2.575 0.001 2.189 0.001 HO4-B 5.226 0.008 6.624 0.006 2.570 0.002 2.183 0.001 Salt Spring Water 5.246 0.008 6.661 0.006 2.585 0.002 2.201 0.002 Gas Samples Seep 1.55-1 5.214 0.008 6.593 0.006 2.555 0.002 2.165 0.002 Seep 1.55-2 5.207 0.008 6.601 0.006 2.562 0.002 2.171 0.002 Salt Spring-1 5.227 0.007 6.644 0.005 2.576 0.002 2.190 0.002 Salt Spring-2 5.221 0.007 6.626 0.006 2.566 0.002 2.185 0.002 Air a 5.213 6.607 2.563 2.176

4 40 4 20 20 36 4 -6 Sample R/Ra +/- He/ Ar* +/- He/ Ne +/- Ne/ Ar +/- He/CH4(x10 ) +/- Water Samples HO2-A 0.0237 0.0005 - 34.85 0.69 0.146 0.003 198.57 2.98 HO2-B 0.0231 0.0005 - 36.85 0.73 0.148 0.003 216.83 3.25 HO4-A 0.0589 0.0009 7.49 2.34 29.80 0.59 0.153 0.003 32.32 0.48 HO4-B 0.0583 0.0011 9.01 5.18 29.60 0.59 0.152 0.003 32.78 0.49 Salt Spring Water 0.0165 0.0006 - 80.41 1.60 0.083 0.002 187.33 2.81 Gas Samples Seep 1.55-1 0.0114 0.0005 98.29 9.59 473.61 9.40 0.411 0.008 - Seep 1.55-2 0.0116 0.0004 82.57 10.03 488.74 9.70 0.406 0.007 - Salt Spring-1 0.0151 0.0005 - 276.73 5.49 0.266 0.005 - Salt Spring-2 0.0160 0.0004 - 265.00 5.26 0.270 0.005 - Air a 1.0000 - 0.318 0.524 0.010 - a (40)

151

Table E5. Hydrocarbon - This table appears almost identical to that which was published in SI for Woda et al. (2018). A few data entries (noted by a footnote) were added where they were mistakenly left out of the published table, or were corrected where mistakes were found in the published version Sampling δ13C- δD- Ethan δ13C- Sampling a Water Propane C1/C2 Latitude Longitude Data Source b Technique CH4 (µg/L) CH4 CH4 e C2H6 Date Type c (µg/L) (molar) for CH4 (‰) (‰) (µg/L) (‰) SR 0.4 7/10/2017 41.2308 -76.7078 this study SW Stream 0.8 ------SR 0.5 7/10/2017 41.2313 -76.7071 this study SW Stream 0.8 ------SR 1.0

Heilweil et al. 5/21/2013 41.2359 -76.6956 SW Stream 5.0 ------(2014; 2015)

Heilweil et al. 6/27/2013 41.2359 -76.6956 SW Stream 2.3 ------(2014; 2015)

Heilweil et al. 11/12/2013 41.2359 -76.6956 SW Stream 1.3 ------(2014; 2015)

Wendt et al. 7/9/2015 41.2359 -76.6956 SW Stream 5.0 ------(2018) SR 1.1 Wendt et al. 6/13/2016 41.2372 -76.6937 SW Stream 20.4 ------(2018) 8/15/2016 41.2372 -76.6937 this study SW Stream 1.0 ------9/15/2016 41.2372 -76.6937 this study SW Stream 1.3 -54.4 - - - - - 10/26/2016 41.2372 -76.6937 this study SW Stream 6.6 -13.2 - - - - - 11/16/2016 41.2372 -76.6937 this study SW Stream 4.7 ------1/16/2017 41.2372 -76.6937 this study SW Stream 11.4 ------

152

Heilweil et al. 11/12/2013 41.2377 -76.6933 SW Stream 6.4 -16.9 -70.1 - - - - (2014; 2015)

Wendt et al. 7/9/2015 41.2377 -76.6933 SW Stream 10.0 ------(2018) 6/27/2016 41.2377 -76.6933 this study SW Stream 10.9 ------8/15/2016 41.2377 -76.6933 this study SW Stream 2.9 ------9/15/2016 41.2377 -76.6933 this study SW Stream 0.7 -19.6 - - - - - 11/16/2016 41.2377 -76.6933 this study SW Stream 8.4 ------1/16/2017 41.2377 -76.6933 this study SW Stream 12.3 ------2/24/2017 41.2377 -76.6933 this study SW Stream 13.1 ------5/10/2017 41.2377 -76.6933 this study SW Stream 10.9 ------6/22/2017 41.2377 -76.6933 this study SW Stream 8.6 ------

153

Heilweil et al. 11/12/2013 41.2381 -76.6930 SW Stream 16.8 -22.1 -111 - - - - (2014; 2015)

12/9/2014 41.2381 -76.6930 this study SW Stream 11.2 ------Wendt et al. 7/9/2015 41.2381 -76.6930 SW Stream 13.3 ------(2018) 8/15/2016 41.2381 -76.6930 this study SW Stream 7.1 ------9/15/2016 41.2381 -76.6930 this study SW Stream 1.5 -13.9 - - - - - 10/26/2016 41.2381 -76.6930 this study SW Stream 16.3 ------11/16/2017 41.2381 -76.6930 this study SW Stream 12.6 ------1/16/2017 41.2381 -76.6930 this study SW Stream 12.3 ------2/24/2017 41.2381 -76.6930 this study SW Stream 16.6 ------5/10/2017 41.2381 -76.6930 this study SW Stream 10.9 ------6/22/2017 41.2381 -76.6930 this study SW Stream 17.1 ------SR 1.5

Heilweil et al. 6/27/2013 41.2391 -76.6923 SW Stream 67 ------(2014; 2015)

Heilweil et al. 11/12/2013 41.2391 -76.6923 SW Stream 28.2 -25.2 -135 - - - - (2014; 2015)

Wendt et al. 12/9/2014 41.2391 -76.6923 SW Stream 8.4 ------(2018) Wendt et al. 7/9/2015 41.2391 -76.6923 SW Stream 10.6 ------(2018) Wendt et al. 6/13/2016 41.2391 -76.6923 SW Stream 22.9 ------(2018) 8/15/2016 41.2391 -76.6923 this study SW Stream 16.7 ------

154

Sampling δ13C- δD- Ethan δ13C- Sampling a Water Propane C1/C2 Latitude Longitude Data Source b Technique CH4 (µg/L) CH4 CH4 e C2H6 Date Type c (µg/L) (molar) for CH4 (‰) (‰) (µg/L) (‰) 9/15/2016 41.2391 -76.6923 this study SW Stream 14.6 -11.4 - - - - - 10/26/2016 41.2391 -76.6923 this study SW Stream 61.6 -20.8 - - - - - 11/16/2016 41.2391 -76.6923 this study SW Stream 50.2 -22.7 - - - - - 1/16/2017 41.2391 -76.6923 this study SW Stream 9.9 ------2/24/2017 41.2391 -76.6923 this study SW Stream 11.6 ------5/10/2017 41.2391 -76.6923 this study SW Stream 17.6 ------6/22/2017 41.2391 -76.6923 this study SW Stream 24.3 -23.8 - - - - - 7/10/2017 41.2391 -76.6923 this study SW Stream 32.5 ------9/8/2017 41.2391 -76.6923 this study SW Stream 26.8 ------10/2/2017 41.2391 -76.6923 this study SW Stream 26.6 ------SR 1.55 Wendt et al. 7/9/2015 41.2398 -76.6919 SW Stream 7.5 ------(2018) 8/15/2016 41.2398 -76.6919 this study SW Stream 12.7 ------9/15/2016 41.2398 -76.6919 this study SW Stream 76.6 -15.2 - - - - - 10/26/2016 41.2398 -76.6919 this study SW Stream 24.2 -20.0 - - - - - 11/16/2016 41.2398 -76.6919 this study SW Stream 17.6 -21.8 - - - - - 1/16/2017 41.2398 -76.6919 this study SW Stream 10.6 ------2/24/2017 41.2398 -76.6919 this study SW Stream 10.4 ------5/10/2017 41.2398 -76.6919 this study SW Stream 6.5 ------6/22/2017 41.2398 -76.6919 this study SW Stream 16.9 ------9/8/2017 41.2398 -76.6919 this study SW Stream 14.0 ------SR 1.6

Heilweil et al. 11/12/2013 41.2403 -76.6912 SW Stream 5.9 -23.8 -121 - - - - (2014; 2015)

Wendt et al. 12/9/2014 41.2403 -76.6912 SW Stream 6.2 ------(2018)

155

Sampling δ13C- δD- Ethan δ13C- Sampling a Water Propane C1/C2 Latitude Longitude Data Source b Technique CH4 (µg/L) CH4 CH4 e C2H6 Date Type c (µg/L) (molar) for CH4 (‰) (‰) (µg/L) (‰) Wendt et al. 7/9/2015 41.2403 -76.6912 SW Stream 7.1 ------(2018) Wendt et al. 6/13/2016 41.2403 -76.6912 SW Stream 5.8 ------(2018) 8/15/2016 41.2403 -76.6912 this study SW Stream 2.1 ------9/15/2016 41.2403 -76.6912 this study SW Stream 25.2 -19.4 - - - - - SR 1.8

Heilweil et al. 11/12/2013 41.2409 -76.6910 SW Stream 4.2 -17.9 -66.7 - - - - (2014; 2015)

Wendt et al. 7/9/2015 41.2409 -76.6910 SW Stream 7.1 ------(2018) Wendt et al. 6/13/2016 41.2409 -76.6910 SW Stream 5.8 ------(2018) 8/15/2016 41.2409 -76.6910 this study SW Stream 2.6 ------9/15/2016 41.2409 -76.6910 this study SW Stream 3.6 -42.1 - - - - - 10/26/2016 41.2409 -76.6910 this study SW Stream 4.0 -17.0 - - - - - 1/16/2017 41.2409 -76.6910 this study SW Stream 10.1 ------SR 2

Heilweil et al. 5/21/2013 41.2407 -76.6898 SW Stream 20 ------(2014; 2015)

Heilweil et al. 6/27/2013 41.2407 -76.6898 SW Stream 7.2 ------(2014; 2015)

Heilweil et al. 11/12/2013 41.2407 -76.6898 SW Stream 5.7 -16.3 -58.2 - - - - (2014; 2015)

Wendt et al. 7/9/2015 41.2407 -76.6898 SW Stream 9.3 ------(2018)

156

Heilweil et al. 5/21/2013 41.2431 -76.6830 SW Stream 2.2 ------(2014; 2015)

Heilweil et al. 6/27/2013 41.2431 -76.6830 SW Stream 2.4 ------(2014; 2015)

7/10/2017 41.2431 -76.6830 this study SW Stream 4.9 ------SR 4

Heilweil et al. 5/21/2013 41.2438 -76.6800 SW Stream 1.1 ------(2014; 2015)

2/24/2017 41.2438 -76.6800 this study SW Stream 3.9 ------SR 6

Heilweil et al. 5/21/2013 41.2494 -76.6687 SW Stream 0.5 ------(2014; 2015)

SR 7

Heilweil et al. 5/21/2013 41.2537 -76.6646 SW Stream 0.4 ------(2014; 2015)

157

Heilweil et al. 11/12/2013 41.2395 -76.6922 Seep Stream 2290 -26.7 -142 34 -32.6 - 126 (2014; 2015)

Wendt et al. 6/13/2016 41.2395 -76.6922 Seep Stream 216 ------(2018) 8/15/2016 41.2395 -76.6922 this study Seep Stream 117 ------9/15/2016 41.2395 -76.6922 this study Seep Stream 830 -52.3 - - - - - 10/26/2016 41.2395 -76.6922 this study Seep Stream 4350 -18.0 - - - - - 11/16/2016 41.2395 -76.6922 this study Seep Stream 4350 -21.3 - - - - - 1/16/2017 41.2395 -76.6922 this study Seep Stream 1.25 -2.0 - - - - - 2/24/2017 41.2395 -76.6922 this study Seep Stream 355 3.4 - - - - - 5/10/2017 41.2395 -76.6922 this study Seep Stream 1260 -20.7 - - - - - 6/22/2017 41.2395 -76.6922 this study Seep Stream 557 -10.4 - - - - - 6/28/2017 41.2395 -76.6922 this study Seep Stream 1820 ------7/10/2017 41.2395 -76.6922 this study Seep Stream 3170 -16.9 - 17 - - 359 9/8/2017 41.2395 -76.6922 this study Seep Stream 2040 -24.0 - - - - - 10/2/2017 41.2395 -76.6922 this study Seep Stream 4410 - - 40 - - 206 10/16/2017 41.2395 -76.6922 this study Seep Stream 1600 ------12/3/2017 41.2395 -76.6922 this study Seep Stream 322 -21.0 - - - - - Seep 1.55 9/16/2016 41.2398 -76.6921 this study Seep Stream ------10/26/2016 41.2398 -76.6921 this study Seep Stream 19.5 37.2 - - - - -

158

Sampling δ13C- δD- Ethan δ13C- Sampling a Water Propane C1/C2 Latitude Longitude Data Source b Technique CH4 (µg/L) CH4 CH4 e C2H6 Date Type c (µg/L) (molar) for CH4 (‰) (‰) (µg/L) (‰) 11/16/2016 41.2398 -76.6921 this study Seep Stream 167 13.0 - - - - - 1/16/2017 41.2398 -76.6921 this study Seep Stream 0.1 37.6 - - - - - 2/24/2017 41.2398 -76.6921 this study Seep Stream 0.2 -3.8 - - - - - 5/10/2017 41.2398 -76.6921 this study Seep Stream 378 -17.2 - - - - - 6/2/2017 41.2398 -76.6921 this study Seep Stream 626 ------6/22/2017 41.2398 -76.6921 this study Seep Stream 581 -18.3 - - - - - 6/28/2017 41.2398 -76.6921 this study Seep Stream 315 ------7/10/2017 41.2398 -76.6921 this study Seep Stream ------9/8/2017 41.2398 -76.6921 this study Seep Stream ------12/3/2017 41.2398 -76.6921 this study Seep Stream 8.1 26.7 - - - - - Seep 1.6 10/26/2016 41.2400 -76.6915 this study Seep Stream 4570 -25.4 - - - - - 11/16/2016 41.2400 -76.6915 this study Seep Stream 6010 -26.7 - - - - - 1/16/2017 41.2400 -76.6915 this study Seep Stream 5850 -25.0 - - - - - 2/24/2017 41.2400 -76.6915 this study Seep Stream 8590 -28.3 - - - - - 5/10/2017 41.2400 -76.6915 this study Seep Stream 6750 -25.8 - - - - - 6/2/2017 41.2400 -76.6915 this study Seep Stream 4220 ------6/22/2017 41.2400 -76.6915 this study Seep Stream 2500 -22.5 - - - - - 6/28/2017 41.2400 -76.6915 this study Seep Stream 1690 ------7/10/2017 41.2400 -76.6915 this study Seep Stream 981 4.9 - - - - - 7/26/2017 41.2400 -76.6915 this study Seep Stream 2850 -23.5 - - - - - 9/8/2017 41.2400 -76.6915 this study Seep Stream 1450 -18.4 - - - - - 10/2/2017 41.2400 -76.6915 this study Seep Stream 1670 - - 5.4 - - 579 10/16/2017 41.2400 -76.6915 this study Seep Stream 3340 ------12/3/2017 41.2400 -76.6915 this study Seep Stream 6570 -28.0 - 60 - - 204 SRS Seep 1.5

159

Sampling δ13C- δD- Ethan δ13C- Sampling a Water Propane C1/C2 Latitude Longitude Data Source b Technique CH4 (µg/L) CH4 CH4 e C2H6 Date Type c (µg/L) (molar) for CH4 (‰) (‰) (µg/L) (‰) Not 2/24/2017 41.2461 -76.6809 this study Seep collected ------for CH4 HO1 - Not 10/26/2016 - - this study GW collected ------for CH4 2/24/2017 - - this study GW Inverted 1990 -30.3 - - - - - 5/10/2017 - - this study GW Inverted 2530 ------7/10/2017 - - this study GW Inverted 3000 -26.3 - - - - - 10/19/2017 - - PADEP GW Unknown 4290 ------HO2 7/18/2017 - - this study GW Inverted 16600 - - 51 - - 608 9/8/2017 - - this study GW Inverted 26400 -11.2 - - - - - 9/8/2017 - - this study GW Isoflask 28000 -29.8 -175.4 120 - <0.3 437 10/19/2017 - - PADEP GW Unknown 26070 - - 101 - <14.2 484 HO3 - 6/13/2016 - - PADEP GW Unknown 27500g - - 578 - <14.2 0.1 11/16/2016 - - this study GW Inverted 31500 -28.4 - - - - - HO4 7/12/2010 - - RR GW Unknown 4450 ------12/5/2011 - - RR GW Unknown 20500 - - 224 - - 172 1/24/2012 - - PADEP GW Unknown 72800 ------1/30/2012 - - RR GW Unknown 22520 - - 429 - - 98 4/10/2012 - - RR GW Unknown 35000 - - 330 - - 199 4/26/2012 - - RR GW Unknown 16980 - - 560 - - 57 5/8/2012 - - RR GW Unknown 10210 - - 270 - - 71

160

Sampling δ13C- δD- Ethan δ13C- Sampling a Water Propane C1/C2 Latitude Longitude Data Source b Technique CH4 (µg/L) CH4 CH4 e C2H6 Date Type c (µg/L) (molar) for CH4 (‰) (‰) (µg/L) (‰) 5/22/2012 - - RR GW Unknown 27730 - - 765 - - 68 7/18/2012 - - RR GW Unknown 39260 - - 336 - - 219 7/18/2012 - - PADEP GW Unknown 69400 ------6/14/2016 - - PADEP GW Unknown 61600 - - 1790 - - 65 7/26/2017 - - this study GW Inverted 29100 - - 590 - - 92 7/26/2017 - - this study GW Isoflask 49000 -27.5 -170.1 1200 - 18 77 Not 7/26/2017e this study GW collected ------for CH4 HO5 11/10 - - RR GW Unknown 7820 - - ND - - - 1/18/2012 - - PADEP GW Unknown 41200 ------1/30/2012 - - RR GW Unknown 16800 - - 258 - - 122 3/8/2012 - - PADEP GW Unknown 37500 ------4/10/2012 - - RR GW Unknown 13460 - - 278 - - 91 4/26/2012 - - RR GW Unknown 13230 - - 303 - - 82 5/8/2012 - - RR GW Unknown 12856 - - 296 - - 81 5/22/2012 - - RR GW Unknown 18553 - - 415 - - 84 7/18/2012 - - RR GW Unknown 14799 - - 336 - - 83 7/18/2012 - - PADEP GW Unknown 44700g ------HO6 4/13/2011 - - RR GW Unknown 2690 - - ND - - - 11/21/2011 - - RR GW Unknown 19200 - - 108 - - 333 1/25/2012 - - PADEP GW Unknown 47400 ------1/30/2012 - - RR GW Unknown 20040 - - 191.5 - - 196 3/8/2012 - - PADEP GW Unknown 21000 ------4/10/2012 - - RR GW Unknown 29990 - - 351 - - 160

161

Sampling δ13C- δD- Ethan δ13C- Sampling a Water Propane C1/C2 Latitude Longitude Data Source b Technique CH4 (µg/L) CH4 CH4 e C2H6 Date Type c (µg/L) (molar) for CH4 (‰) (‰) (µg/L) (‰) 4/26/2012 - - RR GW Unknown 21865 - - 243 - - 169 5/8/2012 - - RR GW Unknown 20352 - - 257 - - 148 5/22/2012 - - RR GW Unknown 17735 - - 224 - - 148 7/18/2012 - - RR GW Unknown 33108 - - 470 - - 132 7/18/2012 - - RR GW Unknown 75200 ------6/16/2016 - - PADEP GW Unknown 90200 - - 1820 - 23.5 93 HO7 11/5/2010 - - RR GW Unknown 4060 ------2/8/2012 - - RR GW Unknown 1400 ------HO8-A 11/2/2010 - - RR GW Unknown 24020 ------11/10/2010 - - RR GW Unknown 15480 ------2/10/2012 - - RR GW Unknown 290 ------HO8-B 11/2/2010 - - RR GW Unknown 33910 ------11/10/2010 - - RR GW Unknown 29110 ------2/10/2012 - - RR GW Unknown 35000 ------HO8-A/Bg 6/13/2016 - - PADEP GW Unknown 4000 - - 54.5 - - 138 6/13/2016 - - PADEP GW Unknown 74300 - - 1880 - - 74 HO9 11/2/2010 - - RR GW Unknown 0 ------2/7/2012 - - RR GW Unknown 1900 ------H10 11/9/2010 - - RR GW Unknown 0 ------2/7/2012 - - RR GW Unknown 7200 ------

162

Heilweil et al. 11/12/2013 - - GW Stream 4590 -27.2 -139 42.86 -33 - 201 (2014; 2015)

a RR: Range Resources; PADEP: Pennsylvania Department of Environmental Protectoin b SW: surface water; GW: groundwater; Seep: water discharging from seepage c Stream: Heilweil et al. (2014) d ND: not detected; -: not measured eSample was not filtered before acidification. See text under "Methane Impacts on Groundwater" fHydrocarbon concentrations and carbon isotopic ratios were analyzed by Isotech Laboratories, Inc. g Adjusted from published version

163

Table E6. Cations – all mg/L unless otherwise stated - This table appears almost identical to that which was published in SI for Woda et al. (2018) Sampling Water Data Sourcea Al Ba Ca Fe K Mg Mn Na Si Sr Asg Ug Date Typeb SR 0.4 7/10/2017 this study SW ------SR 0.5 7/10/2017 this study SW ------SR 1.0 Heilweil et al. (2014; 5/21/2013 SW ------2015) Heilweil et al. (2014; 6/27/2013 SW ------2015) Heilweil et al. (2014; 11/12/2013 SW <0.01 0.02 17.4 <0.02 1.33 3.75 0.003 4.9 2.3 0.06 - - 2015) 7/9/2015 Wendt et al. (2018) SW ------SR 1.1 6/13/2016 Wendt et al. (2018) SW ------8/15/2016 this study SW ------9/15/2016 this study SW 0.001 0.02 18.3 0.02 1.79 3.83 <0.005 5.39 3.12 0.06 0.52 0.032 10/26/2016 this study SW ------11/16/2016 this study SW 0.002 0.02 17.2 0.01 1.27 3.75 0.00 5.07 1.89 0.05 0.21 0.006 1/16/2017 this study SW 0.003 0.02 10.1 0.01 1.07 2.64 0.01 3.89 2.82 0.03 0.16 0.005 2/24/2017 this study SW <0.01 0.02 10.2 0.02 1.3 2.6 0.01 4.54 2.28 0.04 - - 5/10/2017 this study SW 0.007 0.02 10.1 0.02 1.09 2.51 0.01 3.43 2.91 0.04 0.21 0.006 6/22/2017 this study SW 0.014 0.03 17.7 0.03 1.49 3.69 0.01 5.18 3.06 0.06 - - 7/10/2017 this study SW 0.009 0.03 18.7 0.08 1.68 3.71 0.02 5.39 2.72 0.06 - - 7/26/2017 this study SW ------9/8/2017 this study SW <0.005 0.03 19.6 0.04 1.55 3.86 0.01 5.83 2.78 0.06 - - 10/2/2017 this study SW 0.007 0.03 21.9 0.03 1.85 4.46 0.01 6.54 2.45 0.07 0.46 0.036

164

Sampling Water Data Sourcea Al Ba Ca Fe K Mg Mn Na Si Sr Asg Ug Date Typeb 12/3/2017 this study SW <0.005 0.02 14.2 0.03 1.14 3.22 0.01 4.44 1.91 0.04 - - SR 1.15 7/9/2015 Wendt et al. (2018) SW ------8/15/2016 this study SW ------9/15/2016 this study SW ------10/26/2016 this study SW ------SR 1.2 Heilweil et al. (2014; 11/12/2013 SW <0.01 0.03 17.2 <0.02 1.32 3.7 0.01 4.81 2.3 0.06 - - 2015) 7/9/2015 Wendt et al. (2018) SW ------6/27/2016 this study SW ------8/15/2016 this study SW ------9/15/2016 this study SW ------11/16/2016 this study SW ------1/16/2017 this study SW ------2/24/2017 this study SW ------5/10/2017 this study SW ------6/22/2017 this study SW ------SR 1.4 Heilweil et al. (2014; 11/12/2013 SW <0.01 0.02 17.2 <0.02 1.32 3.68 0.02 4.84 2.4 0.06 - - 2015) 12/9/2014 this study SW ------7/9/2015 Wendt et al. (2018) SW ------8/15/2016 this study SW ------9/15/2016 this study SW ------10/26/2016 this study SW ------11/16/2017 this study SW ------1/16/2017 this study SW ------

165

Sampling Water Data Sourcea Al Ba Ca Fe K Mg Mn Na Si Sr Asg Ug Date Typeb 2/24/2017 this study SW ------5/10/2017 this study SW ------6/22/2017 this study SW ------SR 1.5 Heilweil et al. (2014; 6/27/2013 SW ------2015) Heilweil et al. (2014; 11/12/2013 SW <0.01 0.02 17.2 <0.02 1.33 3.7 0.04 4.83 2.5 0.06 - - 2015) 12/9/2014 Wendt et al. (2018) SW ------7/9/2015 Wendt et al. (2018) SW ------6/13/2016 Wendt et al. (2018) SW ------8/15/2016 this study SW ------9/15/2016 this study SW 0.001 0.03 18.3 0.02 1.79 3.82 0.00 5.36 3.15 0.06 0.50 0.023 10/26/2016 this study SW 0.002 0.03 19 0.01 1.52 4.57 0.02 5.54 2.42 0.07 ND 0.024 11/16/2016 this study SW 0.019 0.02 16.8 0.05 1.29 3.71 0.03 4.86 1.95 0.05 - - 1/16/2017 this study SW ------2/24/2017 this study SW <0.01 0.02 10.2 0.02 1.22 2.61 0.01 4.66 2.23 0.03 - - 5/10/2017 this study SW ------6/22/2017 this study SW ------7/10/2017 this study SW ------9/8/2017 this study SW ------10/2/2017 this study SW ------SR 1.55 7/9/2015 Wendt et al. (2018) SW ------8/15/2016 this study SW ------9/15/2016 this study SW 0.002 0.03 18.3 0.10 1.79 3.82 0.02 5.36 3.15 0.06 0.58 0.02 10/26/2016 this study SW <0.005 0.03 19 <0.02 1.52 4.56 0.0003 5.46 2.4 0.07 ND 0.024 11/16/2016 this study SW ------

166

Sampling Water Data Sourcea Al Ba Ca Fe K Mg Mn Na Si Sr Asg Ug Date Typeb 1/16/2017 this study SW ------2/24/2017 this study SW ------5/10/2017 this study SW ------6/22/2017 this study SW ------9/8/2017 this study SW ------SR 1.6 Heilweil et al. (2014; 11/12/2013 SW <0.01 0.03 17.4 <0.02 1.36 3.72 0.01 4.85 2.4 0.06 - - 2015) 12/9/2014 Wendt et al. (2018) SW ------7/9/2015 Wendt et al. (2018) SW ------6/13/2016 Wendt et al. (2018) SW ------8/15/2016 this study SW ------9/15/2016 this study SW ------SR 1.8 Heilweil et al. (2014; 11/12/2013 SW <0.01 0.02 17.3 <0.02 1.36 3.72 0.02 4.82 2.4 0.06 - - 2015) 7/9/2015 Wendt et al. (2018) SW ------6/13/2016 Wendt et al. (2018) SW ------8/15/2016 this study SW ------9/15/2016 this study SW ------10/26/2016 this study SW ------1/16/2017 this study SW ------SR 2 Heilweil et al. (2014; 5/21/2013 SW ------2015) Heilweil et al. (2014; 6/27/2013 SW ------2015) Heilweil et al. (2014; 11/12/2013 SW <0.01 0.02 17.2 <0.02 1.36 3.7 0.02 4.85 2.4 0.06 - - 2015)

167

Sampling Water Data Sourcea Al Ba Ca Fe K Mg Mn Na Si Sr Asg Ug Date Typeb 7/9/2015 Wendt et al. (2018) SW ------6/13/2016 Wendt et al. (2018) SW ------9/15/2016 this study SW ------10/26/2016 this study SW ------1/16/2017 this study SW ------2/24/2017 this study SW ------6/22/2017 this study SW ------7/10/2017 this study SW ------SR 3 Heilweil et al. (2014; 5/21/2013 SW ------2015) Heilweil et al. (2014; 6/27/2013 SW ------2015) 7/10/2017 this study SW ------SR 4 Heilweil et al. (2014; 5/21/2013 SW ------2015) 2/24/2017 this study SW <0.01 0.02 10.4 0.02 1.34 2.67 0.01 4.71 2.39 0.04 - - SR 6 Heilweil et al. (2014; 5/21/2013 SW ------2015) SR 7 Heilweil et al. (2014; 5/21/2013 SW ------2015) SR 7.5 2/24/2017 this study SW <0.01 0.04 11.5 0.06 1.31 2.68 0.01 4.46 2.68 0.03 - - SR 8 7/10/2017 this study SW ------

168

Sampling Water Data Sourcea Al Ba Ca Fe K Mg Mn Na Si Sr Asg Ug Date Typeb Seep 1.5 Heilweil et al. (2014; 11/12/2013 Seep <0.01 0.03 5.05 0.05 0.75 2.17 0.10 1.44 4 0.03 - - 2015) 6/13/2016 Wendt et al. (2018) Seep ------8/15/2016 this study Seep ------9/15/2016 this study Seep <0.005 0.04 8.00 0.42 1.25 2.87 0.63 2.51 4.47 0.05 1.62 ND 10/26/2016 this study Seep 0.004 0.02 8.13 0.12 1.02 3.31 0.00 2.2 4.05 0.06 1.79 0.002 11/16/2016 this study Seep 0.001 0.03 7.83 0.12 0.98 2.89 0.64 2.12 3.79 0.05 0.55 ND 1/16/2017 this study Seep 0.007 0.01 3.81 0.28 0.59 1.55 0.17 1.18 3.12 0.02 0.28 0.006 2/24/2017 this study Seep 0.010 0.01 4.44 0.33 0.62 1.78 0.26 1.55 3.3 0.02 0.67 0.014 5/10/2017 this study Seep 0.016 0.02 5.51 1.14 0.77 2.22 0.50 1.54 3.8 0.03 2.36 0.016 6/22/2017 this study Seep 0.010 0.02 6.45 1.29 0.88 2.59 0.40 1.85 4.16 0.03 2.65 0.031 6/28/2017 this study Seep 0.009 0.03 7.82 4.18 1.05 2.96 0.68 2.05 4.49 0.05 5.81 0.01 7/10/2017 this study Seep 0.009 0.03 8.09 5.28 1.13 3.08 0.71 2.03 4.58 0.06 7.70 0.008 9/8/2017 this study Seep 0.018 0.04 8.64 5.40 1.27 3.1 0.71 3.31 4.57 0.06 7.15 0.01 10/2/2017 this study Seep 0.002 0.04 10.0 5.49 1.18 3.53 0.87 3.74 4.31 0.07 6.37 ND 10/16/2017 this study Seep 0.003 0.04 8.75 4.99 1.19 3.18 0.66 2.6 4.78 0.05 6.32 ND 12/3/2017 this study Seep 0.003 0.02 4.7 0.45 0.64 1.9 0.11 1.78 3.34 0.02 ND 0.007 Seep 1.55 9/16/2016 this study Seep 0.006 0.02 5.53 0.02 0.95 2.39 <0.005 2.1 4.31 0.03 0.49 0.002 10/26/2016 this study Seep 0.008 0.02 5.26 0.25 0.97 2.37 0.23 1.81 3.93 0.03 0.92 0.003 11/16/2016 this study Seep 0.008 0.02 5.29 0.10 0.76 2.27 0.08 1.81 3.86 0.03 0.41 0.018 1/16/2017 this study Seep 0.006 0.02 3.84 0.01 0.64 1.65 0.03 1.2 3.13 0.02 0.15 0.004 2/24/2017 this study Seep <0.01 0.02 3.87 0.00 0.73 1.68 0.00 1.54 3.27 0.02 0.11 0.001 5/10/2017 this study Seep 0.006 0.02 4.61 0.15 0.65 1.91 0.09 1.6 3.92 0.02 0.29 0.006 6/2/2017 this study Seep 0.004 0.02 4.88 0.41 0.71 2.04 0.13 1.78 3.96 0.03 0.56 0.007 6/22/2017 this study Seep 0.005 0.02 5.04 0.41 0.69 2.12 0.14 1.89 3.92 0.03 0.75 0.011 6/28/2017 this study Seep 0.004 0.02 5.11 0.25 0.73 2.16 0.10 1.85 4.1 0.03 0.59 0.006

169

Sampling Water Data Sourcea Al Ba Ca Fe K Mg Mn Na Si Sr Asg Ug Date Typeb 7/10/2017 this study Seep 0.004 0.02 5.37 0.29 0.83 2.3 0.10 1.79 4.22 0.03 0.70 0.006 9/8/2017 this study Seep <0.005 0.03 6.19 0.27 0.94 2.57 0.10 2.27 4.42 0.03 - - 12/3/2017 this study Seep ------Seep 1.6 10/26/2016 this study Seep 0.002 0.07 17.9 0.46 2.01 4.57 2.71 11.7 3.68 0.29 0.92 ND 11/16/2016 this study Seep 0.000849 0.05 16.6 0.16 1.77 4.48 2.44 12 3.57 0.29 0.82 ND 1/16/2017 this study Seep 0.005 0.06 13.1 11.10 1.05 3.51 1.54 11.3 3.45 0.24 4.32 0.003 2/24/2017 this study Seep 0.005 0.07 15.4 11.80 1.7 4.38 1.86 14.9 3.95 0.33 4.04 0.004 5/10/2017 this study Seep 0.009 0.07 16.2 5.28 1.52 4.41 1.82 13.7 4.19 0.34 2.49 0.004 6/2/2017 this study Seep 0.002 0.08 17.5 1.59 1.98 4.84 2.01 18.5 4.46 0.37 2.10 0.004 6/22/2017 this study Seep 0.002 0.08 20.3 0.23 2.14 5.67 1.93 23.9 4.17 0.43 1.69 0.004 6/28/2017 this study Seep 0.003 0.08 20.1 0.06 2.25 5.68 1.98 24.5 3.99 0.41 1.60 0.004 7/10/2017 this study Seep 0.004 0.07 18.4 0.03 1.96 5.00 1.89 18.6 3.95 0.39 1.30 0.004 7/26/2017 this study Seep 0.003 0.05 13.3 0.45 1.65 3.16 1.49 8.51 4.14 0.20 3.03 0.003 9/8/2017 this study Seep <0.005 0.08 17.4 0.27 2.23 4.66 1.81 15.9 3.86 0.36 1.66 ND 10/2/2017 this study Seep 0.001 0.09 18.8 0.91 3.19 5.16 2.49 18.1 3.89 0.38 1.85 ND 10/16/2017 this study Seep 0.005 0.09 16.3 7.45 2.43 4.35 2.41 14 4.62 0.34 4.79 ND 12/3/2017 this study Seep 0.006 0.10 17.5 17.80 1.37 5.12 2.35 19.3 3.96 0.44 5.80 0.005 SRS Seep 1.5 2/24/2017 this study Seep 0.019 0.01 4.56 0.12 0.61 0.567 0.02 69.4 4.38 0.09 1.22 0.008 HO1 10/26/2016 this study GW 0.002 0.20 27.3 0.00 0.27 9.13 0.06 12.8 5.95 1.78 ND 0.065 2/24/2017 this study GW 0.008 0.17 28.9 0.02 0.32 9.67 0.07 12.2 6.35 1.72 ND 0.04 5/10/2017 this study GW 0.006 0.18 28.5 0.02 0.33 9.34 0.08 15 6.31 1.85 ND 0.054 7/10/2017 this study GW 0.003 0.19 27.6 0.01 0.38 9.02 0.06 13.5 6.25 1.83 ND 0.053 10/19/2017 PADEP GW - - 28.3 0.04 - - - 16 - - - -

170

Sampling Water Data Sourcea Al Ba Ca Fe K Mg Mn Na Si Sr Asg Ug Date Typeb HO2 7/18/2017 this study GW 0.003 0.35 17.7 0.00 0.55 4.14 0.04 21.7 5.95 1.17 ND ND 9/8/2017 this study GW 0.002 0.34 18.5 0.00 0.38 4.53 0.05 15.7 5.99 1.02 0.20 0.019 9/8/2017 this study GW ------10/19/2017 PADEP GW - - 18.5 <0.02 - - - 21.2 - - - - HO3 6/13/2016 PADEP GW - - 5.7 1.73 - - - 23.3 - - - - 11/16/2016 this study GW 0.0010 0.08 5.27 0.00 0.37 1.56 0.16 41.1 4.38 0.31 0.52 0.013 HO4 7/12/2010 RR GW - 0.11 - ND - - - 35.1 - 0.70 - - 12/5/2011 RR GW - 0.09 - 0.10 - - - - - 0.59 - - 1/24/2012 PADEP GW ------1/30/2012 RR GW - 0.10 13.1 0.17 ND 2.72 0.08 30 - 0.66 - - 4/10/2012 RR GW - 0.12 15 0.12 ND 3.15 0.06 30 - 0.80 - - 4/26/2012 RR GW ------5/8/2012 RR GW - 0.11 14 0.12 ND 2.99 0.06 29 - 0.74 - - 5/22/2012 RR GW - 0.11 14.1 0.11 ND 2.91 0.49 29.8 - 0.77 - - 7/18/2012 RR GW - 0.11 14.3 0.58 0.6 3.04 0.05 32.5 - 0.78 - - 7/18/2012 PADEP GW ------6/14/2016 PADEP GW - - 14.6 0.15 - - - 38.6 - - - - 7/26/2017 this study GW 0.01 0.03 16.4 0.01 0.64 2.93 0.13 32.8 4.72 0.73 0.59 0.12 7/26/2017 this study GW ------7/26/2017e this study GW 0.01 0.03 16.5 0.03 0.62 2.93 0.13 33.2 4.68 0.76 0.59 0.122 HO5 10-Nov RR GW - 0.16 14.2 0.02 ND 2.91 0.02 48.2 - 0.81 - - 1/18/2012 PADEP GW ------1/30/2012 RR GW - 0.16 14 ND ND 2.79 0.02 43.4 - 1.00 - -

171

Sampling Water Data Sourcea Al Ba Ca Fe K Mg Mn Na Si Sr Asg Ug Date Typeb 3/8/2012 PADEP GW ------4/10/2012 RR GW - 0.16 13.2 ND ND 2.6 0.02 48.2 - 0.94 - - 4/26/2012 RR GW ------5/8/2012 RR GW - 0.14 12.7 ND ND 2.54 0.02 46.1 - 0.89 - - 5/22/2012 RR GW - 0.15 12.7 ND ND 2.44 0.02 47.2 - 0.91 - - 7/18/2012 RR GW - 0.14 12.5 0.14 0.58 2.49 0.02 44.4 - 0.84 - - 7/18/2012 PADEP GW ------HO6 4/13/2011 RR GW - 0.05 - 0.02 - - - 68.1 - 0.20 - - 11/21/2011 RR GW - 0.10 - 1.07 - - - 75 - 0.22 - - 1/25/2012 PADEP GW ------1/30/2012 RR GW - 0.12 1.99 1.60 0.65 0.73 0.05 71.6 - 0.24 - - 3/8/2012 PADEP GW ------4/10/2012 RR GW - 0.14 1.92 0.96 2.53 0.96 0.05 72.3 - 0.23 - - 4/26/2012 RR GW ------5/8/2012 RR GW - 0.11 1.9 1.74 0.97 0.75 0.05 70.8 - 0.22 - - 5/22/2012 RR GW - 0.13 1.87 3.02 1.3 0.99 0.06 70.6 - 0.23 - - 7/18/2012 RR GW - 0.13 1.87 2.18 2.94 0.99 0.05 68.8 - 0.21 - - 7/18/2012 RR GW ------6/16/2016 PADEP GW - - 1.58 0.53 - - - 67.7 - - - - HO7 11/5/2010 RR GW ------2/8/2012 RR GW ------HO8-A 11/2/2010 RR GW ------11/10/2010 RR GW ------2/10/2012 RR GW ------

172

Sampling Water Data Sourcea Al Ba Ca Fe K Mg Mn Na Si Sr Asg Ug Date Typeb HO8-B 11/2/2010 RR GW ------11/10/2010 RR GW ------2/10/2012 RR GW ------HO9 11/2/2010 RR GW ------2/7/2012 RR GW ------H10 11/9/2010 RR GW ------2/7/2012 RR GW ------SR 1.5 Piezo Heilweil et al. (2014; 11/12/2013 GW <0.01 0.17 13.4 0.38 0.53 2.83 0.13 29.7 2.5 0.90 - - 2015) a RR: Range Resources; PADEP: Pennsylvania Department of Environmental Protectoin b SW: surface water; GW: groundwater; Seep: water discharging from seepage c Stream: Heilweil et al. (2014) d ND: not detected; -: not measured eSample was not filtered before acidification. See text under "Methane Impacts on Groundwater" fHydrocarbon concentrations and carbon isotopic ratios were analyzed by Isotech Laboratories, Inc. g µg/L

173

Table E7. Anions, YSI measurements and Sr isotopes - This table appears almost identical to that which was published in SI for Woda et al. (2018)

Sampling Data Water NO3 DO SPC Temp 87 86 Cl (mg/L) SO4 (mg/L) Br (mg/L) pH ORP Sr/ Sr Date Sourcea Typeb (mg/L) (%) (µS/cm) (ᵒC) SR 0.4 7/10/2017 this study SW ------SR 0.5 7/10/2017 this study SW ------SR 1.0 Heilweil et 5/21/2013 al. (2014; SW - - - - 7.2 - 111 - 19.5 - 2015) Heilweil et 6/27/2013 al. (2014; SW - - - - 7.3 87 136 - 19.5 - 2015) Heilweil et 11/12/2013 al. (2014; SW 10.7 10.8 0.09 5.66 7.6 110 156 - 5.1 0.71342 2015) Wendt et al. 7/9/2015 SW - - - - 7.4 - - - 17.1 - (2018) SR 1.1 Wendt et al. 6/13/2016 SW ------(2018) 8/15/2016 this study SW ------9/15/2016 this study SW 8.85 10 0.09 4.97 ------10/26/2016 this study SW ------11/16/2016 this study SW 8.61 9.68 0.06 6.91 ------1/16/2017 this study SW 7.71 9.43 0.07 9.94 ------2/24/2017 this study SW 8.08 10.1 0.05 7.79 8.1 98 103 - - - 5/10/2017 this study SW 4.58 8.89 0.13 6.73 7.3 86 - - 12.7 - 6/22/2017 this study SW 7.13 8.67 0.09 6.63 7.5 100 147 241 17.6 -

174

Sampling Data Water NO3 DO SPC Temp 87 86 Cl (mg/L) SO4 (mg/L) Br (mg/L) pH ORP Sr/ Sr Date Sourcea Typeb (mg/L) (%) (µS/cm) (ᵒC) 7/10/2017 this study SW 8.1 8.64 0.05 5.29 7.6 85 180 - 19.7 - 7/26/2017 this study SW - - - - 7.3 110 128 177 17.4 - 9/8/2017 this study SW 8.14 8.67 0.06 4.88 7.8 91 174 211 15.3 - 10/2/2017 this study SW 10.2 8.64 0.08 4.42 7.1 93 177 214 14.1 - 12/3/2017 this study SW 7.32 9.75 0.10 9.47 7.1 - 159 189 3.5 - SR 1.15 Wendt et al. 7/9/2015 SW ------(2018) 8/15/2016 this study SW ------9/15/2016 this study SW ------10/26/2016 this study SW ------SR 1.2 Heilweil et 11/12/2013 al. (2014; SW 10.5 10.8 <0.09 5.67 7.7 110 155 - 4.8 0.71353 2015) Wendt et al. 7/9/2015 SW - - - - 6.8 - - - 17.4 - (2018) 6/27/2016 this study SW ------8/15/2016 this study SW ------9/15/2016 this study SW ------11/16/2016 this study SW ------1/16/2017 this study SW ------2/24/2017 this study SW ------5/10/2017 this study SW ------6/22/2017 this study SW ------SR 1.4 Heilweil et 11/12/2013 al. (2014; SW 10.5 10.8 <0.09 5.73 7.6 110 155 - 4.8 0.71362 2015)

175

Sampling Data Water NO3 DO SPC Temp 87 86 Cl (mg/L) SO4 (mg/L) Br (mg/L) pH ORP Sr/ Sr Date Sourcea Typeb (mg/L) (%) (µS/cm) (ᵒC) 12/9/2014 this study SW ------Wendt et al. 7/9/2015 SW ------(2018) 8/15/2016 this study SW ------9/15/2016 this study SW ------10/26/2016 this study SW ------11/16/2017 this study SW ------1/16/2017 this study SW ------2/24/2017 this study SW ------5/10/2017 this study SW ------6/22/2017 this study SW ------SR 1.5 Heilweil et 6/27/2013 al. (2014; SW - - - - 7.5 97 136 - 21.1 - 2015) Heilweil et 11/12/2013 al. (2014; SW 10.5 11.1 <0.09 5.61 7.6 110 154 - 4.9 0.71367 2015) Wendt et al. 12/9/2014 SW ------(2018) Wendt et al. 7/9/2015 SW ------(2018) Wendt et al. 6/13/2016 SW ------(2018) 8/15/2016 this study SW ------9/15/2016 this study SW 8.86 10.2 0.08 5.47 ------10/26/2016 this study SW 9.71 9.88 0.07 2.9 ------11/16/2016 this study SW 8.54 9.76 0.06 7 ------1/16/2017 this study SW ------2/24/2017 this study SW 8.26 10.1 0.06 8.04 8.2 100 104 - 10.2 -

176

Sampling Data Water NO3 DO SPC Temp 87 86 Cl (mg/L) SO4 (mg/L) Br (mg/L) pH ORP Sr/ Sr Date Sourcea Typeb (mg/L) (%) (µS/cm) (ᵒC) 5/10/2017 this study SW ------6/22/2017 this study SW ------7/10/2017 this study SW ------9/8/2017 this study SW ------10/2/2017 this study SW ------SR 1.55 Wendt et al. 7/9/2015 SW ------(2018) 8/15/2016 this study SW ------9/15/2016 this study SW 8.9 10.2 0.08 5.61 ------10/26/2016 this study SW 9.61 9.81 0.08 2.67 ------11/16/2016 this study SW ------1/16/2017 this study SW ------2/24/2017 this study SW ------5/10/2017 this study SW ------6/22/2017 this study SW ------9/8/2017 this study SW ------SR 1.6 Heilweil et 11/12/2013 al. (2014; SW 10.7 10.8 <0.09 5.99 7.7 110 156 - 5.1 0.71367 2015) Wendt et al. 12/9/2014 SW ------(2018) Wendt et al. 7/9/2015 SW ------(2018) Wendt et al. 6/13/2016 SW ------(2018) 8/15/2016 this study SW ------9/15/2016 this study SW ------

177

Sampling Data Water NO3 DO SPC Temp 87 86 Cl (mg/L) SO4 (mg/L) Br (mg/L) pH ORP Sr/ Sr Date Sourcea Typeb (mg/L) (%) (µS/cm) (ᵒC) SR 1.8 Heilweil et 11/12/2013 al. (2014; SW 10.7 9.93 <0.08 5.75 7.6 15 155 - 5 0.71365 2015) Wendt et al. 7/9/2015 SW ------(2018) Wendt et al. 6/13/2016 SW ------(2018) 8/15/2016 this study SW ------9/15/2016 this study SW ------10/26/2016 this study SW ------1/16/2017 this study SW ------SR 2 Heilweil et 5/21/2013 al. (2014; SW - - - - 7.5 - 111 - 20.9 - 2015) Heilweil et 6/27/2013 al. (2014; SW - - - - 7.7 99 136 - 22.1 - 2015) Heilweil et 11/12/2013 al. (2014; SW 10.7 10.9 <0.09 6.12 7.6 120 154 - 5 - 2015) Wendt et al. 7/9/2015 SW ------(2018) Wendt et al. 6/13/2016 SW ------(2018) 9/15/2016 this study SW ------10/26/2016 this study SW ------1/16/2017 this study SW ------2/24/2017 this study SW ------

178

Sampling Data Water NO3 DO SPC Temp 87 86 Cl (mg/L) SO4 (mg/L) Br (mg/L) pH ORP Sr/ Sr Date Sourcea Typeb (mg/L) (%) (µS/cm) (ᵒC) 6/22/2017 this study SW ------7/10/2017 this study SW ------SR 3 Heilweil et 5/21/2013 al. (2014; SW - - - - 7.7 - 112 - 21 - 2015) Heilweil et 6/27/2013 al. (2014; SW - - - - 7.5 98 138 - 21.5 - 2015) 7/10/2017 this study SW ------SR 4 Heilweil et 5/21/2013 al. (2014; SW - - - - 7.6 - 113 - 20.9 - 2015) 2/24/2017 this study SW 8.56 10.3 0.07 8.51 7.8 91 108 - 11.7 - SR 6 Heilweil et 5/21/2013 al. (2014; SW - - - - 7.5 - 118 - 20.2 - 2015) SR 7 Heilweil et 5/21/2013 al. (2014; SW - - - - 7.5 - 118 - 20.1 - 2015) SR 7.5 2/24/2017 this study SW 8.68 10.2 0.05 8.83 7.5 - 109 - 11 - SR 8 7/10/2017 this study SW ------Seep 1.5

179

Sampling Data Water NO3 DO SPC Temp 87 86 Cl (mg/L) SO4 (mg/L) Br (mg/L) pH ORP Sr/ Sr Date Sourcea Typeb (mg/L) (%) (µS/cm) (ᵒC) Heilweil et 11/12/2013 al. (2014; Seep 2.23 10.4 <0.09 <0.51 5.8 20.3 62 - 9.8 0.71484 2015) Wendt et al. 6/13/2016 Seep ------(2018) 8/15/2016 this study Seep ------9/15/2016 this study Seep 2.11 7.9 0.06 0.17 ------10/26/2016 this study Seep 2.81 7.74 0.07 0.20 ------11/16/2016 this study Seep 2.19 8.96 0.08 0.54 ------1/16/2017 this study Seep 1.64 9.49 0.04 1.01 ------2/24/2017 this study Seep 1.17 9.93 0.06 0.28 6.9 44 51 - 6.7 - 5/10/2017 this study Seep 1.06 8.16 - 0.57 6.6 40 - - 11.3 - 6/22/2017 this study Seep 1.19 7.18 0.06 0.51 6.4 21 105 -13 15.2 - 6/28/2017 this study Seep 1.29 7.96 0.07 0.55 6.4 14 98 -68 14.8 - 7/10/2017 this study Seep 1.35 6.88 0.05 0.54 6.4 12 114 -35 17.1 - 9/8/2017 this study Seep 2.49 6.44 0.08 1.55 6.7 35 115 13 - - 10/2/2017 this study Seep 5.34 3.01 0.05 0.57 6.4 26 119 2 12.2 - 10/16/2017 this study Seep 2.25 7.13 0.06 0.61 6.1 32 107 13 14.8 - 12/3/2017 this study Seep 2.21 9.96 - 0.75 6.5 - 74 117 6.8 0.71417 Seep 1.55 9/16/2016 this study Seep 2.12 9.89 0.07 0.18 ------10/26/2016 this study Seep 2.89 9.69 0.07 ------11/16/2016 this study Seep 2.55 10.1 0.05 0.53 ------1/16/2017 this study Seep 1.36 9.45 0.04 1.88 ------2/24/2017 this study Seep 1.22 10.3 - 0.56 6.3 42 52 - 5.5 - 5/10/2017 this study Seep 1.05 8.89 - 0.48 6.1 28 - - 10.5 - 6/2/2017 this study Seep 1.48 8.74 0.09 0.56 ------6/22/2017 this study Seep 1.19 8.89 0.05 0.48 ------

180

Sampling Data Water NO3 DO SPC Temp 87 86 Cl (mg/L) SO4 (mg/L) Br (mg/L) pH ORP Sr/ Sr Date Sourcea Typeb (mg/L) (%) (µS/cm) (ᵒC) 6/28/2017 this study Seep 1.21 8.7 0.07 0.48 5.9 37 58 90 13.9 - 7/10/2017 this study Seep 1.24 7.98 0.06 0.49 6 37 76 94 15.1 - 9/8/2017 this study Seep 1.69 7.79 0.07 0.62 6.2 36 83 93 15.5 - 12/3/2017 this study Seep - - - - 6.2 - 70 169 9.4 - Seep 1.6 10/26/2016 this study Seep 22.7 1.27 0.12 0.48 ------11/16/2016 this study Seep 25.4 1.07 0.08 0.49 ------1/16/2017 this study Seep 25.5 6.16 0.08 1.29 ------2/24/2017 this study Seep 41.2 3.02 0.10 0.30 6.6 49 242 - 7.1 - 5/10/2017 this study Seep 26.8 3.49 0.12 0.50 6.4 11 - - 11.2 - 6/2/2017 this study Seep 32.3 2.78 0.13 0.55 ------6/22/2017 this study Seep 46.6 1.81 0.10 0.51 6.5 27 283 55 17.9 - 6/28/2017 this study Seep 45.7 2.21 0.10 0.54 6.5 14 289 12 14.8 - 7/10/2017 this study Seep 38.4 2.02 0.09 0.50 6.4 15 296 34 17.4 - 7/26/2017 this study Seep 12.2 5.52 0.09 0.66 6.3 20 161 31 18.5 - 9/8/2017 this study Seep 30.7 1.66 0.11 0.71 6.5 25 257 6 14.5 - 10/2/2017 this study Seep 35.1 0.843 0.10 0.49 6.5 32 - - - - 10/16/2017 this study Seep 21.5 0.709 0.09 0.51 6.5 19 224 -90 14.5 - 12/3/2017 this study Seep 53.6 1.51 0.41 0.32 6.2 - 425 -40 5.5 0.71168

SRS Seep 1.5

2/24/2017 this study Seep 23.6 5.41 0.08 0.56 9.1 31 300 - 8.8 - HO1 10/26/2016 this study GW 2.14 22.9 0.14 0.83 ------2/24/2017 this study GW 2.16 26.6 0.09 0.39 7.8 10 269 - 12 - 5/10/2017 this study GW 1.87 23.9 0.11 0.57 7.7 8 - - 12.4 - 7/10/2017 this study GW 1.87 21.6 0.09 0.55 7.6 9 304 -210 14.7 -

181

Sampling Data Water NO3 DO SPC Temp 87 86 Cl (mg/L) SO4 (mg/L) Br (mg/L) pH ORP Sr/ Sr Date Sourcea Typeb (mg/L) (%) (µS/cm) (ᵒC) 10/19/2017 PADEP GW 1.5 20.7 ------HO2 7/18/2017 this study GW 3.3 5.86 0.08 0.57 8.1 10 206 -220 12.8 - 9/8/2017 this study GW 3.56 3.98 0.09 0.66 8.1 13 210 -277 12.1 - 9/8/2017 this study GW - - - - 8.1 13 210 -277 12.1 - 10/19/2017 PADEP GW 3.28 3.46 ------HO3 6/13/2016 PADEP GW 6.36 2.34 ------11/16/2016 this study GW 13.3 1 0.14 0.44 8.9 - - - - - HO4 7/12/2010 RR GW 7.77 16.6 - - 8.7 - - - - - 12/5/2011 RR GW 5.46 15.3 - - 8.5 - - - - - 1/24/2012 PADEP GW ------1/30/2012 RR GW 6.1 18.3 - - 8.2 - 201 - - - 4/10/2012 RR GW 6.1 18.2 - - 8 - 224 - - - 4/26/2012 RR GW ------5/8/2012 RR GW 6.4 18.7 - - 8.2 - 224 - - - 5/22/2012 RR GW 6.8 17.4 - - 8.3 - 209 - - - 7/18/2012 RR GW 6.6 18.7 - - 8.2 - 236 - - - 7/18/2012 PADEP GW ------6/14/2016 PADEP GW 9.65 8.21 ------7/26/2017 this study GW 14.4 5.01 0.13 0.56 8.1 13 276 -146 12.1 - 7/26/2017 this study GW ------7/26/2017e this study GW ------HO5 10-Nov RR GW 10.1 21 - - 8.2 - 289 - - - 1/18/2012 PADEP GW ------

182

Sampling Data Water NO3 DO SPC Temp 87 86 Cl (mg/L) SO4 (mg/L) Br (mg/L) pH ORP Sr/ Sr Date Sourcea Typeb (mg/L) (%) (µS/cm) (ᵒC) 1/30/2012 RR GW 9.5 21.8 - - 8 - 259 - - - 3/8/2012 PADEP GW ------4/10/2012 RR GW 10 18 - - 8 - 289 - - - 4/26/2012 RR GW ------5/8/2012 RR GW 9.8 18.3 - - 7.9 - 284 - - - 5/22/2012 RR GW 10 20.6 - - 8.4 - 265 - - - 7/18/2012 RR GW 7.7 21.2 - - 8.1 - 274 - - - 7/18/2012 PADEP GW ------HO6 4/13/2011 RR GW 10.7 0.5 - - 9.2 - 289 - - - 11/21/2011 RR GW 10.2 0.4 - - 9.3 - 310 - - - 1/25/2012 PADEP GW ------1/30/2012 RR GW 11.2 ND - - 9 - - - - - 3/8/2012 PADEP GW ------4/10/2012 RR GW 10.8 ND - - 9.1 - 302 - - - 4/26/2012 RR GW ------288 - - - 5/8/2012 RR GW 10.2 ND - - 9 - 303 - - - 5/22/2012 RR GW 10.8 ND - - 9.2 - - - - - 7/18/2012 RR GW 9.8 ND - - 9 - - - - - 7/18/2012 RR GW ------6/16/2016 PADEP GW 10.81 ND ------HO7 11/5/2010 RR GW ------2/8/2012 RR GW ------HO8-A 11/2/2010 RR GW ------11/10/2010 RR GW ------

183

Sampling Data Water NO3 DO SPC Temp 87 86 Cl (mg/L) SO4 (mg/L) Br (mg/L) pH ORP Sr/ Sr Date Sourcea Typeb (mg/L) (%) (µS/cm) (ᵒC) 2/10/2012 RR GW ------HO8-B 11/2/2010 RR GW ------11/10/2010 RR GW ------2/10/2012 RR GW ------HO9 11/2/2010 RR GW ------2/7/2012 RR GW ------H10 11/9/2010 RR GW ------2/7/2012 RR GW ------

SR 1.5 Piezo

Heilweil et 11/12/2013 al. (2014; GW 14.4 0.62 <0.09 <0.51 8 68 226 - 9.8 0.71141 2015) a RR: Range Resources; PADEP: Pennsylvania Department of Environmental Protectoin b SW: surface water; GW: groundwater; Seep: water discharging from seepage c Stream: Heilweil et al. (2014) d ND: not detected; -: not measured eSample was not filtered before acidification. See text under "Methane Impacts on Groundwater" fHydrocarbon concentrations and carbon isotopic ratios were analyzed by Isotech Laboratories, Inc. gµg/L

184

Tables for Chapter 3

Table E8. Measured hydrocarbons in waters sampled at wellheads of leaking or abandoned wells cited in Chapter 3

13 13 KOH δ C- C2H6 δ C- Sampling Date Site name (Y/N) Lat Long Sampler CH4 (µg/L) CH4 (µg/L) C2H6 C1/C2

ANF 4/9/2018 ANF Leaking Well 1 N 41.5095 -78.7926 PSU 23200 -49.1 1830 -27.4 13 4/9/2018 ANF Leaking Well 2 N 41.6121 -78.7913 PSU - - - - - 4/9/2018 ANF Leaking Well 3 N 41.4964 78.7945 PSU 7500 -38.7 254 - 30 7/10/2018 ANF Leaking Well 4 N 41.7724 -78.6817 PSU 19800 -44.2 46.3 - 428 10/29/2018 ANF Leaking Well 5 Y 41.7701 -78.8621 PSU 7700 - 1580 - 6 10/29/2018 ANF Leaking Well 5 N 41.7701 -78.8621 PSU 7200 - 1294 - 5 10/29/2018 ANF Leaking Well 6 Y 41.6379 -78.8524 PSU 309 - - - - TeenShale Network 12/1/2017 Wallace Run well N 40.9766 -77.9005 TeenShalea 9800 -44.3 130 -34.1 75 2/27/2018 Wallace Run well N 40.9766 -77.9005 PSU + TeenShalea 8500 -39.4 234 -38.4 36 Grieve (2014)b 5/29/2012 Grindstone Hollow Y 40.98 -77.88 Grieve (2014)b - -37.0 - -40.5 - a State College High School watershed group led by Jennifer Williams (Penn State) and Eugene Ruocchio (State College High School) b Reference (87)

185

Table E9. Cation concentrations in waters sampled at wellheads of leaking or abandoned wells cited in Chapter 3

Filtered Date Site (Y/N) Al Ba Ca Fe K Mg Mn Na P Si Sr Ti CrC AsC UC

ANF 4/9/2018 ANF Leaking well 1 Y <0.005 2.20 9.7 0.02 2.05 1.67 0.03 80.3 0.05 3.64 0.20 <0.005 0.051 1.2 0.003 4/9/2018 ANF Leaking well 2 Y <0.005 1.58 27.4 0.02 2.56 4.03 0.05 54.0 <0.02 4.22 0.44 <0.005 0.048 2.0 0.002 4/9/2018 ANF Leaking well 3 Y <0.005 1.16 17.4 0.11 3.03 2.88 0.03 96.2 0.02 3.77 0.33 <0.005 0.040 2.8 0.002 7/10/2018 ANF Leaking well 4 Y <0.01 0.67 18.2 0.30 1.93 3.10 0.12 92.7 0.05 3.04 0.37 <0.005 0.050 4.5 <0.001 10/29/2018 ANF Leaking well 5 Y <0.005 0.93 16.0 0.12 0.72 2.46 0.02 112 0.01 0.05 0.43 <0.005 0.072 10.5 <0.001 10/29/2018 ANF Leaking well 6 Y 0.030 0.40 32.9 0.57 1.62 3.90 0.15 11.8 0.04 3.41 0.22 <0.005 0.215 <1.0 <0.001 10/29/2018 ANF Leaking well 6 N <0.005 0.41 32.7 0.57 1.56 3.90 0.16 11.8 0.03 3.39 0.22 <0.005 - - - TeenShale Sitesa 12/1/2017 Wallace Run well Y <0.005 0.49 15.5 0.02 2.39 4.21 0.07 97.5 0.04 3.98 0.54 <0.005 - - - 2/27/2018 Wallace Run well Y <0.02 0.49 15.2 0.02 2.30 4.09 0.07 103 0.04 4.05 0.52 <0.005 0.039 19.2 0.159 Sites Reported by Grieve (2014)b 5/29/2012 Grindstone Hollow Y <0.01 0.62 21.9 <0.01 2.47 5.99 0.08 127 <0.05 3.92 0.90 - - - - a State College High School watershed group led by Jennifer Williams (Penn State) and Eugene Ruocchio (State College High School) b Reference (87) C Analyte measured in ICP-MS

186

Table E10. Anion concentrations and YSI measurements in waters sampled at leaking or abandoned wells cited in Chapter 3

DO SPC Temp Date Site F CL SO4 Br NO3 pH % (µS/cm) ORP ᵒC

ANF 4/9/2018 ANF Leaking well 1 0.60 20.6 2.28 0.48 0.60 8.09 0.6 691 -107 8.6 4/9/2018 ANF Leaking well 2 0.31 38.4 0.85 0.79 0.47 7.79 0.8 721 -81 8.8 4/9/2018 ANF Leaking well 3 0.66 55.0 0.75 1.18 0.52 7.87 3.6 930 -115 8.9 7/10/2018 ANF Leaking well 4 0.56 102 0.37 2.09 0.42 7.49 14.6 581 -31 11.3 10/29/2018 ANF Leaking well 5 0.39 3300 5.80 - - 5.99 32 10500 23 10.6 10/29/2018 ANF Leaking well 6 0.24 14.1 9.25 - - 6.42 31.4 252 -80 8.6 TeenShale Sitesa 12/1/2017 Wallace Run well 0.55 99.0 2.32 2.15 0.10 7.95 - - - - 2/27/2018 Wallace Run well - - - - - 7.95 - - - - Sites Reported by Grieve (2014)b 5/29/2012 Grindstone Hollow 1.98 157 3.29 0.50 1.31 8.32 - - - 23.3 a State College High School watershed group led by Jennifer Williams (Penn State) and Eugene Ruocchio (State College High School) b Reference (87)

187

Table E11. Measured hydrocarbons in GLD and Salt Springs as cited in Chapters 2 and 3 (not including LAWs)

KOH CH4 δ13C- C2H6 δ13C- C1/C2 Date Site (Y/N) Lat Long Sampler (µg/L) CH4 (µg/L) C2H6 molar

Sugar Run (Lycoming) 11/12/2013 Seep 1.5 N 41.2395 -76.6922 Heilweil 2290 -26.7 34 -32.6 126 6/13/2016 Seep 1.5 N 41.2395 -76.6922 PSU 216 - - - - 8/15/2016 Seep 1.5 N 41.2395 -76.6922 PSU 117 - - - - 9/15/2016 Seep 1.5 N 41.2395 -76.6922 PSU 830 -52.3 - - - 10/26/2016 Seep 1.5 N 41.2395 -76.6922 PSU 4300 -18.0 - - - 11/16/2016 Seep 1.5 N 41.2395 -76.6922 PSU 3800 -21.3 - - - 1/16/2017 Seep 1.5 N 41.2395 -76.6922 PSU 1.3 -2.0 - - - 2/24/2017 Seep 1.5 N 41.2395 -76.6922 PSU 355 3.4 - - - 5/10/2017 Seep 1.5 N 41.2395 -76.6922 PSU 1260 -20.7 - - - 6/22/2017 Seep 1.5 N 41.2395 -76.6922 PSU 557 -10.4 - - - 6/28/2017 Seep 1.5 N 41.2395 -76.6922 PSU 1800 - - - - 7/10/2017 Seep 1.5 N 41.2395 -76.6922 PSU 3170 -16.9 16.6 - 359 9/8/2017 Seep 1.5 N 41.2395 -76.6922 PSU 2040 -24.0 - - 10/2/2017 Seep 1.5 N 41.2395 -76.6922 PSU 4410 - 40.2 - - 10/16/2017 Seep 1.5 N 41.2395 -76.6922 PSU 1600 - - - - 12/3/2017 Seep 1.5 N 41.2395 -76.6922 PSU 322 -21 - - - 10/26/2016 Seep 1.55 N 41.2398 -76.6921 PSU 19.5 37.2 - - - 11/16/2016 Seep 1.55 N 41.2398 -76.6921 PSU 170 13.0 - - - 1/16/2017 Seep 1.55 N 41.2398 -76.6921 PSU 0 37.6 - - - 2/24/2017 Seep 1.55 N 41.2398 -76.6921 PSU 0 -3.8 - - - 5/10/2017 Seep 1.55 N 41.2398 -76.6921 PSU 380 -17.2 - - - 6/2/2017 Seep 1.55 N 41.2398 -76.6921 PSU 630 - - - - 6/22/2017 Seep 1.55 N 41.2398 -76.6921 PSU 580 -18.3 - - - 6/28/2017 Seep 1.55 N 41.2398 -76.6921 PSU 320 - - - -

188

KOH CH4 δ13C- C2H6 δ13C- C1/C2 Date Site (Y/N) Lat Long Sampler (µg/L) CH4 (µg/L) C2H6 molar 12/3/2017 Seep 1.55 N 41.2398 -76.6921 PSU 8.1 26.7 - - - 1/16/2017 Seep 1.53 N 41.2397 -76.6921 PSU 55.3 30.4 - - - 2/24/2017 Seep 1.53 N 41.2397 -76.6921 PSU 360 3.4 - - - 5/10/2017 Seep 1.53 N 41.2397 -76.6921 PSU 1610 -25.6 - - - 6/22/2017 Seep 1.53 N 41.2397 -76.6921 PSU 1530 -18.9 - - - 6/28/2017 Seep 1.53 N 41.2397 -76.6921 PSU 32.6 - - - - 7/26/2017 Seep 1.53 N 41.2397 -76.6921 PSU 1060 - - - - 10/26/2016 Seep 1.6 N 41.2400 -76.6915 PSU 4570 -25.4 - - - 11/16/2016 Seep 1.6 N 41.2400 -76.6915 PSU 6010 -26.7 - - - 1/16/2017 Seep 1.6 N 41.2400 -76.6915 PSU 5850 -25 - - - 2/24/2017 Seep 1.6 N 41.2400 -76.6915 PSU 8590 -28.3 - - - 5/10/2017 Seep 1.6 N 41.2400 -76.6915 PSU 6750 -25.8 - - - 6/2/2017 Seep 1.6 N 41.2400 -76.6915 PSU 4220 - - - - 6/22/2017 Seep 1.6 N 41.2400 -76.6915 PSU 2500 -22.5 - - - 6/28/2017 Seep 1.6 N 41.2400 -76.6915 PSU 1690 - - - - 7/10/2017 Seep 1.6 N 41.2400 -76.6915 PSU 981 4.9 - - - 7/26/2017 Seep 1.6 N 41.2400 -76.6915 PSU 2850 -23.5 - - - 9/8/2017 Seep 1.6 N 41.2400 -76.6915 PSU 1450 -18.4 - - - 10/2/2017 Seep 1.6 N 41.2400 -76.6915 PSU 1670 - 5.4 - 578 10/16/2017 Seep 1.6 N 41.2400 -76.6915 PSU 3340 - - - - 12/3/2017 Seep 1.6 N 41.2400 -76.6915 PSU 6570 -28 60.4 - 204 ANF Seepages 10/23/2017 Ludlow Seepage N 41.7120 -78.9120 PSU 11200 -33.5 735 -30.6 29 10/24/2017 Ludlow Seepage N 41.7120 -78.9120 PSU 10600 -41.8 803 -31.0 25 4/9/2018 Ludlow Seepage N 41.7120 -78.9120 PSU 10300 -36.6 660 -28.2 29 4/9/2018 Ludlow Seepage Y 41.7120 -78.9120 PSU 9540 -42.8 - - - 7/11/2018 Ludlow Seepage N 41.7120 -78.9120 PSU 8650 -45.4 782 - 21

189

KOH CH4 δ13C- C2H6 δ13C- C1/C2 Date Site (Y/N) Lat Long Sampler (µg/L) CH4 (µg/L) C2H6 molar 10/29/2018 Ludlow Seepage Y 41.7120 -78.9120 PSU 8910 - 778 - 21 10/29/2018 Ludlow Seepage N 41.7120 -78.9120 PSU 7370 - 66.3 - 208 4/10/2018 Oil Well discharge N 41.7403 -79.0823 PSU 750 - - - - Morrison Creek 4/10/2018 Seepage N 41.7915 -79.1703 Volunteer 3200 -62.6 1664 -34.7 4 Morrison Creek 7/10/2018 Seepage N 41.7915 -79.1703 PSU 4210 -59.4 2455 -35.7 3 Morrison Creek 10/29/2018 Seepage Y 41.7915 -79.1703 PSU 3940 - 2308 - 3 Morrison Creek 10/29/2018 Seepage N 41.7915 -79.1703 PSU 3420 - 1496 - 4 10/29/2018 Morrison Road Seepage Y 41.8134 -79.1231 PSU 271 - 66.3 - 8 7/10/2018 EBTC Seepage N 41.6841 -78.9180 PSU 4724 -44.3 2380 4 10/29/2018 Blood Run Seep A N 41.6915 -78.9219 PSU 110 - 0.58 - 355 10/29/2018 Blood Run Seep A Y 41.6915 -78.9219 PSU 282 - 18.8 - 28 10/29/2018 Blood Run Seep B Y 41.6912 -78.9217 PSU 550 - - - - 10/30/2018 Iron Pool seepage N 41.8169 -78.6167 Volunteer 5960 - - - - 11/11/2018 Maple Run Seep N 41.5271 -78.8597 Volunteer 398 - - - - Moshannon Forest 4/13/2017 LR 1 seep N 41.1850 -78.5161 PSU & volunteer 10.7 - - - - 4/13/2017 LR 1.5 Seep N 41.1848 -78.5160 PSU & volunteer 61.2 - - - - 4/13/2017 LR 2A Seep N 41.1808 -78.5192 PSU & volunteer 262 - - - - 9/18/2017 LR 2B Seep N 41.1808 -78.5192 PSU & volunteer 1110 -54.1 8.1 -34.7 258 9/18/2017 LR 3A seep N 41.1808 -78.5192 PSU & volunteer 143 -47.6 - - - 9/18/2017 LR 3B Seep N 41.1808 -78.5192 PSU & volunteer 27.5 -65.7 - - - 4/13/2017 LR 4 Seep N 41.1823 -78.5164 PSU & volunteer 338 - - - - 9/18/2017 LR 4 Seep N 41.1823 -78.5164 PSU & volunteer 2580 -64.4 - - - 4/13/2017 LR Discharge pipe N 41.1670 -78.5394 PSU & volunteer 789 - - - -

190

KOH CH4 δ13C- C2H6 δ13C- C1/C2 Date Site (Y/N) Lat Long Sampler (µg/L) CH4 (µg/L) C2H6 molar 9/18/2017 LR Discharge pipe N 41.1670 -78.5394 PSU & volunteer 3130 -57.20 6.6 -31.4 883 Gas-rich Seep (Natural) 7/17/2017 Salt Spring State Park N 41.9109 -75.8647 PSU 16100 -44.3 - - -

191

Table E12: Measured cations in GLD and Salt Springs as cited in Chapters 2 and 3 (not including LAWs). Data from Seeps in Sugar Run were reported previously. Filt Date Site (Y/N) Al Ba Ca Fe K Mg Mn Na P Si Sr Cra Asa Ua Sugar Run (Lycoming) 11/12/2013 Seep 1.5 Y <0.01 0.025 5.1 0.1 0.8 2.2 0.1 1.4 0.1 4.0 0.026 - - - 6/13/2016 Seep 1.5 Y ------8/15/2016 Seep 1.5 Y ------9/15/2016 Seep 1.5 Y <0.005 0.03 8.0 0.4 1.2 2.9 0.6 2.5 <0.06 4.5 0.5 0.025 1.6 <0.001 10/26/2016 Seep 1.5 Y <0.005 0.02 8.1 0.1 1.0 3.3 <0.005 2.2 <0.12 4.1 0.056 0.036 1.8 0.0 11/16/2016 Seep 1.5 Y 0.02 0.03 7.8 0.1 1.0 2.9 0.6 2.1 0.0 3.8 0.048 <0.046 0.5 <0.001 1/16/2017 Seep 1.5 Y 0.01 0.02 3.8 0.3 0.6 1.6 0.2 1.2 0.1 3.1 0.02 <0.046 0.3 0.006 2/24/2017 Seep 1.5 Y <0.01 0.01 4.4 0.3 0.6 1.8 0.3 1.6 <0.03 3.3 0.02 0.066 0.7 0.014 5/10/2017 Seep 1.5 Y 0.01 0.02 5.5 1.1 0.8 2.2 0.5 1.5 <0.11 3.8 0.03 0.085 2.4 0.016 6/22/2017 Seep 1.5 Y 0.01 0.02 6.5 1.3 0.9 2.6 0.4 1.9 <0.07 4.2 0.03 0.084 2.6 0.031 6/28/2017 Seep 1.5 Y 0.01 0.03 7.8 4.2 1.0 3.0 0.7 2.1 <0.07 4.5 0.05 0.137 5.8 0.010 7/10/2017 Seep 1.5 Y <0.005 0.03 8.1 5.3 1.1 3.1 0.7 2.0 0.1 4.6 0.06 0.082 7.7 0.008 9/8/2017 Seep 1.5 Y 0.01 0.04 8.6 5.4 1.3 3.1 0.7 3.3 <0.05 4.6 0.06 0.133 7.2 0.010 10/2/2017 Seep 1.5 Y <0.005 0.04 10.0 5.5 1.2 3.5 0.9 3.7 0.1 4.3 0.07 0.077 6.4 <0.001 10/16/2017 Seep 1.5 Y <0.005 0.04 8.8 5.0 1.2 3.2 0.7 2.6 0.1 4.8 0.05 0.076 6.3 <0.001 12/3/2017 Seep 1.5 Y <0.005 0.02 4.7 0.5 0.6 1.9 0.1 1.8 <0.12 3.3 0.05 0.125 <1.0 0.007 4/23/2018 Seep 1.5 Y 0.01 0.02 4.7 0.4 0.7 1.9 0.1 1.8 <0.01 3.3 0.03 0.087 1.4 0.007 9/16/2016 Seep 1.55 Y <0.005 0.02 5.5 0.0 0.9 2.4 <0.005 2.1 <0.06 4.3 0.03 0.035 0.5 0.002 10/26/2016 Seep 1.55 Y <0.005 0.02 5.3 0.3 1.0 2.4 0.2 1.8 <0.12 3.9 0.03 0.060 0.9 0.003 11/16/2016 Seep 1.55 Y 0.02 0.02 5.3 0.1 0.8 2.3 0.1 1.8 0.04 3.9 0.03 0.080 0.4 0.018 1/16/2017 Seep 1.55 Y <0.01 0.02 3.8 0.0 0.6 1.6 0.0 1.2 0.07 3.1 0.02 0.094 0.1 0.004 2/24/2017 Seep 1.55 Y <0.01 0.02 3.9 0.0 0.7 1.7 0.0 1.5 0.07 3.3 0.02 0.063 0.1 0.001 5/10/2017 Seep 1.55 Y <0.01 0.02 4.6 0.1 0.6 1.9 0.1 1.6 <0.11 3.9 0.03 0.073 0.3 0.006 6/2/2017 Seep 1.55 Y <0.01 0.02 4.9 0.4 0.7 2.0 0.1 1.8 <0.11 4.0 0.03 0.038 0.6 0.007

192

Filt Date Site (Y/N) Al Ba Ca Fe K Mg Mn Na P Si Sr Cra Asa Ua 6/22/2017 Seep 1.55 Y 0.01 0.02 5.0 0.4 0.7 2.1 0.1 1.9 <0.07 3.9 0.03 0.037 0.7 0.011 6/28/2017 Seep 1.55 Y 0.01 0.02 5.1 0.2 0.7 2.2 0.1 1.8 <0.07 4.1 0.03 0.020 0.6 0.006 7/10/2017 Seep 1.55 Y <0.005 0.02 5.4 0.3 0.8 2.3 0.1 1.8 0.04 4.2 0.03 0.036 0.7 0.006 9/8/2017 Seep 1.55 Y <0.005 0.03 6.2 0.3 0.9 2.6 0.1 2.3 <0.05 4.4 0.03 - - - 12/3/2017 Seep 1.55 Y ------1/16/2017 Seep 1.53 Y 0.01 0.02 3.9 0.2 0.5 1.7 0.2 1.2 <0.02 3.1 0.02 0.078 0.3 0.004 2/24/2017 Seep 1.53 Y ------5/10/2017 Seep 1.53 Y 0.01 0.02 4.8 1.1 0.7 1.9 0.4 1.6 <0.11 3.9 0.03 0.054 1.7 0.010 6/22/2017 Seep 1.53 Y 0.01 0.02 5.4 1.1 0.7 2.2 0.4 1.9 <0.07 4.1 0.03 - - - 6/28/2017 Seep 1.53 Y 0.02 0.02 5.3 0.5 0.8 2.2 0.3 2.3 <0.07 4.1 0.03 - - - 7/26/2017 Seep 1.53 Y ------10/26/2016 Seep 1.6 Y <0.005 0.06 17.9 0.5 2.0 4.6 2.7 11.7 <0.09 3.7 0.29 <0.046 0.9 <0.001 11/16/2016 Seep 1.6 Y 0.02 0.05 16.6 0.2 1.8 4.5 2.4 12.0 <0.03 3.6 0.29 <0.046 0.8 <0.001 1/16/2017 Seep 1.6 Y <0.01 0.06 13.1 11.1 1.0 3.5 1.5 11.3 0.06 3.4 0.24 <0.046 4.3 0.003 2/24/2017 Seep 1.6 Y <0.01 0.07 15.4 11.8 1.7 4.4 1.9 14.9 0.07 3.9 0.33 0.073 4.0 0.004 5/10/2017 Seep 1.6 Y <0.01 0.07 16.2 5.3 1.5 4.4 1.8 13.7 <0.11 4.2 0.34 0.029 2.5 0.004 6/2/2017 Seep 1.6 Y <0.01 0.08 17.5 1.6 2.0 4.8 2.0 18.5 <0.11 4.5 0.36 0.019 2.1 0.004 6/22/2017 Seep 1.6 Y 0.03 0.08 20.3 0.2 2.1 5.7 1.9 23.9 <0.07 4.2 0.43 0.023 1.7 0.004 6/28/2017 Seep 1.6 Y 0.01 0.08 20.1 0.0 2.3 5.7 2.0 24.5 <0.07 4.0 0.41 0.022 1.6 0.004 7/10/2017 Seep 1.6 Y <0.005 0.07 18.4 0.0 2.0 5.0 1.9 18.6 0.04 4.0 0.39 0.017 1.3 0.004 7/26/2017 Seep 1.6 Y <0.005 0.05 13.3 0.5 1.7 3.2 1.5 8.5 <0.03 4.1 0.20 0.026 3.0 0.003 9/8/2017 Seep 1.6 Y <0.005 0.08 17.4 0.3 2.2 4.7 1.8 15.9 <0.05 3.9 0.36 0.036 1.7 <0.001 10/2/2017 Seep 1.6 Y <0.005 0.09 18.8 0.9 3.2 5.2 2.5 18.1 0.05 3.9 0.38 0.104 1.9 <0.001 10/16/2017 Seep 1.6 Y <0.005 0.09 16.3 7.4 2.4 4.4 2.4 14.0 0.04 4.6 0.34 0.039 4.8 <0.001 12/3/2017 Seep 1.6 Y <0.005 0.10 17.5 17.8 1.4 5.1 2.3 19.3 <0.12 4.0 0.44 0.050 5.8 0.005 4/23/2018 Seep 1.6 Y <0.005 0.07 13.4 4.0 1.2 4.2 1.2 14.3 <0.01 3.6 0.33 0.105 <1.0 0.005 2/24/2017 SRS Seep 1.5 Y 0.02 0.01 4.6 0.1 0.6 0.6 0.0 69.4 0.1 4.4 0.08 <0.046 1.2 0.008 4/23/2018 SRS Seep 1.5 Y ------

193

Filt Date Site (Y/N) Al Ba Ca Fe K Mg Mn Na P Si Sr Cra Asa Ua ANF Seepages Ludlow 10/23/2017 Seepage Y <0.005 0.10 5.2 4.7 1.0 2.2 1.8 0.4 0.1 2.9 0.03 0.06 11.0 <0.001 Ludlow 4/9/2018 Seepage Y <0.005 0.06 4.2 3.6 0.9 1.7 1.4 0.3 0.1 2.6 0.02 0.079 6.7 0.014 Ludlow 4/9/2018 Seepage N 0.129 0.07 4.3 6.8 0.9 1.8 1.4 0.3 0.1 2.7 0.02 0.106 16.0 0.061 Ludlow 7/11/2018 Seepage Y <0.01 0.10 4.8 4.4 1.1 2.1 1.7 0.4 0.1 2.7 0.03 0.067 10.0 <0.001 Ludlow 10/29/2018 Seepage Y 0.01 0.09 5.0 4.5 0.9 2.1 1.7 0.3 0.1 2.8 0.03 <0.046 7.8 <0.001 Ludlow 10/29/2018 Seepage N <0.005 0.09 5.0 0.2 0.9 2.1 1.6 0.3 0.0 2.7 0.02 <0.046 <1.0 <0.001 Oil Well 10/24/2017 discharge Y <0.005 0.26 23.6 0.4 1.6 7.6 0.2 12.5 0.0 5.1 0.2 0.02 0.7 <0.001 Morrison Creek 7/10/2018 Seepage Y <0.01 3.06 26.8 0.3 2.3 5.6 0.3 33.5 0.0 6.1 0.6 <0.046 <1.0 <0.001 Morrison Creek 10/29/2018 Seepage Y <0.005 3.12 31.3 0.4 2.7 5.9 0.3 37.8 0.0 6.4 0.6 <0.046 1.1 <0.001 Morrison Creek 10/29/2018 Seepage N 0.01 3.13 31.6 1.1 2.7 5.8 0.3 38.0 0.0 6.3 0.6 <0.046 1.4 <0.001 Morrison Road 10/29/2018 Seepage Y 0.05 0.15 11.8 1.2 1.2 4.3 0.3 11.3 0.0 4.5 0.10 0.078 1.4 <0.001 Blood Run 7/10/2018 Seep A Y <0.01 0.15 24.2 11.6 1.2 6.4 2.4 3.5 0.0 4.6 0.13 0.036 20.1 <0.001 Blood Run 10/29/2018 Seep A Y <0.005 0.04 9.9 1.6 0.8 2.6 0.6 1.1 0.0 3.5 0.04 0.072 2.9 <0.001 Blood Run 10/29/2018 Seep A N 0.02 0.05 10.0 2.8 0.8 2.7 0.6 1.2 0.1 3.7 0.04 0.124 8.1 <0.001 Moshannon Forest Seepages

194

Filt Date Site (Y/N) Al Ba Ca Fe K Mg Mn Na P Si Sr Cra Asa Ua 4/13/2017 LR 1 seep Y 0.02 0.14 6.0 3.7 1.1 2.1 0.6 1.5 <0.07 3.2 0.03 - - - 4/13/2017 LR 2A Seep Y 0.09 0.13 6.4 3.1 1.0 1.9 0.3 3.0 <0.07 2.4 0.02 - - - 9/18/2017 LR 2A Seep Y 0.05 0.19 10.5 5.0 0.9 3.0 0.5 3.2 <0.05 2.8 0.04 - - - 9/18/2017 LR 2B Seep Y 0.01 0.19 9.9 17.5 0.9 3.1 0.9 2.8 0.1 3.1 0.04 - - - 9/18/2017 LR 3A seep Y 0.02 0.07 4.2 1.2 1.2 1.5 0.8 0.7 <0.05 1.7 0.02 - - - 9/18/2017 LR 3B Seep Y <0.005 0.27 12.6 4.3 0.9 2.6 0.5 1.4 <0.05 3.0 0.03 - - - LR Discharge 9/18/2017 pipe Y <0.005 0.33 10.5 5.4 1.1 2.7 0.4 21.1 <0.05 3.1 0.09 - - - Gas-rich Seep (Natural) Salt Spring 7/17/2017 State Park Y 0.03 110 339 1.30 34.3 46.9 0.24 1,920 0.19 3.6 62.8 - - - aAnalyte measured using ICP-MS

195

Table E13. Measured anions and YSI parameters in GLD and Salt Springs as cited in Chapters 2 and 3 (not including LAWs). Inorganic analytes are mg/L unless otherwise stated SPC Temp Date Site F Cl SO4 Br NO3 pH DO % (µS/cm) ORP ᵒC

Sugar Run (Lycoming) 11/12/2013 Seep 1.5 - 2.2 10.4 <0.09 <0.51 5.77 20.3 62 - 9.8 9/15/2016 Seep 1.5 0.02 2.1 7.9 0.1 0.2 - - - - - 10/26/2016 Seep 1.5 0.02 2.8 7.7 0.1 0.2 - - - - - 11/16/2016 Seep 1.5 0.04 2.2 9.0 0.1 0.5 - - - - - 1/16/2017 Seep 1.5 0.02 1.6 9.5 0.0 1.0 - - - - - 2/24/2017 Seep 1.5 0.03 1.2 9.9 0.1 0.3 6.89 43.9 51 - 6.7 5/10/2017 Seep 1.5 0.06 1.1 8.2 - 0.6 6.60 40.2 - - 11.3 6/22/2017 Seep 1.5 0.06 1.2 7.2 0.1 0.5 6.42 20.9 105 -13 15.2 6/28/2017 Seep 1.5 0.06 1.3 8.0 0.1 0.6 6.38 13.5 98 -68 14.8 7/10/2017 Seep 1.5 0.05 1.3 6.9 0.0 0.5 6.40 12.1 114 -35 17.1 9/8/2017 Seep 1.5 0.04 2.5 6.4 0.1 1.6 6.73 35.1 115 13 10/2/2017 Seep 1.5 0.17 5.3 3.0 0.1 0.6 6.41 26.1 119 2 12.2 10/16/2017 Seep 1.5 0.05 2.2 7.1 0.1 0.6 6.12 31.8 107 13 14.8 12/3/2017 Seep 1.5 0.06 2.2 10.0 - 0.7 6.45 - 74 117 6.8 4/23/2018 Seep 1.5 0.02 1.3 9.3 0.1 1.0 7.03 31 58 118 8.5 9/16/2016 Seep 1.55 0.02 2.1 9.9 0.1 0.2 - - - - - 10/26/2016 Seep 1.55 0.02 2.9 9.7 0.1 ------11/16/2016 Seep 1.55 0.05 2.6 10.1 0.0 0.5 - - - - - 1/16/2017 Seep 1.55 0.02 1.4 9.5 0.0 1.9 - - - - - 2/24/2017 Seep 1.55 0.03 1.2 10.3 - 0.6 6.25 41.7 52 - 5.5 5/10/2017 Seep 1.55 0.06 1.0 8.9 - 0.5 6.06 27.6 - - 10.5 6/2/2017 Seep 1.55 0.05 1.5 8.7 0.1 0.6 - - - - - 6/22/2017 Seep 1.55 0.06 1.2 8.9 0.0 0.5 - - - - - 6/28/2017 Seep 1.55 0.06 1.2 8.7 0.1 0.5 5.86 36.8 58 90 13.9 7/10/2017 Seep 1.55 0.05 1.2 8.0 0.1 0.5 6.01 36.6 76 94 15.1

196

SPC Temp Date Site F Cl SO4 Br NO3 pH DO % (µS/cm) ORP ᵒC 9/8/2017 Seep 1.55 0.03 1.7 7.8 0.1 0.6 6.16 36.2 83 93 15.5 12/3/2017 Seep 1.55 - - - - - 6.23 - 70 169 9.4 1/16/2017 Seep 1.53 0.01 1.3 9.2 0.1 1.2 - - - - - 5/10/2017 Seep 1.53 - - - - - 6.07 28.6 - - 11 6/22/2017 Seep 1.53 0.06 1.2 8.5 0.0 0.5 6.11 29.4 64 92 14.9 6/28/2017 Seep 1.53 0.06 1.2 8.7 0.1 0.5 6.17 37.9 61 88 13.9 7/26/2017 Seep 1.53 - - - - - 6.45 46.2 77 61 16.5 10/26/2016 Seep 1.6 0.06 22.7 1.3 0.1 0.5 - - - - - 11/16/2016 Seep 1.6 0.07 25.4 1.1 0.1 0.5 - - - - - 1/16/2017 Seep 1.6 0.03 25.5 6.2 0.1 1.3 - - - - - 2/24/2017 Seep 1.6 0.04 41.2 3.0 0.1 0.3 6.59 48.7 242 - 7.1 5/10/2017 Seep 1.6 0.06 26.8 3.5 0.1 0.5 6.38 11.2 - - 11.2 6/2/2017 Seep 1.6 0.07 32.3 2.8 0.1 0.5 - - - - - 6/22/2017 Seep 1.6 0.08 46.6 1.8 0.1 0.5 6.47 26.8 283 55 17.9 6/28/2017 Seep 1.6 0.08 45.7 2.2 0.1 0.5 6.51 13.5 289 12 14.8 7/10/2017 Seep 1.6 0.08 38.4 2.0 0.1 0.5 6.44 15.3 296 34 17.4 7/26/2017 Seep 1.6 0.01 12.2 5.5 0.1 0.7 6.31 20.0 161 31 18.5 9/8/2017 Seep 1.6 0.07 30.7 1.7 0.1 0.7 6.45 25.1 257 6 14.5 10/2/2017 Seep 1.6 0.07 35.1 0.8 0.1 0.5 6.48 31.9 - - - 10/16/2017 Seep 1.6 0.29 21.5 0.7 0.1 0.5 6.45 19.3 224 -90 14.5 12/3/2017 Seep 1.6 0.10 53.6 1.5 0.4 0.3 6.20 - 425 -40 5.5 4/23/2018 Seep 1.6 0.03 33.4 1.9 0.2 0.6 6.57 13.1 304 -55 6.6 2/24/2017 SRS Seep 1.5 0.23 23.6 5.4 0.1 0.6 9.07 31.0 300 - 8.8 4/23/2018 SRS Seep 1.5 - - - - - 8.05 31.9 276 -114 9.9 ANF Seepages 10/23/2017 Ludlow Seepage 0.04 0.3 0.3 0.05 0.5 6.15 - - - - 4/9/2018 -filt Ludlow Seepage 0.01 0.3 1.1 0.1 0.5 6.20 2.7 97 -26 8.6 7/11/2018 Ludlow Seepage 0.00 0.5 1.7 <0.01 0.2 6.06 10.1 - - -

197

SPC Temp Date Site F Cl SO4 Br NO3 pH DO % (µS/cm) ORP ᵒC 10/29/2018 Ludlow Seepage 0.03 0.6 1.9 - - 6.02 5.4 - -29 10.5 10/29/2018 Ludlow Seepage 0.03 0.6 1.9 - - 6.02 5.4 - -29 10.5 10/24/2017 Oil Well discharge 0.05 9.8 6.4 0.1 0.5 5.90 - 238 -38 8.7 Morrison Creek 7/10/2018 Seepage 0.44 11.7 1.7 <0.01 0.2 7.07 - - - - Morrison Creek 10/29/2018 Seepage 0.38 12.7 0.8 - - 6.55 6.0 363 -64 9 Morrison Creek 10/29/2018 Seepage 0.38 12.7 0.8 - - 6.55 6.0 363 -64 9 10/29/2018 Morrison Road Seepage 0.21 3.5 0.4 - - 6.72 83.0 145 55 9.2 7/10/2018 Blood Run Seep A 0.15 1.0 3.6 0.2 0.2 - - - - - 10/29/2018 Blood Run Seep A 0.08 0.5 0.3 - - 5.95 20.0 98 95 10.4 Moshannon Forest Seepages 4/13/2017 LR 1 seep 0.08 0.7 5.9 0.1 0.4 5.99 - - - - 4/13/2017 LR 2A Seep 0.10 1.3 2.3 0.1 0.4 6.03 - - - - 9/18/2017 LR 2A Seep 0.08 1.5 1.4 0.1 1.7 - - - - - 9/18/2017 LR 2B Seep 0.08 1.4 1.3 0.1 0.8 - - - - - 9/18/2017 LR 3A seep 0.08 0.6 2.9 0.1 0.7 5.99 - - - - 9/18/2017 LR 3B Seep 0.08 0.5 2.2 0.1 0.7 6.02 - - - - 9/18/2017 LR Discharge Pipe 0.15 14.3 0.9 0.2 0.7 6.65 - - - - Gas-rich Seep (Natural) 7/17/2017 Salt Spring State Park 0.51 3345 8.9 31.9 17.4 7.32 10 12000 -119 9

198

Table E14. Measured hydrocarbon concentrations and isotopes in AMD discharge cited in Chapter 3

KOH CH4 δ13C- Date Site (Y/N) Lat Long Sampler (µg/L) CH4 Anthracite Mines 2/4/2017 Gravity Slope N 41.4812 -75.5630 PSU + Volunteer 21 - 2/4/2017 Old Forge 1 N 41.3590 -75.7513 PSU + Volunteer 106 - 2/4/2017 Old Forge 1a N 41.3590 -75.7513 PSU + Volunteer 118 - 2/4/2017 Old Forge 2 N 41.3590 -75.7513 PSU + Volunteer 112 - 2/4/2017 Honey Pot N 41.2069 -76.0061 PSU + Volunteer 164 - 2/4/2017 Honey Pot 2 N 41.2069 -76.0061 PSU + Volunteer 157 - Pitt Area 11/15/2016 Coal Run N 40.3481 -80.1134 Fred Zelt 93 - 12/3/2106 Coal Run N 40.3481 -80.1134 Fred Zelt + PSU 117 - 2/4/2017 Coal Run N 40.3481 -80.1135 Fred Zelt 136 - 5/13/2017 Coal Run N 40.3481 -80.1135 Fred Zelt + PSU 149 - 11/15/2016 Gladding N 40.3422 -80.1708 Fred Zelt 114 - 12/3/2016 Gladding N 40.3422 -80.1708 Fred Zelt + PSU 116 - 2/4/2017 Gladding N 40.3424 -80.1708 Fred Zelt 115 - 5/13/2017 Gladding N Fred Zelt + PSU 199 -31.0 11/15/2016 Mclaughlin N 40.3548 -80.1039 Fred Zelt 74 - 12/3/2016 Presto-Sygan N 40.3666 -80.1247 Fred Zelt + PSU 227 - 2/4/2017 Presto-Sygan N 40.3665 -80.1247 Fred Zelt 222 - 5/13/2017 Presto-Sygan N Fred Zelt + PSU 246 - 12/3/2016 Scrubgrass N 40.3821 -80.0896 Fred Zelt + PSU 50 - 2/4/2017 Scrubgrass N 40.3820 -80.0896 Fred Zelt 55 - 2/4/2017 Tinkers Run N 40.3333 -79.7125 Fred Zelt 99 - 2/4/2017 Blythdale N 40.2564 -79.7951 Fred Zelt 70 - 5/13/2017 Blythdale N 40.2564 -79.7951 Fred Zelt + PSU 109 -43.0

199

KOH CH4 δ13C- Date Site (Y/N) Lat Long Sampler (µg/L) CH4 2/4/2017 Elrama N 40.2430 -79.9531 Fred Zelt 88 - 2/4/2017 Elrama N 40.2428 -79.9535 Fred Zelt 77 - 2/4/2017 Elrama N 40.2427 -79.9538 Fred Zelt 131 - Other PSU + Greg 6/8/2017 Vindale borehole (middle) N 40.4848 -78.9227 Mount 127 - PSU + Greg 6/8/2017 Vindale bore. (upstream) N 40.4849 -78.9227 Mount 550 - PSU + Greg 6/8/2017 Passive Treatment N 40.6940 -79.0508 Mount 103 -

200

Table E15. Measured cation concentrations (mg/L) in AMD discharge cited in Chapter 3.

Filt Date Site (Y/N) Al Ba Ca Fe K Mg Mn Na P Si Sr Ti

Pitt Area 5/13/2017 Gladding Y 0.03 0.01 82.0 59.5 7.23 32.2 0.60 202 <0.11 7.91 1.17 <0.005 5/13/2017 Blythdale Y <0.01 0.01 95.9 36.9 4.24 36.8 0.61 128 <0.11 5.35 1.53 <0.005 2/4/2017 Elrama Y 1.77 0.01 113 62.0 5.7 42 1.5 128 <0.11 12 1.3 <0.005

201

Table E16. Measured anion concentrations and YSI parameters in AMD discharge cited in Chapter 3

DO SPC Temp Date Site F CL SO4 Br NO3 pH % (µS/cm) ORP ᵒC Anthracite Mines 12/3/2016 Gladding 0.18 71.2 499 0.46 0.90 5.89 - 1400 - 12.3 5/13/2017 Gladding 0.29 87.7 442 0.71 0.91 6.02 - - - - 12/3/2016 Presto-Sygan 0.47 150 575 0.36 0.76 5.59 - 1660 - 12.4 12/3/2016 Scrubgrass 0.14 381 587 0.59 0.81 6.09 - 2610 - 12.8 5/13/2017 Blythdale 0.23 87.6 442 0.72 0.94 6.27 - - - - 5/13/2017 Elrama 0.22 80.4 359 0.37 1.01 3.42 54 1720 - 19.2

202

Table E17. Stream methane samples from all datasets cited in and presented in Chapter 3

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane 6/22/2015 Wendt et al., 2018 40.2500 -80.2060 3.0 3.0 Abandoned well - Alleghany SF - Wendt et al., 2018 - Test Elk Bar Run 3/9/2016 Dataset 41.7660 -78.7190 0.2 Abandoned well - Alleghany SF - Wendt et al., 2018 - Test Elk Bar Run 3/9/2016 Dataset 41.7665 -78.7181 2.1 Abandoned well - Alleghany SF - Wendt et al., 2018 - Test Elk Bar Run 3/9/2016 Dataset 41.7665 -78.7182 2.0 Abandoned well - Alleghany SF - Wendt et al., 2018 - Test Elk Bar Run 3/9/2016 Dataset 41.7666 -78.7184 2.1 Abandoned well - Alleghany SF - Wendt et al., 2018 - Test Elk Bar Run 3/9/2016 Dataset 41.7668 -78.7187 2.5 1.8 Abandoned well - Bennett Branch Wendt et al., 2018 - Test Sinnemahoning 3/9/2016 Dataset 41.2761 -78.4012 0.4 Abandoned well - Bennett Branch Wendt et al., 2018 - Test Sinnemahoning 3/9/2016 Dataset 41.2764 -78.4009 0.5 Abandoned well - Bennett Branch Wendt et al., 2018 - Test Sinnemahoning 3/9/2016 Dataset 41.2765 -78.4009 0.3 Abandoned well - Bennett Branch Wendt et al., 2018 - Test Sinnemahoning 3/9/2016 Dataset 41.2767 -78.4009 0.4 0.4 ARNORU001 10/24/2017 ANF Snapshot Day 1 41.7390 -79.0806 4.2 ARNORU001 4/10/2018 ANF Snapshot Day 2 41.7390 -79.0806 0.9 ARNORU001 10/30/2018 ANF Snapshot Day 3 41.7390 -79.0806 2.0 2.4 ARNORU002 4/10/2018 ANF Snapshot Day 2 41.7198 -79.0758 0.8 0.8 ASAPRU002 6/24/2017 Pine Creek Snapshot Day 41.7819 -77.4330 0.2 0.2 BABBCR003 6/24/2017 Pine Creek Snapshot Day 41.6420 -77.2084 0.7 0.7 BABBCR004 6/24/2017 Pine Creek Snapshot Day 41.5729 -77.3353 0.4 0.4 Bailey Run (lower) 9/26/2015 Wendt et al., 2018 41.5123 -78.0458 1.0 1.0 Bailey Run (upper) 9/26/2015 Wendt et al., 2018 41.5236 -78.0660 0.5 0.5 Bailey Run 1 12/14/2016 PSU 41.7619 -76.5547 0.5 0.5 BALDRU002 6/24/2017 Pine Creek Snapshot Day 41.7939 -77.3045 1.2 1.2

203

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane Barberry 1/14/2016 Wendt et al., 2018 40.5760 -80.138 0.4 Barberry 1/14/2016 Wendt et al., 2018 40.5760 -80.138 0.5 0.4 BC Orvistan road down 11/2/2016 PSU 41.0850 -77.8676 0.4 0.4 BC Orvistan road down up 11/2/2016 PSU 41.0858 -77.8672 2.2 2.2 BC2 - downstream 7/11/2018 PSU 41.3984 -78.8223 0.7 0.7 BEARCR006 10/24/2017 ANF Snapshot Day 1 41.4907 -78.8607 10.9 -49.6 BEARCR006 4/10/2018 ANF Snapshot Day 2 41.4907 -78.8607 6.9 -54.4 BEARCR006 7/11/2018 PSU 41.4904 -78.8606 7.4 8.4 -50 BEARCR007 4/10/2018 ANF Snapshot Day 2 41.4904 -78.8613 6.6 BEARCR007 10/30/2018 ANF Snapshot Day 3 41.4904 -78.8613 5.1 5.8 BEARCR008 4/10/2018 ANF Snapshot Day 2 41.4913 -78.8600 6.8 6.8 Beauty Run_Kato Rd 9/16/2015 Wendt et al., 2018 41.0783 -77.9073 0.5 Beauty Run_Kato Rd 3/1/2016 Wendt et al., 2018 41.0783 -77.9073 0.4 Beauty Run_Kato Rd 5/5/2016 Wendt et al., 2018 41.0783 -77.9073 0.1 0.3 BeavRu01 10/30/2018 ANF Snapshot Day 3 41.5924 -79.3018 2.7 2.74 Beech Creek 11/10/2015 Wendt et al., 2018 41.1075 -77.6936 0.2 0.2 BeechCreek_Monument 8/10/2015 Wendt et al., 2018 41.1133 -77.7046 0.5 BeechCreek_Monument 4/11/2016 Wendt et al., 2018 41.1133 -77.7046 0.2 0.3 Berge Run 9/26/2015 Wendt et al., 2018 41.4886 -78.0518 0.1 0.1 Big Nelson Run 9/26/2015 Wendt et al., 2018 41.5564 -78.0341 0.3 0.3 BigRun 8/10/2015 Wendt et al., 2018 41.1107 -77.7324 0.5 Big Run 9/16/2015 Wendt et al., 2018 41.1107 -77.7324 1.0 LHU_Big_Run 11/9/2015 Wendt et al., 2018 41.1107 -77.7324 0.3 BigRun 4/11/2016 Wendt et al., 2018 41.1107 -77.7324 0.1 0.5 BIGRUN001 6/24/2017 Pine Creek Snapshot Day 41.5438 -77.3619 0.2 0.2 Billy Buck Run 9/26/2015 Wendt et al., 2018 41.5874 -78.4416 0.2 0.2 BIMICR001 4/10/2018 ANF Snapshot Day 2 41.4534 -78.7875 5.3

204

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane BIMICR001 10/30/2018 ANF Snapshot Day 3 41.4534 -78.7875 4.0 4.6 BIMICR002 4/10/2018 ANF Snapshot Day 2 41.5438 -78.7875 0.6 BIMICR002 10/30/2018 ANF Snapshot Day 3 41.5438 -78.7875 0.74 0.7 BiMiCr003 10/30/2018 ANF Snapshot Day 3 41.4913 -78.7920 1.6 1.58 Birch Run 9/26/2015 Wendt et al., 2018 41.5579 -77.9511 0.8 0.8 BLACCR003 6/24/2017 Pine Creek Snapshot Day 41.5660 -77.1346 2.4 2.4 BLACFO001 10/1/2016 West Virginia Snapshot Day 39.0723 -79.6331 0.6 0.6 Black Mo meets Red Mo - Site 3 6/24/2015 Wendt et al., 2018 41.0360 -78.0598 0.3 Black Mo meets Red Mo - Site 3 6/24/2015 Wendt et al., 2018 41.0360 -78.0598 0.2 Black Mo meets Red Mo - Site 3 7/8/2015 Wendt et al., 2018 41.0360 -78.0598 0.1 Black Mo meets Red Mo - Site 3 7/8/2015 Wendt et al., 2018 41.0360 -78.0598 0.1 Black Mo meets Red Mo - Site 3 7/8/2015 Wendt et al., 2018 41.0360 -78.0598 0.1 Black Mo meets Red Mo - Site 3 8/19/2015 Wendt et al., 2018 41.0360 -78.0598 0.2 Black Mo meets Red Mo - Site 3 8/19/2015 Wendt et al., 2018 41.0360 -78.0598 0.3 Black Mo meets Red Mo - Site 3 8/19/2015 Wendt et al., 2018 41.0360 -78.0598 0.2 Black Mo meets Red Mo - Site 3 9/16/2015 Wendt et al., 2018 41.0360 -78.0598 2.2 0.4 Black Moshannon at Bridge - Site 2 6/24/2015 Wendt et al., 2018 41.0161 -78.0215 0.6 Black Moshannon at Bridge - Site 2 6/24/2015 Wendt et al., 2018 41.0161 -78.0215 0.4 Black Moshannon at Bridge - Site 2 6/24/2015 Wendt et al., 2018 41.0161 -78.0215 0.2 Black Moshannon at Bridge - Site 2 7/8/2015 Wendt et al., 2018 41.0161 -78.0215 0.5 Black Moshannon at Bridge - Site 2 7/8/2015 Wendt et al., 2018 41.0161 -78.0215 0.4 Black Moshannon at Bridge - Site 2 7/8/2015 Wendt et al., 2018 41.0161 -78.0215 0.4 Black Moshannon at Bridge - Site 2 8/19/2015 Wendt et al., 2018 41.0161 -78.0215 0.7

205

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane Black Moshannon at Bridge - Site 2 8/19/2015 Wendt et al., 2018 41.0161 -78.0215 0.8 Black Moshannon at Bridge - Site 2 8/19/2015 Wendt et al., 2018 41.0161 -78.0215 0.9 Black Moshannon at Bridge - Site 2 9/16/2015 Wendt et al., 2018 41.0161 -78.0215 0.5 Black Moshannon at Bridge - Site 2 9/16/2015 Wendt et al., 2018 41.0161 -78.0215 0.2 Black Moshannon at Bridge - Site 2 9/16/2015 Wendt et al., 2018 41.0161 -78.0215 0.4 Black Moshannon at Bridge - Site 2 10/16/2015 Wendt et al., 2018 41.0161 -78.0216 0.4 Black Moshannon at Bridge - Site 2 10/16/2015 Wendt et al., 2018 41.0161 -78.0216 0.6 Black Moshannon at Bridge - Site 2 10/16/2015 Wendt et al., 2018 41.0161 -78.0215 0.6 Black Moshannon at Bridge - Site 2 10/16/2015 Wendt et al., 2018 41.0161 -78.0216 0.4 Black Moshannon at Bridge - Site 2 10/16/2015 Wendt et al., 2018 41.0160 -78.0218 0.6 Black Moshannon at Bridge - Site 2 10/16/2015 Wendt et al., 2018 41.0160 -78.0216 2.1 Black Moshannon at Bridge - Site 2 10/16/2015 Wendt et al., 2018 41.0160 -78.0219 0.8 Black Moshannon at Bridge - Site 2 11/10/2015 Wendt et al., 2018 41.0161 -78.0215 0.5 Black Moshannon at Bridge - Site 2 11/10/2015 Wendt et al., 2018 41.0161 -78.0215 0.3 Black Moshannon at Bridge - Site 2 11/10/2015 Wendt et al., 2018 41.0161 -78.0215 0.5 Black Moshannon at Bridge - Site 2 11/10/2015 Wendt et al., 2018 41.0161 -78.0215 0.4 Black Moshannon at Bridge - Site 2 11/10/2015 Wendt et al., 2018 41.0161 -78.0215 0.3 0.6 Black Moshannon State Park (near outflow to Dam) 8/19/2015 Wendt et al., 2018 41.0161 -78.0215 25.6 25.6

206

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane Black Moshannon State Park - Site 1 6/24/2015 Wendt et al., 2018 40.9190 -78.0590 0.6 Black Moshannon State Park - Site 1 6/24/2015 Wendt et al., 2018 40.9190 -78.0590 0.1 Black Moshannon State Park - Site 1 6/24/2015 Wendt et al., 2018 40.9190 -78.0590 0.5 Black Moshannon State Park - Site 1 7/8/2015 Wendt et al., 2018 40.9190 -78.0590 7.9 Black Moshannon State Park - Site 1 7/8/2015 Wendt et al., 2018 40.9190 -78.0590 7.6 Black Moshannon State Park - Site 1 7/8/2015 Wendt et al., 2018 40.9190 -78.0590 8.8 Black Moshannon State Park - Site 1 8/19/2015 Wendt et al., 2018 40.9190 -78.0590 14.7 Black Moshannon State Park - Site 1 8/19/2015 Wendt et al., 2018 40.9190 -78.0590 15.2 Black Moshannon State Park - Site 1 3/16/2016 Wendt et al., 2018 40.9190 -78.0590 6.2 Black Moshannon State Park - Site 1 3/16/2016 Wendt et al., 2018 40.9190 -78.0590 7.8 Black Moshannon State Park - Site 1 3/16/2016 Wendt et al., 2018 40.9190 -78.0590 6.7 Black Moshannon State Park - Site 1 3/16/2016 Wendt et al., 2018 40.9190 -78.0590 6.2 Black Moshannon State Park - Site 1 3/16/2016 Wendt et al., 2018 40.9190 -78.0590 7.4 Black Moshannon State Park - Site 1 3/16/2016 Wendt et al., 2018 40.9190 -78.0590 6.0 Black Moshannon State Park - Site 1 3/16/2016 Wendt et al., 2018 40.9190 -78.0590 7.0 Black Moshannon State Park - Site 1 3/16/2016 Wendt et al., 2018 40.9190 -78.0590 5.5 Black Moshannon State Park - Site 1 3/16/2016 Wendt et al., 2018 40.9190 -78.0590 6.0 6.7 BLHOCR002 6/24/2017 Pine Creek Snapshot Day 41.5603 -77.1049 1.0 1.0 BloodR001 7/10/2018 PSU 41.6915 -78.9219 23.3 23.3 -40.9 2.09

207

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane BLUECR001 10/24/2017 ANF Snapshot Day 1 41.5432 -79.0566 0.5 BLUECR001 4/10/2018 ANF Snapshot Day 2 41.5433 -79.0556 0.3 0.4 BOONRU001 6/24/2017 Pine Creek Snapshot Day 41.3491 -77.3579 0.3 0.3 BR001 7/19/2016 Anna 41.7702 -76.5439 2.5 2.5 BR002 7/18/2016 Anna 41.7809 -76.5335 2.7 2.7 BRBUFO001 10/1/2016 West Virginia Snapshot Day 38.5293 -79.6900 0.6 0.6 BROWRU01 10/30/2018 ANF Snapshot Day 3 41.8186 -79.0930 0.24 0.2 BROWRU003 6/24/2017 Pine Creek Snapshot Day 41.3426 -77.3993 0.4 0.4 Buffalo Creek 10/18/2016 Volunteer 40.1843 -80.4866 2.7 2.7 BUFFFO001 10/1/2016 West Virginia Snapshot Day 38.5286 -79.6905 1.0 1.0 BULIRU001 10/24/2017 ANF Snapshot Day 1 41.8774 -78.8005 2.6 BULIRU001 4/10/2018 ANF Snapshot Day 2 41.8774 -78.8005 1.2 BULIRU001 10/30/2018 ANF Snapshot Day 3 41.8774 -78.8005 1.1 1.6 BULIRU002 10/24/2017 ANF Snapshot Day 1 41.8368 -78.7701 0.3 BULIRU002 4/10/2018 ANF Snapshot Day 2 41.8368 -78.7701 0.4 BULIRU002 10/30/2018 ANF Snapshot Day 3 41.8368 -78.7701 0.45 0.4 Burdick Creek 7/5/2016 SRBC 41.7175 -75.8727 0.99 Burdick Creek 10/18/2016 SRBC 41.7175 -75.8727 0.92 1.0 Butler Creek 7/5/2016 SRBC 41.7293 -75.6747 1.6 Butler Creek 10/18/2016 SRBC 41.7293 -75.6747 -0.23 0.7 Caleb Run 7/29/2015 Wendt et al., 2018 41.3356 -76.9553 0.3 0.3 CampRun 10/30/2018 ANF Snapshot Day 3 41.5852 -79.3396 0.85 0.9 Wendt et al., 2018 - Test Canadaway Creek 7/3/2016 12:38 Dataset 42.4378 -79.3335 8.0 Wendt et al., 2018 - Test Canadaway Creek 7/3/2016 15:09 Dataset 42.4755 -79.3646 0.8 Wendt et al., 2018 - Test Canadaway Creek 7/3/2016 15:36 Dataset 42.4328 -79.3137 4.3

208

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane Wendt et al., 2018 - Test Canadaway Creek 7/3/169 11:50 Dataset 42.4417 -79.3916 2.8 Wendt et al., 2018 - Test Canadaway Creek 7/38/16 13:09 Dataset 42.4385 -79.3372 1.5 3.5 CANARU001 6/24/2017 Pine Creek Snapshot Day 41.7827 -77.3749 0.2 0.2 Coal Run Green 10/18/2016 Volunteer 40.3481 -80.1140 0.3 0.3 CEDARI002 6/24/2017 Pine Creek Snapshot Day 41.5553 -77.4593 0.2 0.2 PSU + Chartiers Creek CH #2 (20 feet upstream) 11/7/2016 Watershed 40.1455 -80.2677 14.0 14.0 PSU + Chartiers Creek CH #3 11/7/2016 Watershed 40.1451 -80.2669 30.3 30.3 CHAPFO001 4/10/2018 ANF Snapshot Day 2 41.8167 -78.8579 0.6 0.6 PSU + Chartiers Creek Chartiers Main stem 11/7/2016 Watershed 40.1455 -80.2677 13.6 13.6 Chartiers Run x 6/22/2015 Wendt et al., 2018 40.2480 -80.2120 2.6 Chartiers Run x 1/14/2016 Wendt et al., 2018 40.2480 -80.2120 2.3 2.4 Chartiers Run 6/22/2015 Wendt et al., 2018 40.2580 -80.2570 2.1 2.1 PSU + Chartiers Creek Chartiers Run #1 11/7/2016 Watershed 40.2524 -80.2491 4.3 4.3 PSU + Chartiers Creek Chartiers Run #2 11/7/2016 Watershed 40.2484 -80.2134 2.8 2.8 PSU + Chartiers Creek Chartiers Run #3 11/7/2016 Watershed 40.2485 -80.2147 3.1 3.1 CHEARI001 10/1/2016 West Virginia Snapshot Day 39.1649 -79.7057 0.9 0.9 CHERRU001 4/10/2018 ANF Snapshot Day 2 41.6342 -78.9959 0.2 CHERRU001 10/30/2018 ANF Snapshot Day 3 41.6342 -78.9959 0.4 0.3 ChurRu01 10/30/2018 ANF Snapshot Day 3 41.3770 -78.9736 0.1 0.1 CLOVRU001 10/1/2016 West Virginia Snapshot Day 39.1482 -79.7131 0.7 0.7 Coal Run001 10/18/2016 Volunteer 40.3480 -80.1123 19.6 Coal Run001 11/15/2016 Volunteer 40.3480 -80.1124 24.3 Coal Run001 12/3/2016 Volunteer + PSU 40.3481 -80.1124 27.9 23.9

209

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane Coal Run002 11/15/2016 Volunteer 40.3480 -80.1135 19.4 Coal Run002 12/3/2016 Volunteer + PSU 40.3481 -80.1135 20.2 19.8 Coal Run003 11/15/2016 Volunteer 40.3480 -80.1137 1.3 Coal Run003 12/3/2016 Volunteer + PSU 40.3481 -80.1138 0.2 0.7 Cold Creek 7/6/2016 SRBC 41.7477 -76.2076 2.31 -52.8 Cold Creek 10/19/2016 SRBC 41.7477 -76.2076 2.18 2.2 -47.9 COONRU01 10/30/2018 ANF Snapshot Day 3 41.5822 -78.8980 1.5 1.49 Council Run 10/12/2015 Wendt et al., 2018 41.0907 -77.8190 0.3 Council Run 8/10/2015 Wendt et al., 2018 41.0907 -77.8190 0.2 Council Run 11/9/2015 Wendt et al., 2018 41.0907 -77.8190 0.2 Council Run 5/5/2016 Wendt et al., 2018 41.0907 -77.8190 <0.06 0.2 CR 1 11/2/2016 PSU 41.0910 -77.8184 0.4 0.4 Crooked Creek 3 2/19/2017 Volunteer 41.9490 -80.3600 0.8 Crooked Creek Lower 2/19/2017 Volunteer 42.0030 -80.4310 0.9 Crooked Creek miidle 2/19/2017 Volunteer 41.9840 -80.4040 1.5 1.1 Crooked Creek near Shortsville, PA Anna Wendt 41.8630 -77.3462 1.3 1.3 Cross Creek 10/18/2016 Volunteer 40.2590 -80.3747 4.3 4.3 Cross Creek 10/18/2016 Volunteer 40.2859 -80.4920 1.1 1.1 CROURU001 10/1/2016 West Virginia Snapshot Day 38.6792 -79.8557 1.3 1.3 CUSHHO001 6/24/2017 Pine Creek Snapshot Day 41.8215 -77.7082 0.4 0.4 DAMRUN001 6/24/2017 Pine Creek Snapshot Day 41.3267 -77.3549 0.2 0.2 DAMRUN002 6/24/2017 Pine Creek Snapshot Day 41.3294 -77.3445 0.2 0.2 DIXIRU001 6/24/2017 Pine Creek Snapshot Day 41.5586 -77.3315 0.2 0.2 DLCK 7/1/2016 SRBC 41.7738 -76.0580 8.2 -25.5 DLCK 10/1/2016 SRBC 41.7738 -76.0580 2.2 -36.4 DLCK 5/1/2017 SRBC 41.7738 -76.0580 1.1 DLCK 7/1/2017 SRBC 41.7738 -76.0580 2.0

210

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane DLCK 10/1/2017 SRBC 41.7738 -76.0580 3.6 -31.8 DLCK 9/5/2018 PSU+SRBC 41.7738 -76.0580 2.7 3.3 -39.3 DLCK DS1 5/1/2017 SRBC 41.7751 -76.0596 1.7 DLCK DS1 7/1/2017 SRBC 41.7751 -76.0596 7.5 -34.22 DLCK DS1 10/1/2017 SRBC 41.7751 -76.0596 25.0 -35.03 DLCK DS1 9/5/2018 PSU+SRBC 41.7751 -76.0596 7.3 -33.5 DLCK DS1 - KOH 9/5/2018 PSU+SRBC 41.7751 -76.0596 8.9 10.1 -33.5 DLCK US 2 5/1/2017 SRBC 41.7698 -76.0555 2.7 -49 DLCK US 2 7/1/2017 SRBC 41.7698 -76.0555 2.3 -40.17 DLCK US 2 10/1/2017 SRBC 41.7698 -76.0555 4.6 3.2 -37.79 DLCK US 3 5/1/2017 SRBC 41.7500 -76.0331 12.2 -55 DLCK US 3 7/1/2017 SRBC 41.7500 -76.0331 16.2 -51.52 DLCK US 3 10/1/2017 SRBC 41.7500 -76.0331 2.7 10.4 -35.8 DLCKUS 1 5/1/2017 SRBC 41.7721 -76.0563 0.7 DLCKUS 1 7/1/2017 SRBC 41.7721 -76.0563 2.4 -51.16 DLCKUS 1 10/1/2017 SRBC 41.7721 -76.0563 3.4 -41.46 DLCKUS 1 9/5/2018 PSU+SRBC 41.7721 -76.0563 4.0 2.6 -39.6 Driftwood Branch (Emporium) 9/26/2015 Wendt et al., 2018 41.5082 -78.2358 1.0 1.0 DRYRUN002 10/1/2016 West Virginia Snapshot Day 39.1486 -79.6382 1.9 1.9 Dunkard Creek 11/15/2016 Volunteer 39.7642 -79.9383 2.1 2.1 Dutch Fork 10/18/2016 Volunteer 40.1836 -80.4865 0.9 0.9 Wendt et al., 2018 - Test East Van Buren point 7/3/2016 16:20 Dataset 42.4455 -79.4197 11.6 11.6 EAHICR001 4/10/2018 ANF Snapshot Day 2 41.6419 -79.3371 0.8 EAHICR001 10/30/2018 ANF Snapshot Day 3 41.6419 -79.3371 1.8 1.3 EAHICR002a 4/10/2018 ANF Snapshot Day 2 41.7074 -79.2499 0.2 0.2 EAHICR002b 10/30/2018 ANF Snapshot Day 3 41.7061 -79.2523 6.3 6.3 East Branch of Cowley Run 9/26/2015 Wendt et al., 2018 41.5968 -78.1834 0.6 0.6

211

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane EBMICR001 4/10/2018 ANF Snapshot Day 2 41.4282 -79.0313 0.2 0.2 EBMICR002 10/30/2018 ANF Snapshot Day 3 41.4379 -79.0077 1.5 1.5 EBMiSt002 4/10/2018 ANF Snapshot Day 2 41.4379 -79.0077 0.4 0.4 EBSPCR001 4/10/2018 ANF Snapshot Day 2 41.5616 -78.9375 1.7 EBSPCR001 10/30/2018 ANF Snapshot Day 3 41.5616 -78.9375 1.0 1.4 EBTC04 10/30/2018 ANF Snapshot Day 3 41.6359 -78.8502 9.2 9.2 EBTC 2 7/10/2018 PSU 41.6403 -78.8558 22.5 22.5 -34.1 0.99 EBTC 4 7/10/2018 PSU 41.6396 -78.8541 36.3 36.3 -35.6 1.42 EBTC1 4/10/2018 ANF Snapshot Day 2 41.6886 -78.9240 4.3 EBTC1 7/10/2018 PSU 41.6885 -78.9233 2.5 EBTC1 10/30/2018 ANF Snapshot Day 3 41.6886 -78.9240 4.1 3.6 EBTC3 4/10/2018 ANF Snapshot Day 2 41.6428 -78.8600 12.7 -34.9 EBTC3 7/10/2018 PSU 41.6428 -78.8600 6.0 -32.7 EBTC3 10/30/2018 ANF Snapshot Day 3 41.6419 -78.8586 15.7 11.5 -43.99 Eddy Lick Run 9/16/2015 Wendt et al., 2018 41.1137 -77.8123 0.4 0.4 EFGLFO001 10/1/2016 West Virginia Snapshot Day 38.7735 -79.7034 1.7 1.7 EFGRRI002 10/1/2016 West Virginia Snapshot Day 38.5800 -79.7053 0.4 0.4 Elk Creek Lower 2/18/2017 Volunteer 42.0214 -80.3742 0.6 0.6 Elk Creek Middle 2/18/2017 Volunteer 41.9942 -80.3260 0.3 0.3 Elk Creek Upper 2/18/2017 Volunteer 41.9994 -80.1549 0.6 0.6 Elk Creek Upper 2 2/18/2017 Volunteer 41.9985 -80.0616 3.0 3.0 ELKCRE001 4/10/2018 ANF Snapshot Day 2 41.4213 -78.5755 0.8 0.8 ELKCRE002 4/10/2018 ANF Snapshot Day 2 41.4256 -78.7270 1.1 1.1 ELKHRU01 10/30/2018 ANF Snapshot Day 3 41.7620 -79.1756 10.0 10.1 -52.7 Elk Lake Stream 7/5/2016 SRBC 41.7801 -76.0455 2.5 -48.3 Elk Lake Stream 10/19/2016 SRBC 41.7801 -76.0455 2.0 2.2 -47.0 Elklick Run 9/26/2015 Wendt et al., 2018 41.5218 -78.0265 1.0 1.0

212

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane ELKRU003 6/24/2017 Pine Creek Snapshot Day 41.7121 -77.5784 1.2 1.2 Elk Run 7/11/2016 SRBC 41.8481 -77.0123 11.8 -55.3 Elk Run 10/20/2016 SRBC 41.8481 -77.0123 7.8 9.8 -48.4 ENGLRU001 6/24/2017 Pine Creek Snapshot Day 41.3542 -77.3477 0.2 0.2 ENGLRU002 6/24/2017 Pine Creek Snapshot Day 41.4439 -77.2976 0.2 0.2 Wendt et al., 2018 - Test Erie #1 (Walnut Creek) 5/30/2016 Dataset 42.0618 -80.0268 7.4 Wendt et al., 2018 - Test Erie #1 (Walnut Creek) 5/30/2016 Dataset 42.0618 -80.0268 7.1 7.3 -43.6 Wendt et al., 2018 - Test Erie #2 (Trib. 1 Walnut Creek) 5/30/2016 Dataset 42.0610 -80.0572 14.6 Wendt et al., 2018 - Test Erie #2 (Trib. 1 Walnut Creek) 5/30/2016 Dataset 42.0610 -80.0572 25.5 20.0 -56.9 Wendt et al., 2018 - Test Erie #3 (Trib. 2 Walnut Creek) 5/30/2016 Dataset 42.0458 -80.0708 2.5 Wendt et al., 2018 - Test Erie #3 (Trib. 2 Walnut Creek) 5/30/2016 Dataset 42.0458 -80.0708 4.9 3.7 -34.7 Fall Brook near Franklin Forks, PA Anna Wendt 41.9117 -75.8629 0.3 0.3 FARNBR001 4/10/2018 ANF Snapshot Day 2 41.7436 -79.1392 0.7 FARNBR002 4/10/2018 ANF Snapshot Day 2 41.7111 -79.1741 0.2 0.5 Fern Hollow 1/14/2016 Wendt et al., 2018 40.5734 -80.1580 2.3 Fern Hollow 1/14/2016 Wendt et al., 2018 40.5734 -80.1580 2.2 2.2 FIBIFO001 6/24/2017 Pine Creek Snapshot Day 41.4006 -77.4946 0.4 0.4 FILIRU001 10/1/2016 West Virginia Snapshot Day 38.7416 -79.6853 0.4 0.4 First Fork Larrys Creek near Salladasburg, PA Anna Wendt 41.2802 -77.2702 0.1 0.1 First Fork Larrys 10/20/2016 SRBC 41.2673 -77.2353 0.02 0.02 First Fork Sinnemahoning Creek (@ SP) 9/26/2015 Wendt et al., 2018 41.4510 -78.0467 2.8 2.8 FLICKRU001 6/24/2017 Pine Creek Snapshot Day 41.4815 -77.2114 0.4 0.4 FoolCr001 4/10/2018 ANF Snapshot Day 2 41.6519 -79.1475 0.4

213

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane FoolCr001 10/30/2018 ANF Snapshot Day 3 41.6519 -79.1475 0.8 0.6 ForkRu01 10/30/2018 ANF Snapshot Day 3 41.5443 -79.2654 0.5 0.5 FOXRUN001 10/1/2016 West Virginia Snapshot Day 38.6905 -79.7907 0.6 0.6 Freeman Run 9/26/2015 Wendt et al., 2018 41.6006 -78.0643 0.7 0.7 French Lick Run 7/11/2016 SRBC 41.5599 -76.9801 1.4 French Lick Run 10/20/2016 SRBC 41.5599 -76.9801 0.3 0.8 GAMBRU001 6/24/2017 Pine Creek Snapshot Day 41.2414 -77.3365 0.3 0.3 GANDCR001 10/1/2016 West Virginia Snapshot Day 38.7430 -79.6022 0.5 0.5 GENEFO002 6/24/2017 Pine Creek Snapshot Day 41.8273 -77.7061 2.9 2.9 GLADFO002 10/1/2016 West Virginia Snapshot Day 38.9573 -79.6041 0.9 0.9 GLADRU001 4/10/2018 ANF Snapshot Day 2 41.6939 -78.7433 0.9 0.9 GRANBR001 10/1/2016 West Virginia Snapshot Day 38.7285 -79.6147 0.2 0.2 Grays Run near Fields Station, PA Anna Wendt 41.4236 -77.0226 0.1 0.1 Gregs Run 6/4/2018 PSU 41.2430 -76.7071 6.0 Gregs Run (repeat sample) 6/4/2018 PSU 41.2430 -76.7071 6.7 6.4 -34.4 GRUNRU001 4/10/2018 ANF Snapshot Day 2 41.8366 -79.2215 0.2 0.2 GURGRU001 10/24/2017 ANF Snapshot Day 1 41.4373 -79.0109 2.1 GURGRU001 4/10/2018 ANF Snapshot Day 2 41.4373 -79.0109 0.6 GURGRU001 10/30/2018 ANF Snapshot Day 3 41.4378 -79.0131 1.7 1.4 Hagerman Run 7/29/2015 Wendt et al., 2018 41.4216 -77.0487 0.1 0.1 Hamilton Run 1 11/15/2016 Volunteer 39.7390 -80.4147 0.6 0.6 Hamilton Run 2 11/15/2016 Volunteer 39.7580 -80.4235 0.4 0.4 HASTRu001 4/10/2018 ANF Snapshot Day 2 41.6055 -79.0996 0.2 HASTRu001 10/30/2018 ANF Snapshot Day 3 41.6055 -79.0996 0.2 0.2 Hayes Run 11/9/2015 Wendt et al., 2018 41.1053 -77.7587 0.2 Hayes Run 5/5/2016 Wendt et al., 2018 41.1053 -77.7587 <0.06 0.2 HEDGRU001 4/10/2018 ANF Snapshot Day 2 41.7851 -79.2808 0.2 0.2

214

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane HILERU001 10/1/2016 West Virginia Snapshot Day 39.1850 -79.6005 0.4 0.4 Hoagland Run 7/11/2016 SRBC 41.3271 -77.1162 1.7 Hoagland Run 10/20/2016 SRBC 41.3271 -77.1162 0.0 0.8 HOFFRU001 10/24/2017 ANF Snapshot Day 1 41.6025 -78.7357 2.7 HOFFRU001 4/10/2018 ANF Snapshot Day 2 41.6025 -78.7357 0.7 1.7 HORSRU003 10/1/2016 West Virginia Snapshot Day 39.1821 -79.6020 3.0 3.0 Horton Run 9/26/2015 Wendt et al., 2018 41.6161 -77.8748 4.2 4.2 HR 1 11/2/2016 PSU 41.0851 -77.7895 0.2 0.2 HUNTCR005 10/24/2017 ANF Snapshot Day 1 41.5096 -78.9336 8.6 -52 HUNTCR005 4/10/2018 ANF Snapshot Day 2 41.5096 -78.9336 1.2 HUNTCR005 10/30/2018 ANF Snapshot Day 3 41.5096 -78.9336 2.2 4.0 HUNTCR006 4/10/2018 ANF Snapshot Day 2 41.5193 -78.9237 6.0 -51.2 HUNTCR006 7/11/2018 PSU 41.5194 -78.9235 7.2 -50.3 HUNTCR006 10/30/2018 ANF Snapshot Day 3 41.5193 -78.9237 3.2 5.5 HUNTCR007 4/10/2018 ANF Snapshot Day 2 41.5104 -78.9335 0.7 0.7 HUNTCR008 4/10/2018 ANF Snapshot Day 2 41.5089 -78.9339 1.1 1.1 HUNTCR009 7/11/2018 PSU 41.5188 -78.9232 5.8 5.8 IndiRu01 10/30/2018 ANF Snapshot Day 3 41.8255 -78.8555 0.2 0.2 IRBRRU001 10/1/2016 West Virginia Snapshot Day 38.6395 -79.8051 0.1 0.1 IrwinR01 10/30/2018 ANF Snapshot Day 3 41.4018 -78.9045 2.9 2.9 Johnathan Run 4/11/2016 Wendt et al., 2018 41.0200 -77.8824 0.4 0.4 JOHNBR001 6/24/2017 Pine Creek Snapshot Day 41.7433 -77.6209 0.2 0.2 KINUCR005 10/30/2018 ANF Snapshot Day 3 41.7467 -78.6213 1.3 1.3 KINZCR003 10/24/2017 ANF Snapshot Day 1 41.7646 -78.7161 4.1 KINZCR003 4/10/2018 ANF Snapshot Day 2 41.7646 -78.7161 1.6 KINZCR003 10/30/2018 ANF Snapshot Day 3 41.7646 -78.7161 2.3 2.7 KINZCR004 10/24/2017 ANF Snapshot Day 1 41.7704 -78.8622 2.2

215

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane KINZCR004 4/10/2018 ANF Snapshot Day 2 41.7704 -78.8622 1.6 1.9 KINZCR005 7/10/2018 PSU + ANF SERVICE 41.7667 -78.7183 1.8 1.8 KINZHW001 4/10/2018 ANF Snapshot Day 2 41.7932 -78.5720 12.9 12.9 -57.8 Lake Creek near Lawton, PA PSU 41.7801 -76.0454 1.7 1.7 Larrys Creek 7/11/2016 SRBC 41.2686 -77.2329 1.5 1.5 LAREFO001 10/1/2016 West Virginia Snapshot Day 38.7400 -79.6925 0.7 0.7 Larrys Creek near White Pine, PA Anna Wendt 41.3985 -77.2004 0.9 0.9 Laurel Cr 005 10/6/2016 Volunteer 39.9157 -80.0865 0.2 0.2 Laurel Run Wendt et al., 2018 - Test 3/9/2016 Dataset 41.1668 -78.5394 0.4 Laurel Run Wendt et al., 2018 - Test 3/9/2016 Dataset 41.1668 -78.5392 0.4 Laurel Run Wendt et al., 2018 - Test 3/9/2016 Dataset 41.1669 -78.5394 0.3 Laurel Run Wendt et al., 2018 - Test 3/9/2016 Dataset 41.1669 -78.5393 0.4 Laurel Run Wendt et al., 2018 - Test 3/9/2016 Dataset 41.1670 -78.5391 8.1 Laurel Run Wendt et al., 2018 - Test 3/9/2016 Dataset 41.1671 -78.5394 34.3 7.3 Laurel Run 2 4/13/2017 PSU + David Matta 41.1865 -78.5144 0.1 0.1 Laurel Run 3 4/13/2017 PSU + David Matta 41.1670 -78.5394 10.7 10.7 Laurel Run 4 4/13/2017 PSU + David Matta 41.1636 -78.5344 0.0 0.0 LAURRU004 10/1/2016 West Virginia Snapshot Day 39.1936 -79.5844 0.3 0.3 LEADRU001 10/1/2016 West Virginia Snapshot Day 39.2112 -79.5723 0.3 0.3 LEWIRU002 10/24/2017 ANF Snapshot Day 1 41.8438 -78.6926 1.1 1.1 LEWIRU003 4/10/2018 ANF Snapshot Day 2 41.8431 -78.6942 0.6 LEWIRU003 10/30/2018 ANF Snapshot Day 3 41.8431 -78.6942 1.2 0.9 LEWIRU004 4/10/2018 ANF Snapshot Day 2 41.8424 -78.6935 0.4 LEWIRU004 10/30/2018 ANF Snapshot Day 3 41.8424 -78.6935 0.8 0.6

216

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane LFCLRU002 10/1/2016 West Virginia Snapshot Day 39.1174 -79.7431 1.1 1.1 LIARRU001 10/24/2017 ANF Snapshot Day 1 41.7403 -79.0851 1.0 LIARRU001 4/10/2018 ANF Snapshot Day 2 41.7403 -79.0851 0.3 LIARRU001 10/30/2018 ANF Snapshot Day 3 41.7403 -79.0851 0.4 0.6 LIBLFO001 10/1/2016 West Virginia Snapshot Day 38.9827 -79.7297 0.2 0.2 Lick Island Run 9/26/2015 Wendt et al., 2018 41.3728 -78.0526 0.2 Lick Island Run 9/26/2015 Wendt et al., 2018 41.3728 -78.0526 0.4 0.3 LICKRU003 6/24/2017 Pine Creek Snapshot Day 41.4323 -77.2711 0.3 0.3 LIFACR001 6/24/2017 Pine Creek Snapshot Day 41.5697 -77.2045 1.2 1.2 LIPICR001 6/24/2017 Pine Creek Snapshot Day 41.4128 -77.3212 0.4 LIPICR002 6/24/2017 Pine Creek Snapshot Day 41.3100 -77.3626 0.9 0.7 LISACR001 10/24/2017 ANF Snapshot Day 1 41.5099 -79.1515 5.1 LISACR001 4/10/2018 ANF Snapshot Day 2 41.5099 -79.1515 2.4 3.8 LISLRU001 6/24/2017 Pine Creek Snapshot Day 41.4637 -77.5055 0.3 0.3 PSU + Chartiers Creek Little Chartier 1 11/7/2016 Watershed 40.2372 -80.1382 2.0 PSU + Chartiers Creek Little Chartier 1b 11/7/2016 Watershed 40.2373 -80.1388 1.8 1.9 PSU + Chartiers Creek Little Chartier 3 11/7/2016 Watershed 40.2283 -80.1439 3.5 3.5 Little Chartiers 8/6/2015 Wendt et al., 2018 40.2282 -80.1436 2.0 Little Chartiers 6/22/2015 Wendt et al., 2018 40.2280 -80.1436 2.7 2.4 Little Chartiers 8/6/2015 Wendt et al., 2018 40.1570 -80.1340 2.9 2.9 Little Chartiers 8/6/2015 Wendt et al., 2018 40.1820 -80.1460 2.6 2.6 Little Chartiers Creek 10/26/2015 Wendt et al., 2018 40.1713 -80.1364 4.2 Little Chartiers Creek 11/15/2015 Wendt et al., 2018 40.1713 -80.1363 6.7 5.4 Little Chartiers Creek 11/15/2015 Wendt et al., 2018 40.1630 -80.1336 2.6 2.6 Little Chartiers Creek 11/15/2015 Wendt et al., 2018 40.1778 -80.1362 2.5 Little Chartiers Creek 11/15/2015 Wendt et al., 2018 40.1780 -80.1364 4.1 3.3

217

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane Little Chartiers Creek 11/15/2015 Wendt et al., 2018 40.1953 -80.1363 3.2 3.2 Little Moores Run 9/26/2015 Wendt et al., 2018 41.6429 -78.0022 0.4 0.4 Little Muncy Creek downstream (near Frenchtown, PA) PSU 41.2073 -76.6353 5.0 5.0 Little Muncy Creek upstream (near Biggerstown, PA) PSU 41.2427 -76.5609 0.5 0.5 Little Portage Creek 9/26/2015 Wendt et al., 2018 41.6044 -78.0667 0.6 0.6 Little Salmon Creek 145 4/10/2018 ANF Snapshot Day 2 41.4995 -79.2032 0.8 0.8 Little Sandy Run 9/16/2015 Wendt et al., 2018 41.0758 -77.9611 1.2 Little Sandy Run 11/10/2015 Wendt et al., 2018 41.0758 -77.9611 0.2 0.7 LITTRI002 10/1/2016 West Virginia Snapshot Day 38.5306 -79.7269 0.3 0.3 LITTRI003 10/1/2016 West Virginia Snapshot Day 38.6164 -79.8066 1.9 1.9 Logan Run 4/10/2018 ANF Snapshot Day 2 41.5824 -79.1373 0.2 Logan Run 10/30/2018 ANF Snapshot Day 3 41.5824 -79.1373 0.5 0.4 LONGRU001 6/24/2017 Pine Creek Snapshot Day 41.7651 -77.5612 0.4 0.4 LONGRU003 6/24/2017 Pine Creek Snapshot Day 41.5967 -77.2895 0.2 0.2 LOUKRU001 10/1/2016 West Virginia Snapshot Day 38.7736 -79.7030 0.6 0.6 LOVERU001 6/24/2017 Pine Creek Snapshot Day 41.3662 -77.3563 0.2 0.2 Lower Hunts Run 9/26/2015 Wendt et al., 2018 41.4527 -78.1744 0.4 0.4 Lower Mingo Creek 10/18/2016 Volunteer 40.2028 -80.0095 0.3 0.3 LPBORU001 6/24/2017 Pine Creek Snapshot Day 41.3038 -77.3950 0.4 0.4 LSR 1 11/2/2016 PSU 41.0522 -77.9437 3.8 3.8 LSR 2 11/2/2016 PSU 41.0540 -77.9437 3.6 3.6 LYMARU001 6/24/2017 Pine Creek Snapshot Day 41.7210 -77.7762 1.1 1.1 MANOFO001 6/24/2017 Pine Creek Snapshot Day 41.5121 -77.5363 1.9 1.9 MARTRU01 10/30/2018 ANF Snapshot Day 3 41.6248 -78.8686 0.3 0.3 MARTRU02 10/30/2018 ANF Snapshot Day 3 41.6146 -78.8767 2.4 2.4 Marrow 1/14/2016 Wendt et al., 2018 40.5583 -80.2006 0.4

218

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane Marrow 1/14/2016 Wendt et al., 2018 40.5583 -80.2006 0.5 0.5 MAXWRU001 10/1/2016 West Virginia Snapshot Day 39.1683 -79.6148 0.2 0.2 MCGERU001 10/1/2016 West Virginia Snapshot Day 38.7073 -79.8378 2.2 2.2 McKinnon Branch 9/26/2015 Wendt et al., 2018 41.4643 -78.1732 0.7 0.7 McLaughlin Run 11/15/2016 Volunteer 40.3549 -80.1037 0.2 0.2 MEADRU001 10/24/2017 ANF Snapshot Day 1 41.7405 -78.7892 0.7 MEADRU001 4/10/2018 ANF Snapshot Day 2 41.7405 -78.7892 0.2 0.4 MEADRU002 4/10/2018 ANF Snapshot Day 2 41.7308 -78.7325 0.8 0.8 MEADRU003 10/24/2017 ANF Snapshot Day 1 41.6291 -79.0350 0.3 MEADRU003 4/10/2018 ANF Snapshot Day 2 41.6291 -79.0350 0.2 MEADRU003 10/30/2018 ANF Snapshot Day 3 41.6291 -79.0350 0.1 0.2 Meshoppen 0 Jul-16 SRBC 41.6827 -75.8872 21.6 -37.4 Meshoppen 0 Oct-16 SRBC 41.6827 -75.8872 11.4 -31.1 Meshoppen 0 May-17 SRBC 41.6827 -75.8872 16.5 -57.8 Meshoppen 0 Jul-17 SRBC 41.6827 -75.8872 13.2 -58.2 Meshoppen 0 Oct-17 SRBC 41.6827 -75.8872 13.8 15.3 -49.8 Meshoppen Bog stream 9/5/2018 PSU+SRBC 41.7179 -75.8708 46.8 46.8 -43.2 Meshoppen DS 1 May-17 SRBC 41.6818 -75.8874 13.5 13.5 -57.0 Meshoppen DS 1.5 Jul-17 SRBC 41.6623 -75.9074 3.8 -49.3 Meshoppen DS 1.5 Oct-17 SRBC 41.6623 -75.9074 6.5 5.1 -46.5 Meshoppen DS 2 7/17/2017 PSU 41.6229 -75.9534 4.9 4.9 Meshoppen mouth 7/19/2016 Wendt et al., 2018 41.6138 -76.0484 0.7 0.7 Trib Meshoppen Creek (MC1 Trib) 11/14/2013 Grieve et al., (87) 41.7180 -75.8710 0.1 0.1 Meshoppen trib 1b 7/17/2017 PSU 41.7305 -75.8780 0.4 0.4 Meshoppen US 1 May-17 SRBC 41.6834 -75.8888 12.0 12.0 -57.0 Meshoppen US 1.5 Jul-17 SRBC 41.6950 -75.8873 8.3 -35.21 Meshoppen US 1.5 7/17/2017 PSU 41.6951 -75.8875 9.5 -22.9

219

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane Meshoppen US 1.5 10/17/2018 SRBC 41.6951 -75.8875 14.7 10.8 -39.2 Meshoppen US 2 6/26/2013 Heilweil et al., 2015 41.7174 -75.8708 68.5 Meshoppen US 2 11/14/2013 Grieve et al., (87) 41.7174 -75.8708 11.6 Meshoppen US 2 May-17 SRBC 41.7173 -75.8708 10.4 -53 Meshoppen US 2 Jul-17 SRBC 41.7173 -75.8708 9.1 -38.03 Meshoppen US 2 7/17/2017 SRBC 41.7173 -75.8708 9.1 -36.39 Meshoppen US 2 7/17/2017 PSU 41.7174 -75.8710 10.6 -13.7 Meshoppen US 2 9/5/2018 PSU+SRBC 41.7173 -75.8708 30.0 21.3 -39.5 Meshoppen US 3 7/17/2017 PSU 41.7495 -75.8504 15.6 15.6 -30.61 Meshoppen US 4 7/17/2017 PSU 41.8035 -75.8384 3.0 3.0 MESSRU001 10/24/2017 ANF Snapshot Day 1 41.6351 -79.0416 0.3 MESSRU001 4/10/2018 ANF Snapshot Day 2 41.6351 -79.0416 0.2 0.3 MFEBCR001 4/10/2018 ANF Snapshot Day 2 41.5513 -78.5968 0.2 MFEBCR001 10/30/2018 ANF Snapshot Day 3 41.5513 -78.5968 0.3 0.2 Middle Creek 10/18/2016 Volunteer 40.0666 -80.5162 2.1 2.1 Middle Hunts Run 9/26/2015 Wendt et al., 2018 41.4738 -78.1511 0.9 0.9 Middle Mingo Creek 10/18/2016 Volunteer 40.1921 -80.0405 0.6 0.6 MIHORU001 6/24/2017 Pine Creek Snapshot Day 41.5555 -77.4584 0.2 0.2 MIKERU001 10/1/2016 West Virginia Snapshot Day 38.6965 -79.7864 0.7 0.7 Mill Creek West 7/29/2015 Wendt et al., 2018 41.3453 -76.9719 0.2 0.2 Miller Run 1 12/3/2016 Volunteer + PSU 40.3434 -80.1699 1.7 1.7 Miller Run Trib 1 12/3/2016 Volunteer + PSU 40.3435 -80.1700 22.0 22.0 MILLRU002 6/24/2017 Pine Creek Snapshot Day 41.4047 -77.4647 0.5 0.5 MILLRU003 6/24/2017 Pine Creek Snapshot Day 41.7528 -77.5033 0.3 0.3 MINERU001 10/1/2016 West Virginia Snapshot Day 39.1810 -79.6790 0.5 0.5 Mingo Creek 10/26/2015 Wendt et al., 2018 40.1953 -80.0421 1.0 1.0 MINICR002 4/10/2018 ANF Snapshot Day 2 41.6220 -79.1550 0.4

220

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane MINICR002 10/30/2018 ANF Snapshot Day 3 41.6220 -79.1550 0.9 0.6 Montour Run 9/26/2015 Wendt et al., 2018 41.3066 -78.0167 0.2 Montour Run 9/26/2015 Wendt et al., 2018 41.3066 -78.0167 0.3 0.2 Monument Run 10/12/2015 Wendt et al., 2018 41.1133 -77.7044 0.3 Monument Run 4/11/2016 Wendt et al., 2018 41.1133 -77.7044 0.1 0.2 Morrison run downstream of w/fall 7/11/2018 PSU 41.7913 -79.1695 2.2 2.2 Morrison run upstream of w/fall 7/11/2018 PSU 41.7915 -79.1703 2.1 2.1 MORRRU01 10/30/2018 ANF Snapshot Day 3 41.8008 -79.1435 1.7 1.7 MORRRU02 10/30/2018 ANF Snapshot Day 3 41.8185 -79.1110 0.2 0.2 Moss 1 12/3/2016 Volunteer + PSU 40.4058 -79.7620 0.4 0.4 Moss 2 12/3/2016 Volunteer + PSU 40.4076 -79.7614 8.5 8.5 Moss 3 12/3/2016 Volunteer + PSU 40.4083 -79.7611 3.7 3.7 MUDDFO001 10/24/2017 ANF Snapshot Day 1 41.4291 -79.0350 0.4 0.4 MUDDFO001 4/10/2018 ANF Snapshot Day 2 41.4291 -79.0350 0.3 0.3 Muddy Creek 11/15/2016 Volunteer 39.8774 -79.9992 6.1 6.1 Murray Creek 7/11/2016 SRBC 41.3913 -76.9461 1.0 Murray Creek 10/20/2016 SRBC 41.3913 -76.9461 -0.5 0.3 NANSBR001 10/1/2016 West Virginia Snapshot Day 38.7487 -79.5961 0.1 0.1 NARRRU001 10/1/2016 West Virginia Snapshot Day 38.7209 -79.6195 0.1 0.1 NAVARU001 6/24/2017 Pine Creek Snapshot Day 41.4581 -77.5165 0.5 0.5 NBSIXR01 10/30/2018 ANF Snapshot Day 3 41.7591 -79.0529 0.4 0.4 NBSR 1 7/17/2017 PSU 41.6386 -76.2949 4.2 4.2 -44.5 NBSR 2 7/19/2016 PSU 41.6405 -76.2947 1.0 1.0 NBSURU001 10/24/2017 ANF Snapshot Day 1 41.8911 -78.8891 1.0 1.0 NFBC 11/2/2016 PSU 41.0508 -77.9398 3.7 3.7 NFCHFO001 4/10/2018 ANF Snapshot Day 2 41.8185 -78.8581 1.1 1.1

221

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane NIMIRU001 6/24/2017 Pine Creek Snapshot Day 41.7919 -77.7628 1.0 1.0 North Fork Beech Creek 8/10/2015 Wendt et al., 2018 41.0505 -77.9400 3.0 North Fork Beech Creek 11/10/2015 Wendt et al., 2018 41.0505 -77.9400 3.5 North Fork Beech Creek 3/1/2016 Wendt et al., 2018 41.0505 -77.9400 1.2 North Fork Beech Creek 5/5/2016 Wendt et al., 2018 41.0505 -77.9400 0.1 1.9 Wendt et al., 2018 - Test Oil Creek 7/12/2016 14:37 Dataset 41.6151 -79.6583 3.6 3.6 Wendt et al., 2018 - Test Oil Creek 5/30/2016 Dataset 41.6388 -79.6707 2.8 2.9 Osbourne run 10/6/2016 Volunteer 38.8946 -80.1144 0.0 0.0 OTTEBR002 4/10/2018 ANF Snapshot Day 2 41.7090 -79.1678 0.7 0.7 OTTERU001 6/24/2017 Pine Creek Snapshot Day 41.4070 -77.3347 0.2 0.2 OTTFOR001 6/24/2017 Pine Creek Snapshot Day 41.3180 -77.4155 0.2 0.2 PACARU001 10/1/2016 West Virginia Snapshot Day 38.9574 -79.6010 0.1 0.1 PACKRU001 6/24/2017 Pine Creek Snapshot Day 41.4941 -77.1504 0.6 0.6 PAINRU003 6/24/2017 Pine Creek Snapshot Day 41.7457 -77.4927 0.3 0.3 Panther Run 8/10/2015 Wendt et al., 2018 41.1115 -77.8419 0.2 Panther Run 11/10/2015 Wendt et al., 2018 41.1115 -77.8419 5.7 Panther Run 4/11/2016 Wendt et al., 2018 41.1115 -77.8419 0.1 2.0 PANTRU001 6/24/2017 Pine Creek Snapshot Day 41.3834 -77.3563 10.7 10.7 Parks Creek near Rome, PA PSU 41.8679 -76.3333 7.0 7.0 PHOERU001 6/24/2017 Pine Creek Snapshot Day 41.7716 -77.6116 0.7 0.7 PINCECR006 6/24/2017 Pine Creek Snapshot Day 41.5562 -77.3823 1.2 1.2 Pine Creek 1 7/10/2018 PSU + ANF Service 41.7735 -78.6781 0.6 0.6 PINECR005 6/24/2017 Pine Creek Snapshot Day 41.1803 -77.2784 4.1 4.1 PINECR007 6/24/2017 Pine Creek Snapshot Day 41.7902 -77.7612 1.0 1.0 PINECR008 6/24/2017 Pine Creek Snapshot Day 41.4714 -77.5028 2.1 2.1 PINERU002 10/24/2017 ANF Snapshot Day 1 41.5038 -78.8589 2.3 2.3

222

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane PINERU002B 10/30/2018 ANF Snapshot Day 3 41.5081 -78.8642 0.1 0.2 PINERU004 4/10/2018 ANF Snapshot Day 2 41.5081 -78.8642 0.1 0.1 PINERU04 10/30/2018 ANF Snapshot Day 3 41.5098 -78.7947 0.1 0.1 Pink House 1/14/2016 Wendt et al., 2018 40.5711 -80.1589 0.6 Pink House 1/14/2016 Wendt et al., 2018 40.5711 -80.1589 0.8 0.7 Plum Run 6/22/2015 Wendt et al., 2018 40.2580 -80.2190 5.3 Plum Run 1/14/2016 Wendt et al., 2018 40.2580 -80.2190 0.4 2.8 Plum Run (2) 1/14/2016 Wendt et al., 2018 40.2551 -80.2157 2.6 2.6 PSU + Chartiers Creek Plum Run #1 11/7/2016 Watershed 40.2542 -80.2143 2.8 2.8 PSU + Chartiers Creek Plum Run #2 11/7/2016 Watershed 40.2596 -80.2200 1.2 1.2 POCARU001 10/1/2016 West Virginia Snapshot Day 38.5775 -79.7033 0.3 0.3 RAMSRU001 6/24/2017 Pine Creek Snapshot Day 41.2847 -77.3241 0.4 0.4 RATTRU001 10/1/2016 West Virginia Snapshot Day 38.9743 -79.7529 0.5 0.5 REAGRU001 10/24/2017 ANF Snapshot Day 1 41.6098 -79.1053 0.2 REAGRU001 4/10/2018 ANF Snapshot Day 2 41.6098 -79.1053 0.1 REAGRU001 10/30/2018 ANF Snapshot Day 3 41.6098 -79.1053 0.1 0.1 REDCRE003 10/1/2016 West Virginia Snapshot Day 38.9734 -79.3973 0.1 0.1 REDRUN001 10/1/2016 West Virginia Snapshot Day 38.6315 -79.8806 0.3 0.3 RELIRU001 4/10/2018 ANF Snapshot Day 2 41.4856 -78.8802 1.7 RELIRU001 10/30/2018 ANF Snapshot Day 3 41.4856 -78.8802 1.5 1.6 Rockwell Creek 7/6/2016 SRBC 41.7681 -76.1556 0.9 Rockwell Creek 10/19/2016 SRBC 41.7681 -76.1556 -0.4 0.3 ROSPRU001 10/24/2017 ANF Snapshot Day 1 41.5665 -79.0479 0.3 ROSPRU001 4/10/2018 ANF Snapshot Day 2 41.5665 -79.0479 0.1 0.2 RRDRFO001 10/1/2016 West Virginia Snapshot Day 39.0373 -79.5949 0.2 0.2 Rummerfield Creek 7/6/2016 SRBC 41.7454 -76.3088 1.0

223

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane Rummerfield Creek 10/19/2016 SRBC 41.7454 -76.3088 -0.6 0.2 SAH-13-10 Tunkhannock Creek 7/31/2013 Wendt et al., 2018 41.7033 -75.6709 1.3 SAH-13-10 Tunkhannock Creek 7/31/2013 Wendt et al., 2018 41.7033 -75.6709 1.3 1.3 SAH-13-9 Tunkhannock Creek 5/30/2013 Wendt et al., 2018 41.7070 -75.6719 1.6 SAH-13-11 Tunkhannock Creek 7/31/2013 Wendt et al., 2018 41.7072 -75.6721 1.0 SAH-13-11 Tunkhannock Creek 7/31/2013 Wendt et al., 2018 41.7072 -75.6721 1.5 1.4 SAH-13-8 Tunkhannock Creek 5/30/2013 Wendt et al., 2018 41.7102 -75.6716 1.9 SAH-13-12 Tunkhannock Creek 7/31/2013 Wendt et al., 2018 41.7104 -75.6718 2.2 SAH-13-12 Tunkhannock Creek 7/31/2013 Wendt et al., 2018 41.7104 -75.6718 2.2 SAH-13-12 Tunkhannock Creek 7/31/2013 Wendt et al., 2018 41.7104 -75.6718 2.2 2.1 SAH-13-7 Tunkhannock Creek 5/30/2013 Wendt et al., 2018 41.7107 -75.6718 2.6 2.6 SAH-13-13 9 Partners 5/30/2013 Wendt et al., 2018 41.7119 -75.6713 5.3 5.3 SAH-13-14 Tunkhannock Creek 7/31/2013 Wendt et al., 2018 41.7122 -75.6698 0.4 0.4 SAH-13-24 Tunkhannock Creek 11/13/2013 Wendt et al., 2018 41.7120 -75.6700 0.2 0.2 SAH-13-25 9 Partners 11/13/2013 Wendt et al., 2018 41.7119 -75.6715 1.6 1.6 SAH-13-26 9 Partners 11/13/2013 Wendt et al., 2018 41.7124 -75.6716 1.6 1.6 SAH-13-27 9 Partners 11/13/2013 Wendt et al., 2018 41.7132 -75.6724 1.3 1.3 SAH-13-28 9 Partners 11/13/2013 Wendt et al., 2018 41.7138 -75.6730 1.4 1.4 SAH-13-29 9 Partners 11/13/2013 Wendt et al., 2018 41.7142 -75.6737 1.5 1.5 SAH-13-30 9 Partners 11/13/2013 Wendt et al., 2018 41.7154 -75.6851 1.5 1.5 SAH-13-19 9 Partners 9/1/2013 Wendt et al., 2018 41.7136 -75.6726 2.5 2.5 SAH-13-18 Tunkhannock Creek 8/1/2013 Wendt et al., 2018 41.7150 -75.6683 0.5 SAH-13-18 Tunkhannock Creek 8/1/2013 Wendt et al., 2018 41.7150 -75.6683 0.5 0.5 SAH-13-16 Tunkhannock Creek 7/31/2013 Wendt et al., 2018 41.7173 -75.6983 1.6 SAH-13-17 Tunkhannock Creek 8/1/2013 Wendt et al., 2018 41.7169 -75.6635 0.8 SAH-13-17 Tunkhannock Creek 8/1/2013 Wendt et al., 2018 41.7169 -75.6635 1.0 1.1 SAH-13-15 Trib Tunkhannock Creek 7/31/2013 Wendt et al., 2018 41.7177 -75.6600 0.1 0.1

224

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane SAH-13-6 Tunkhannock Creek 5/30/2013 Wendt et al., 2018 41.7185 -75.6504 0.7 0.7 SAH-13-5 Tunkhannock Creek 5/30/2013 Wendt et al., 2018 41.7204 -75.6487 0.9 0.9 SAH-13-4 Tunkhannock Creek 5/30/2013 Wendt et al., 2018 41.7231 -75.6463 0.7 0.7 SAH-13-20 9 Partners 9/1/2013 Wendt et al., 2018 41.7286 -75.6762 0.4 0.4 SAH-13-21 9 Partners 9/1/2013 Wendt et al., 2018 41.7289 -75.6765 0.4 0.4 SAH-13-1 Tunkhannock Creek 5/30/2013 Wendt et al., 2018 41.7330 -75.6319 0.7 0.7 SAH-13-2 Tunkhannock Creek 5/30/2013 Wendt et al., 2018 41.7334 -75.6299 0.6 0.6 SAH-13-3 Tunkhannock Creek 5/30/2013 Wendt et al., 2018 41.7330 -75.6334 1.0 1.0 SAH-13-22 9 Partners 9/1/2013 Wendt et al., 2018 41.7631 -75.6872 0.2 0.2 SAH-13-23 9 Partners 9/1/2013 Wendt et al., 2018 41.7866 -75.6865 2.6 2.6 SALMCR002 4/10/2018 ANF Snapshot Day 2 41.5067 -79.1220 1.7 1.7 Salmon Creek 4/10/2018 ANF Snapshot Day 2 41.5388 -79.2585 1.2 1.2 Salmon Creek 145 4/10/2018 ANF Snapshot Day 2 41.4708 -79.1995 2.8 2.8 Salt Lick 10/12/2015 Wendt et al., 2018 41.1050 -77.7225 0.2 Salt Lick 4/11/2016 Wendt et al., 2018 41.1050 -77.7225 0.1 0.2 Salt Lick Creek 7/5/2016 SRBC 41.9449 -75.7376 1.1 Salt Lick Creek 10/18/2016 SRBC 41.9449 -75.7376 -0.5 0.3 Salt Run 9/26/2015 Wendt et al., 2018 41.5340 -78.1952 0.5 0.5 Salt Spring Stream 1 7/17/2017 PSU 41.9119 -75.8649 0.6 0.6 Salt Spring Stream 2 7/17/2017 PSU 41.9121 -75.8635 0.6 0.6 Sandy Run 1 11/2/2016 PSU 41.0733 -77.9031 0.8 0.8 SandyRun_Kato 9/16/2015 Wendt et al., 2018 41.0782 -77.9075 4.6 SandyRun_Kato 11/10/2015 Wendt et al., 2018 41.0782 -77.9075 1.1 SandyRun_Kato 3/1/2016 Wendt et al., 2018 41.0782 -77.9075 0.4 SandyRun_Kato 5/5/2016 Wendt et al., 2018 41.0782 -77.9075 <0.06 1.5 SAR #1 10/7/2016 PSU 41.0970 -77.9440 1.0 1.0 SAR #10 10/7/2016 PSU 41.1057 -77.9460 0.7 0.7

225

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane SAR #2 10/7/2016 PSU 41.0973 -77.9438 1.5 1.5 SAR #3 10/7/2016 PSU 41.0980 -77.9442 2.0 2.0 SAR #4 10/7/2016 PSU 41.0987 -77.9445 0.7 0.7 SAR #5 10/7/2016 PSU 41.0994 -77.9451 0.6 0.6 SAR #6 10/7/2016 PSU 41.0994 -77.9451 0.9 0.9 SAR #7 10/7/2016 PSU 41.0996 -77.9450 0.5 0.5 SAR #8 10/7/2016 PSU 41.1023 -77.9456 3.1 3.1 SAR #9 10/7/2016 PSU 41.1027 -77.9458 0.4 0.4 SBKICR001 10/24/2017 ANF Snapshot Day 1 41.7113 -78.7230 3.7 SBKICR001 4/10/2018 ANF Snapshot Day 2 41.7113 -78.7230 1.2 2.5 S. Branch Towanda Creek 7/6/2016 SRBC 41.6271 -76.4341 1.4 S. Branch Towanda Creek 10/19/2016 SRBC 41.6271 -76.4341 0.4 0.9 SBWICR001 10/24/2017 ANF Snapshot Day 1 41.9500 -78.8067 0.3 0.3 SC1 7/18/2016 Anna Wendt 41.7620 -76.6993 1.8 1.8 SC2 7/18/2016 Anna Wendt 41.7626 -76.6883 4.9 4.9 SCAL 6/20/2018 PSU 40.6730 -77.9018 SCBL 6/20/2018 PSU 40.6558 -77.9199 SCO 6/20/2018 PSU 40.6106 -78.0067 SCHOHO001 6/24/2017 Pine Creek Snapshot Day 41.3954 -77.3473 0.2 0.2 SEBIFO001 6/24/2017 Pine Creek Snapshot Day 41.4012 -77.5004 0.3 0.3 SEVERU001 10/30/2018 ANF Snapshot Day 3 41.6311 -78.5770 0.4 0.4 SFRECR001 10/1/2016 West Virginia Snapshot Day 38.9683 -79.3980 0.1 0.1 Shalehillsflume 6/20/2018 PSU 40.6648 -77.9072 3.1 3.1 SHAVFO001 10/1/2016 West Virginia Snapshot Day 38.6634 -79.8629 0.7 0.7 Silver Creek at Franklin Forks, PA Anna Wendt 41.9177 -75.8480 0.6 0.6 Sinnemahoning Portage Creek (Emporium) 9/26/2015 Wendt et al., 2018 41.5131 -78.2197 0.4 0.4

226

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane SIXMIR01 10/30/2018 ANF Snapshot Day 3 41.7522 -79.0594 1.0 1.01 Sixteen mile creek 2 2/18/2017 Volunteer 42.2050 -79.8340 0.2 0.2 Sixteen mile creek lower 2/18/2017 Volunteer 42.2410 -79.8320 0.2 0.2 Slab Cabin 6/11/2015 Wendt et al., 2018 40.8090 -77.8260 1.2 Slab Cabin 6/11/2015 Wendt et al., 2018 40.8090 -77.8260 1.2 Slab Cabin 6/11/2015 Wendt et al., 2018 40.8090 -77.8260 1.2 Slab Cabin 6/11/2015 Wendt et al., 2018 40.8090 -77.8260 1.2 Slab Cabin 6/18/2015 Wendt et al., 2018 40.8090 -77.8260 0.9 Slab Cabin 6/18/2015 Wendt et al., 2018 40.8090 -77.8260 0.8 Slab Cabin 6/18/2015 Wendt et al., 2018 40.8090 -77.8260 1.4 Slab Cabin 6/18/2015 Wendt et al., 2018 40.8090 -77.8260 1.1 Slab Cabin 6/25/2015 Wendt et al., 2018 40.8090 -77.8260 0.6 Slab Cabin 6/25/2015 Wendt et al., 2018 40.8090 -77.8260 0.6 Slab Cabin 6/25/2015 Wendt et al., 2018 40.8090 -77.8260 0.6 Slab Cabin 6/25/2015 Wendt et al., 2018 40.8090 -77.8260 0.7 Slab Cabin 6/29/2015 Wendt et al., 2018 40.8090 -77.8260 1.0 Slab Cabin 6/29/2015 Wendt et al., 2018 40.8090 -77.8260 0.9 Slab Cabin 6/29/2015 Wendt et al., 2018 40.8090 -77.8260 0.8 Slab Cabin 6/29/2015 Wendt et al., 2018 40.8090 -77.8260 0.8 Slab Cabin 7/1/2015 Wendt et al., 2018 40.8090 -77.8260 0.8 Slab Cabin 7/1/2015 Wendt et al., 2018 40.8090 -77.8260 1.2 Slab Cabin 7/1/2015 Wendt et al., 2018 40.8090 -77.8260 1.2 Slab Cabin 7/1/2015 Wendt et al., 2018 40.8090 -77.8260 1.2 Slab Cabin 7/6/2015 Wendt et al., 2018 40.8090 -77.8260 1.2 Slab Cabin 7/6/2015 Wendt et al., 2018 40.8090 -77.8260 1.1 Slab Cabin 7/6/2015 Wendt et al., 2018 40.8090 -77.8260 0.9 Slab Cabin 7/6/2015 Wendt et al., 2018 40.8090 -77.8260 1.0 Slab Cabin 7/15/2015 Wendt et al., 2018 40.8090 -77.8260 0.9

227

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane Slab Cabin 7/15/2015 Wendt et al., 2018 40.8090 -77.8260 1.0 Slab Cabin 7/15/2015 Wendt et al., 2018 40.8090 -77.8260 1.0 Slab Cabin 7/15/2015 Wendt et al., 2018 40.8090 -77.8260 1.0 Slab Cabin 7/29/2015 Wendt et al., 2018 40.8090 -77.8260 0.6 Slab Cabin 7/29/2015 Wendt et al., 2018 40.8090 -77.8260 0.7 Slab Cabin 7/29/2015 Wendt et al., 2018 40.8090 -77.8260 0.6 Slab Cabin 7/29/2015 Wendt et al., 2018 40.8090 -77.8260 0.6 Slab Cabin 8/12/2015 Wendt et al., 2018 40.8090 -77.8260 0.6 Slab Cabin 8/12/2015 Wendt et al., 2018 40.8090 -77.8260 0.6 Slab Cabin 8/12/2015 Wendt et al., 2018 40.8090 -77.8260 0.8 Slab Cabin 8/25/2015 Wendt et al., 2018 40.8090 -77.8260 0.7 Slab Cabin 8/25/2015 Wendt et al., 2018 40.8090 -77.8260 0.5 Slab Cabin 8/25/2015 Wendt et al., 2018 40.8090 -77.8260 0.7 Slab Cabin 9/11/2015 Wendt et al., 2018 40.8090 -77.8260 0.9 Slab Cabin 9/11/2015 Wendt et al., 2018 40.8090 -77.8260 0.9 Slab Cabin 9/11/2015 Wendt et al., 2018 40.8090 -77.8260 0.5 Slab Cabin 10/5/2015 Wendt et al., 2018 40.8090 -77.8260 0.5 Slab Cabin 10/5/2015 Wendt et al., 2018 40.8090 -77.8260 0.5 Slab Cabin 10/5/2015 Wendt et al., 2018 40.8090 -77.8260 0.5 Slab Cabin 10/19/2015 Wendt et al., 2018 40.8090 -77.8260 0.3 Slab Cabin 10/19/2015 Wendt et al., 2018 40.8090 -77.8260 0.3 Slab Cabin 10/19/2015 Wendt et al., 2018 40.8090 -77.8260 0.7 Slab Cabin 11/9/2015 Wendt et al., 2018 40.8090 -77.8260 0.6 Slab Cabin 11/9/2015 Wendt et al., 2018 40.8090 -77.8260 0.7 Slab Cabin 11/9/2015 Wendt et al., 2018 40.8090 -77.8260 0.4 Slab Cabin 11/21/2015 Wendt et al., 2018 40.8090 -77.8260 0.3 Slab Cabin 11/21/2015 Wendt et al., 2018 40.8090 -77.8260 0.3 Slab Cabin 11/21/2015 Wendt et al., 2018 40.8090 -77.8260 0.3

228

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane Slab Cabin 12/12/2015 Wendt et al., 2018 40.8090 -77.8260 0.4 Slab Cabin 12/12/2015 Wendt et al., 2018 40.8090 -77.8260 0.3 Slab Cabin 12/12/2015 Wendt et al., 2018 40.8090 -77.8260 0.3 Slab Cabin 1/8/2016 Wendt et al., 2018 40.8090 -77.8260 0.2 Slab Cabin 1/8/2016 Wendt et al., 2018 40.8090 -77.8260 0.2 Slab Cabin 1/8/2016 Wendt et al., 2018 40.8090 -77.8260 0.2 Slab Cabin 2/3/2016 Wendt et al., 2018 40.8090 -77.8260 0.7 Slab Cabin 2/3/2016 Wendt et al., 2018 40.8090 -77.8260 0.7 Slab Cabin 2/3/2016 Wendt et al., 2018 40.8090 -77.8260 0.7 Slab Cabin 3/18/2016 Wendt et al., 2018 40.8090 -77.8260 0.3 Slab Cabin 3/18/2016 Wendt et al., 2018 40.8090 -77.8260 0.3 Slab Cabin 3/18/2016 Wendt et al., 2018 40.8090 -77.8260 0.3 Slab Cabin 4/13/2016 Wendt et al., 2018 40.8090 -77.8260 0.3 Slab Cabin 4/13/2016 Wendt et al., 2018 40.8090 -77.8260 0.3 Slab Cabin 4/13/2016 Wendt et al., 2018 40.8090 -77.8260 0.3 Slab Cabin 5/1/2016 Wendt et al., 2018 40.8090 -77.8260 1.0 Slab Cabin 5/1/2016 Wendt et al., 2018 40.8090 -77.8260 1.0 Slab Cabin 5/1/2016 Wendt et al., 2018 40.8090 -77.8260 1.0 Slab Cabin 5/16/2016 Wendt et al., 2018 40.8090 -77.8260 0.4 Slab Cabin 5/16/2016 Wendt et al., 2018 40.8090 -77.8260 0.4 Slab Cabin 6/8/2016 Wendt et al., 2018 40.8090 -77.8260 1.0 Slab Cabin 6/8/2016 Wendt et al., 2018 40.8090 -77.8260 1.0 Slab Cabin 6/8/2016 Wendt et al., 2018 40.8090 -77.8260 1.0 Slab Cabin 6/23/2016 Wendt et al., 2018 40.8090 -77.8260 0.9 Slab Cabin 6/23/2016 Wendt et al., 2018 40.8090 -77.8260 0.9 Slab Cabin 6/23/2016 Wendt et al., 2018 40.8090 -77.8260 0.8 Slab Cabin 6/30/2016 Wendt et al., 2018 40.8090 -77.8260 1.1 Slab Cabin 6/30/2016 Wendt et al., 2018 40.8090 -77.8260 0.7

229

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane Slab Cabin 6/30/2016 Wendt et al., 2018 40.8090 -77.8260 0.8 Slab Cabin 7/13/2016 Wendt et al., 2018 40.8090 -77.8260 1.1 Slab Cabin 7/13/2016 Wendt et al., 2018 40.8090 -77.8260 1.0 Slab Cabin 7/13/2016 Wendt et al., 2018 40.8090 -77.8260 1.0 Slab Cabin 7/27/2016 Wendt et al., 2018 40.8090 -77.8260 1.2 Slab Cabin 7/27/2016 Wendt et al., 2018 40.8090 -77.8260 1.2 Slab Cabin 7/27/2016 Wendt et al., 2018 40.8090 -77.8260 1.3 Slab Cabin 8/15/2016 Wendt et al., 2018 40.8090 -77.8260 0.8 Slab Cabin 8/15/2016 Wendt et al., 2018 40.8090 -77.8260 0.8 Slab Cabin 8/15/2016 Wendt et al., 2018 40.8090 -77.8260 0.8 Slab Cabin 8/28/2016 Wendt et al., 2018 40.8090 -77.8260 0.7 Slab Cabin 8/28/2016 Wendt et al., 2018 40.8090 -77.8260 0.9 Slab Cabin 8/28/2016 Wendt et al., 2018 40.8090 -77.8260 0.8 Slab Cabin 9/21/2016 Wendt et al., 2018 40.8090 -77.8260 1.0 Slab Cabin 9/21/2016 Wendt et al., 2018 40.8090 -77.8260 1.0 Slab Cabin 9/21/2016 Wendt et al., 2018 40.8090 -77.8260 1.0 Slab Cabin 10/9/2016 Wendt et al., 2018 40.8090 -77.8260 0.7 Slab Cabin 10/9/2016 Wendt et al., 2018 40.8090 -77.8260 0.7 Slab Cabin 10/9/2016 Wendt et al., 2018 40.8090 -77.8260 0.7 Slab Cabin 10/23/2016 Wendt et al., 2018 40.8090 -77.8260 0.7 Slab Cabin 10/23/2016 Wendt et al., 2018 40.8090 -77.8260 0.6 0.8 SLATRU001 6/24/2017 Pine Creek Snapshot Day 41.5250 -77.5311 0.2 0.2 SLATRU002 6/24/2017 Pine Creek Snapshot Day 41.4715 -77.5056 0.4 0.4 SLATRU003 10/24/2017 ANF Snapshot Day 1 41.5444 -79.0566 8.5 -50.1 SLATRU003 4/10/2018 ANF Snapshot Day 2 41.5444 -79.0566 1.6 5.1 SLIDRU001 10/24/2017 ANF Snapshot Day 1 41.5302 -78.7950 1.3 SLIDRU001 4/10/2018 ANF Snapshot Day 2 41.5302 -78.7950 0.2 SLIDRU001 10/30/2018 ANF Snapshot Day 3 41.5302 -78.7950 0.4 0.6

230

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane Snake Creek near Franklin Forks, PA Wendt et al., 2018 41.9275 -75.8429 0.4 0.4 SNLIRU002 10/1/2016 West Virginia Snapshot Day 38.7118 -79.7773 2.9 2.9 South Fork Beech Creek 9/16/2015 Wendt et al., 2018 41.0240 -77.9041 0.4 South Fork Beech Creek 5/5/2016 Wendt et al., 2018 41.0240 -77.9041 0.3 0.3 South Fork Cross Creek 10/18/2016 Volunteer 40.2832 -80.3650 4.1 4.1 SPC2 10/24/2017 ANF Snapshot Day 1 41.4606 -78.9701 1.3 SPC2 4/10/2018 ANF Snapshot Day 2 41.4606 -78.9701 2.9 2.1 SPC3 10/24/2017 ANF Snapshot Day 1 41.5297 -79.0147 2.2 SPC3 4/10/2018 ANF Snapshot Day 2 41.5297 -79.0147 2.1 2.1 SPC4 10/24/2017 ANF Snapshot Day 1 41.5445 -78.9941 2.1 SPC4 4/10/2018 ANF Snapshot Day 2 41.5445 -78.9941 0.4 1.3 SPENRU01 10/30/2018 ANF Snapshot Day 3 41.5262 -78.7972 0.3 0.32 Spring above W. Branch of Big Run 9/16/2015 Wendt et al., 2018 41.1464 -77.7908 0.3 Spring above West Branch-Big Run 3/1/2016 Wendt et al., 2018 41.1464 -77.7908 0.3 Spring above West Branch-Big Run 5/5/2016 Wendt et al., 2018 41.1464 -77.7908 0.0 0.2 Spring Creek 6/11/2015 Wendt et al., 2018 40.8200 -77.8300 1.7 Spring Creek 6/11/2015 Wendt et al., 2018 40.8200 -77.8300 1.6 Spring Creek 6/11/2015 Wendt et al., 2018 40.8200 -77.8300 1.2 Spring Creek 6/11/2015 Wendt et al., 2018 40.8200 -77.8300 1.6 Spring Creek 6/11/2015 Wendt et al., 2018 40.8200 -77.8300 1.7 Spring Creek 6/25/2015 Wendt et al., 2018 40.8200 -77.8300 1.1 Spring Creek 6/25/2015 Wendt et al., 2018 40.8200 -77.8300 1.1 Spring Creek 6/25/2015 Wendt et al., 2018 40.8200 -77.8300 1.0 Spring Creek 6/25/2015 Wendt et al., 2018 40.8200 -77.8300 1.1 Spring Creek 6/29/2015 Wendt et al., 2018 40.8200 -77.8300 1.3

231

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane Spring Creek 6/29/2015 Wendt et al., 2018 40.8200 -77.8300 1.1 Spring Creek 6/29/2015 Wendt et al., 2018 40.8200 -77.8300 1.3 Spring Creek 6/29/2015 Wendt et al., 2018 40.8200 -77.8300 1.6 Spring Creek 7/1/2015 Wendt et al., 2018 40.8200 -77.8300 1.4 Spring Creek 7/1/2015 Wendt et al., 2018 40.8200 -77.8300 1.4 Spring Creek 7/1/2015 Wendt et al., 2018 40.8200 -77.8300 1.3 Spring Creek 7/6/2015 Wendt et al., 2018 40.8200 -77.8300 1.2 Spring Creek 7/6/2015 Wendt et al., 2018 40.8200 -77.8300 1.7 Spring Creek 7/6/2015 Wendt et al., 2018 40.8200 -77.8300 1.5 Spring Creek 7/6/2015 Wendt et al., 2018 40.8200 -77.8300 1.5 Spring Creek 7/15/2015 Wendt et al., 2018 40.8200 -77.8300 1.2 Spring Creek 7/15/2015 Wendt et al., 2018 40.8200 -77.8300 1.2 Spring Creek 7/15/2015 Wendt et al., 2018 40.8200 -77.8300 1.2 Spring Creek 7/15/2015 Wendt et al., 2018 40.8200 -77.8300 1.1 Spring Creek 7/29/2015 Wendt et al., 2018 40.8200 -77.8300 1.3 Spring Creek 7/29/2015 Wendt et al., 2018 40.8200 -77.8300 1.2 Spring Creek 7/29/2015 Wendt et al., 2018 40.8200 -77.8300 1.1 Spring Creek 7/29/2015 Wendt et al., 2018 40.8200 -77.8300 1.3 Spring Creek 8/12/2015 Wendt et al., 2018 40.8200 -77.8300 1.5 Spring Creek 8/12/2015 Wendt et al., 2018 40.8200 -77.8300 1.5 Spring Creek 8/12/2015 Wendt et al., 2018 40.8200 -77.8300 1.7 Spring Creek 8/25/2015 Wendt et al., 2018 40.8200 -77.8300 1.8 Spring Creek 8/25/2015 Wendt et al., 2018 40.8200 -77.8300 1.7 Spring Creek 8/25/2015 Wendt et al., 2018 40.8200 -77.8300 1.8 Spring Creek 9/11/2015 Wendt et al., 2018 40.8200 -77.8300 1.4 Spring Creek 9/11/2015 Wendt et al., 2018 40.8200 -77.8300 1.0 Spring Creek 9/11/2015 Wendt et al., 2018 40.8200 -77.8300 1.6 Spring Creek 10/5/2015 Wendt et al., 2018 40.8200 -77.8300 1.1

232

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane Spring Creek 10/5/2015 Wendt et al., 2018 40.8200 -77.8300 1.0 Spring Creek 10/5/2015 Wendt et al., 2018 40.8200 -77.8300 1.0 Spring Creek 10/19/2015 Wendt et al., 2018 40.8200 -77.8300 0.3 Spring Creek 10/19/2015 Wendt et al., 2018 40.8200 -77.8300 0.7 Spring Creek 10/19/2015 Wendt et al., 2018 40.8200 -77.8300 0.7 Spring Creek 11/9/2015 Wendt et al., 2018 40.8200 -77.8300 0.4 Spring Creek 11/9/2015 Wendt et al., 2018 40.8200 -77.8300 0.4 Spring Creek 11/9/2015 Wendt et al., 2018 40.8200 -77.8300 0.3 Spring Creek 11/21/2015 Wendt et al., 2018 40.8200 -77.8300 1.0 Spring Creek 11/21/2015 Wendt et al., 2018 40.8200 -77.8300 0.9 Spring Creek 11/21/2015 Wendt et al., 2018 40.8200 -77.8300 1.1 Spring Creek 12/12/2015 Wendt et al., 2018 40.8200 -77.8300 0.7 Spring Creek 12/12/2015 Wendt et al., 2018 40.8200 -77.8300 0.5 Spring Creek 12/12/2015 Wendt et al., 2018 40.8200 -77.8300 0.6 Spring Creek 1/8/2016 Wendt et al., 2018 40.8200 -77.8300 0.5 Spring Creek 1/8/2016 Wendt et al., 2018 40.8200 -77.8300 0.5 Spring Creek 1/8/2016 Wendt et al., 2018 40.8200 -77.8300 0.5 Spring Creek 2/3/2016 Wendt et al., 2018 40.8200 -77.8300 1.0 Spring Creek 2/3/2016 Wendt et al., 2018 40.8200 -77.8300 1.1 Spring Creek 2/3/2016 Wendt et al., 2018 40.8200 -77.8300 0.9 Spring Creek 3/18/2016 Wendt et al., 2018 40.8200 -77.8300 0.5 Spring Creek 3/18/2016 Wendt et al., 2018 40.8200 -77.8300 0.5 Spring Creek 3/18/2016 Wendt et al., 2018 40.8200 -77.8300 0.5 Spring Creek 4/13/2016 Wendt et al., 2018 40.8200 -77.8300 0.6 Spring Creek 4/13/2016 Wendt et al., 2018 40.8200 -77.8300 0.7 Spring Creek 4/13/2016 Wendt et al., 2018 40.8200 -77.8300 0.6 Spring Creek 5/1/2016 Wendt et al., 2018 40.8200 -77.8300 1.2 Spring Creek 5/1/2016 Wendt et al., 2018 40.8200 -77.8300 1.2

233

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane Spring Creek 5/1/2016 Wendt et al., 2018 40.8200 -77.8300 1.3 Spring Creek 5/16/2016 Wendt et al., 2018 40.8200 -77.8300 0.9 Spring Creek 5/16/2016 Wendt et al., 2018 40.8200 -77.8300 0.7 Spring Creek 6/8/2016 Wendt et al., 2018 40.8200 -77.8300 1.4 Spring Creek 6/8/2016 Wendt et al., 2018 40.8200 -77.8300 1.4 Spring Creek 6/8/2016 Wendt et al., 2018 40.8200 -77.8300 1.4 Spring Creek 6/23/2016 Wendt et al., 2018 40.8200 -77.8300 1.4 Spring Creek 6/23/2016 Wendt et al., 2018 40.8200 -77.8300 1.4 Spring Creek 6/23/2016 Wendt et al., 2018 40.8200 -77.8300 1.3 Spring Creek 6/30/2016 Wendt et al., 2018 40.8200 -77.8300 1.2 Spring Creek 6/30/2016 Wendt et al., 2018 40.8200 -77.8300 1.2 Spring Creek 6/30/2016 Wendt et al., 2018 40.8200 -77.8300 1.1 Spring Creek 7/13/2016 Wendt et al., 2018 40.8200 -77.8300 1.2 Spring Creek 7/13/2016 Wendt et al., 2018 40.8200 -77.8300 1.3 Spring Creek 7/13/2016 Wendt et al., 2018 40.8200 -77.8300 1.2 Spring Creek 7/27/2016 Wendt et al., 2018 40.8200 -77.8300 1.3 Spring Creek 7/27/2016 Wendt et al., 2018 40.8200 -77.8300 1.5 Spring Creek 7/27/2016 Wendt et al., 2018 40.8200 -77.8300 1.4 Spring Creek 8/15/2016 Wendt et al., 2018 40.8200 -77.8300 2.4 Spring Creek 8/15/2016 Wendt et al., 2018 40.8200 -77.8300 2.4 Spring Creek 8/15/2016 Wendt et al., 2018 40.8200 -77.8300 2.4 Spring Creek 8/28/2016 Wendt et al., 2018 40.8200 -77.8300 2.3 Spring Creek 8/28/2016 Wendt et al., 2018 40.8200 -77.8300 2.4 Spring Creek 8/28/2016 Wendt et al., 2018 40.8200 -77.8300 2.6 Spring Creek 9/21/2016 Wendt et al., 2018 40.8200 -77.8300 1.6 Spring Creek 9/21/2016 Wendt et al., 2018 40.8200 -77.8300 1.6 Spring Creek 9/21/2016 Wendt et al., 2018 40.8200 -77.8300 1.5 Spring Creek 10/9/2016 Wendt et al., 2018 40.8200 -77.8300 1.3

234

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane Spring Creek 10/9/2016 Wendt et al., 2018 40.8200 -77.8300 1.5 Spring Creek 10/9/2016 Wendt et al., 2018 40.8200 -77.8300 1.4 Spring Creek 10/23/2016 Wendt et al., 2018 40.8200 -77.8300 2.0 Spring Creek 10/23/2016 Wendt et al., 2018 40.8200 -77.8300 1.8 Spring Creek 10/23/2016 Wendt et al., 2018 40.8200 -77.8300 1.8 1.2 SR 0.4 7/10/2017 PSU 41.2308 -76.7078 0.8 0.8 - SR 0.5 7/10/2017 PSU 41.2313 -76.7071 0.8 0.8 - SR 1 5/21/2013 Heilweil et al., 2014, 2015 41.2359 -76.6956 5.0 -

SR 1 6/27/2013 Heilweil et al., 2014, 2015 41.2359 -76.6956 2.3 -

SR 1 11/12/2013 Heilweil et al., 2014, 2015 41.2359 -76.6956 1.3 -

SR 1 7/9/2015 Wendt et al., 2018 41.2359 -76.6956 5.0 - SR 1 7/9/2015 Wendt et al., 2018 41.2359 -76.6956 5.0 3.7 SR 1.1 6/13/2016 Wendt et al., 2018 41.2372 -76.6937 20.4 -

SR 1.1 8/15/2016 PSU 41.2372 -76.6937 1.0 -

SR 1.1 9/15/2016 PSU 41.2372 -76.6937 1.3 -54.4

SR 1.1 10/26/2016 PSU 41.2372 -76.6937 6.6 -13.2

SR 1.1 11/16/2016 PSU 41.2372 -76.6937 4.7 -

SR 1.1 1/16/2017 PSU 41.2372 -76.6937 11.4 -

SR 1.1 2/24/2017 PSU 41.2372 -76.6937 11.1 -

SR 1.1 5/10/2017 PSU 41.2372 -76.6937 8.3 -

SR 1.1 6/22/2017 PSU 41.2372 -76.6937 5.5 -

SR 1.1 7/10/2017 PSU 41.2372 -76.6937 1.9 -

SR 1.1 7/26/2017 PSU 41.2372 -76.6937 6.5 -

SR 1.1 9/8/2017 PSU 41.2372 -76.6937 2.3 -

SR 1.1 10/2/2017 PSU 41.2372 -76.6937 0.9 -

SR 1.1 12/3/2017 PSU 41.2372 -76.6937 8.2 6.4 - SR 1.15 7/9/2015 Wendt et al., 2018 41.2360 -76.6940 5.4 SR 1.15 7/9/2015 Wendt et al., 2018 41.2360 -76.6940 5.4

235

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane

SR 1.15 8/15/2016 PSU 41.2374 -76.6936 1.3 -

SR 1.15 9/15/2016 PSU 41.2374 -76.6936 0.3 SR 1.15 10/26/2016 PSU 41.2374 -76.6936 5.2 3.5

SR 1.2 11/12/2013 Heilweil et al., 2014, 2015 41.2377 -76.6933 6.4 -16.9

SR 1.2 7/9/2015 Wendt et al., 2018 41.2377 -76.6933 10.0 -

SR 1.2 7/9/2015 Wendt et al., 2018 41.2377 -76.6933 10.0

SR 1.2 6/13/2016 Wendt et al., 2018 41.2377 -76.6933 10.9

SR 1.2 6/27/2016 PSU 41.2377 -76.6933 10.9 -

SR 1.2 8/15/2016 PSU 41.2377 -76.6933 2.9 -

SR 1.2 9/15/2016 PSU 41.2377 -76.6933 0.7 -19.6

SR 1.2 11/16/2016 PSU 41.2377 -76.6933 8.4 -

SR 1.2 1/16/2017 PSU 41.2377 -76.6933 12.3 -

SR 1.2 2/24/2017 PSU 41.2377 -76.6933 13.1 -

SR 1.2 5/10/2017 PSU 41.2377 -76.6933 10.9 -

SR 1.2 6/22/2017 PSU 41.2377 -76.6933 8.6 8.8 - SR 1.4 11/12/2013 Heilweil et al., 2014, 2015 41.2381 -76.6930 16.8 -22.1

SR 1.4 7/9/2015 Wendt et al., 2018 41.2381 -76.6930 13.3 SR 1.4 7/9/2015 Wendt et al., 2018 41.2381 -76.6930 13.3 -

SR 1.4 6/13/2016 Wendt et al., 2018 41.2381 -76.6930 18.6

SR 1.4 8/15/2016 PSU 41.2381 -76.6930 7.1 - SR 1.4 9/15/2016 PSU 41.2381 -76.6930 1.5 -13.9 SR 1.4 10/26/2016 PSU 41.2381 -76.6930 16.3 - SR 1.4 11/16/2017 PSU 41.2381 -76.6930 12.6 -

SR 1.4 1/16/2017 PSU 41.2381 -76.6930 12.3 -

SR 1.4 2/24/2017 PSU 41.2381 -76.6930 16.6 -

SR 1.4 5/10/2017 PSU 41.2381 -76.6930 10.9 -

SR 1.4 6/22/2017 PSU 41.2381 -76.6930 17.1 13.0 - SR 1.45 12/9/2014 Wendt et al., 2018 41.2386 -76.6925 9.7

236

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane SR 1.45 7/9/2015 Wendt et al., 2018 41.2386 -76.6925 17.5 SR 1.45 7/9/2015 Wendt et al., 2018 41.2386 -76.6925 17.5 SR 1.45 6/13/2016 Wendt et al., 2018 41.2386 -76.6925 27.4 18.0 SR 1.5 6/27/2013 Heilweil et al., 2014, 2015 41.2391 -76.6923 67.0 -

SR 1.5 11/12/2013 Heilweil et al., 2014, 2015 41.2391 -76.6923 28.2 -25.2 SR 1.5 11/12/2013 Heilweil et al., 2014, 2015 41.2391 -76.6923 29.0 SR 1.5 12/9/2014 Wendt et al., 2018 41.2391 -76.6923 6.7 SR 1.5 12/9/2014 Wendt et al., 2018 41.2391 -76.6923 10.1 SR 1.5 12/9/2014 Wendt et al., 2018 41.2391 -76.6923 7.6 SR 1.5 12/9/2014 Wendt et al., 2018 41.2391 -76.6923 9.0 SR 1.5 7/9/2015 Wendt et al., 2018 41.2391 -76.6923 10.6 SR 1.5 7/9/2015 Wendt et al., 2018 41.2391 -76.6923 10.7 SR 1.5 7/9/2015 Wendt et al., 2018 41.2391 -76.6923 10.4

SR 1.5 6/13/2016 PSU 41.2391 -76.6923 22.9 -

SR 1.5 8/15/2016 PSU 41.2391 -76.6923 16.7 -

SR 1.5 9/15/2016 PSU 41.2391 -76.6923 14.6 -11.4

SR 1.5 10/26/2016 PSU 41.2391 -76.6923 61.6 -20.8

SR 1.5 11/16/2016 PSU 41.2391 -76.6923 50.2 -22.7

SR 1.5 1/16/2017 PSU 41.2391 -76.6923 9.9 -

SR 1.5 2/24/2017 PSU 41.2391 -76.6923 11.6 -

SR 1.5 5/10/2017 PSU 41.2391 -76.6923 17.6 -

SR 1.5 6/22/2017 PSU 41.2391 -76.6923 24.3 -23.8

SR 1.5 7/10/2017 PSU 41.2391 -76.6923 32.5 -

SR 1.5 9/8/2017 PSU 41.2391 -76.6923 26.8 -

SR 1.5 10/2/2017 PSU 41.2391 -76.6923 26.6 22.9 - SR 1.55 12/9/2014 Wendt et al., 2018 41.2398 -76.6917 6.2 SR 1.55 7/9/2015 Wendt et al., 2018 41.2398 -76.6917 7.5 SR 1.55 7/9/2015 Wendt et al., 2018 41.2398 -76.6917 7.5

237

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane

SR 1.55 8/15/2016 PSU 41.2398 -76.6919 12.7 -

SR 1.55 9/15/2016 PSU 41.2398 -76.6919 76.6 -15.2

SR 1.55 10/26/2016 PSU 41.2398 -76.6919 24.2 -20.0

SR 1.55 11/16/2016 PSU 41.2398 -76.6919 17.6 -21.8

SR 1.55 1/16/2017 PSU 41.2398 -76.6919 10.6 -

SR 1.55 2/24/2017 PSU 41.2398 -76.6919 10.4 -

SR 1.55 5/10/2017 PSU 41.2398 -76.6919 6.5 -

SR 1.55 6/22/2017 PSU 41.2398 -76.6919 16.9 - SR 1.55 9/8/2017 PSU 41.2398 -76.6919 14.0 17.6 -

SR 1.6 11/12/2013 Heilweil et al., 2014, 2015 41.2403 -76.6912 5.9 -23.8 SR 1.6 12/9/2014 Wendt et al., 2018 41.2403 -76.6912 6.6 SR 1.6 12/9/2014 Wendt et al., 2018 41.2403 -76.6912 6.5 SR 1.6 12/9/2014 Wendt et al., 2018 41.2403 -76.6912 5.5 SR 1.6 12/9/2014 Wendt et al., 2018 41.2403 -76.6912 6.2 SR 1.6 7/9/2015 Wendt et al., 2018 41.2403 -76.6912 7.1 SR 1.6 7/9/2015 Wendt et al., 2018 41.2403 -76.6912 6.5 SR 1.6 7/9/2015 Wendt et al., 2018 41.2403 -76.6912 7.8

SR 1.6 6/13/2016 Wendt et al., 2018 41.2403 -76.6912 5.8

SR 1.6 8/15/2016 PSU 41.2403 -76.6912 2.1 - SR 1.6 9/15/2016 PSU 41.2403 -76.6912 25.2 7.7 -19.4 SR 1.8 11/12/2013 Heilweil et al., 2014, 2015 41.2409 -76.6910 4.2 -17.9 SR 1.8 7/9/2015 Wendt et al., 2018 41.2409 -76.6910 9.1 SR 1.8 7/9/2015 Wendt et al., 2018 41.2409 -76.6910 9.1 SR 1.8 6/13/2016 Wendt et al., 2018 41.2409 -76.6910 7.1

SR 1.8 8/15/2016 PSU 41.2409 -76.6910 2.6 -

SR 1.8 9/15/2016 PSU 41.2409 -76.6910 3.6 -42.1

SR 1.8 10/26/2016 PSU 41.2409 -76.6910 4.0 -17.0 SR 1.8 1/16/2017 PSU 41.2409 -76.6910 10.1 6.2 -

238

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane SR 2 5/21/2013 Heilweil et al., 2014, 2015 41.2407 -76.6898 20.0 -

SR 2 6/27/2013 Heilweil et al., 2014, 2015 41.2407 -76.6898 7.2 -

SR 2 11/12/2013 Heilweil et al., 2014, 2015 41.2407 -76.6898 5.7 -16.3 SR 2 7/9/2015 Wendt et al., 2018 41.2407 -76.6898 9.3 SR 2 7/9/2015 Wendt et al., 2018 41.2407 -76.6898 9.3

SR 2 6/13/2016 Wendt et al., 2018 41.2407 -76.6898 7.1 -

SR 2 9/15/2016 PSU 41.2407 -76.6898 6.7 -26.3

SR 2 10/26/2016 PSU 41.2407 -76.6898 5.2 -14.7

SR 2 1/16/2017 PSU 41.2407 -76.6898 13.0 -

SR 2 2/24/2017 PSU 41.2407 -76.6898 9.8 -

SR 2 6/22/2017 PSU 41.2407 -76.6898 4.3 - SR 2 7/10/2017 PSU 41.2407 -76.6898 3.5 8.4 - SR 3 5/21/2013 Heilweil et al., 2014, 2015 41.2431 -76.6830 2.2 -

SR 3 6/27/2013 Heilweil et al., 2014, 2015 41.2431 -76.6830 2.4 - SR 3 7/10/2017 PSU 41.2431 -76.6830 4.9 3.2 - SR 4 5/21/2013 Heilweil et al., 2014, 2015 41.2438 -76.6800 1.1 -

SR 4 2/24/2017 PSU 41.2438 -76.6800 3.9 2.5 - SR 6 5/21/2013 Heilweil et al., 2014, 2015 41.2494 -76.6687 0.5 0.5 - SR 7 5/21/2013 Heilweil et al., 2014, 2015 41.2537 -76.6646 0.4 0.4 - SR 7.5 2/24/2017 PSU 41.2537 -76.6646 0.3 0.3 - SR 8 7/10/2017 PSU 41.2625 -76.6570 0.6 0.6 - SSR3 7/19/2016 Anna Wendt 41.6263 -76.2741 1.2 1.2 SSR4 7/17/2017 PSU 41.6270 -76.2747 10.6 10.6 -28.1 SSRR1 7/17/2017 PSU 41.6265 -76.2345 1.0 1.0 SSRR2 7/17/2017 PSU 41.6189 -76.2386 0.9 0.9 Stevens Creek 7/5/2016 SRBC 41.6967 -75.8907 3.8 -49.6 Stevens Creek 10/18/2016 SRBC 41.6967 -75.8907 1.2 2.5

239

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane STONRU001 10/1/2016 West Virginia Snapshot Day 38.6349 -79.8826 0.6 0.6 STRARU001 6/24/2017 Pine Creek Snapshot Day 41.7774 -77.3964 0.2 0.2 Sugar Creek 7/11/2016 SRBC 41.7898 -76.7758 3.0 -46.0 Sugar Creek 10/20/2016 SRBC 41.7898 -76.7758 0.7 1.9 SUGARU001 10/24/2017 ANF Snapshot Day 1 41.8711 -78.8046 0.9 SUGARU001 4/10/2018 ANF Snapshot Day 2 41.8711 -78.8046 0.2 0.6 SUNKBR001 6/24/2017 Pine Creek Snapshot Day 41.6823 -77.7946 0.4 0.4 SWRORU001 10/1/2016 West Virginia Snapshot Day 38.7700 -79.5568 0.1 0.1 TC 2 12/14/2016 PSU 41.6527 -76.7608 1.8 1.8 TC 3 12/14/2016 PSU 41.6527 -76.7379 1.1 1.1 TC 4 12/14/2016 PSU 41.6728 -76.7066 0.5 0.5 TC 5 12/14/2016 PSU 41.6969 -76.5788 0.9 0.9 TC 6 12/14/2016 PSU 41.6980 -76.5788 2.5 2.5 TC 7 12/14/2016 PSU 41.6982 -76.5809 0.5 0.5 TC 7.1 12/14/2016 PSU 41.6982 -76.5809 0.6 0.6 TC 8 12/14/2016 PSU 41.7022 -76.5542 0.5 0.5 TC 8.1 12/14/2016 PSU 41.7022 -76.5542 0.5 0.5 TC 9 12/14/2016 PSU 41.7016 -76.5553 0.6 0.6 TC 9.1 12/14/2016 PSU 41.7016 -76.5553 0.5 0.5 TC 9.2 12/14/2016 PSU 41.7016 -76.5553 0.2 0.2 TC 9.3 12/14/2016 PSU 41.7004 -76.5367 0.2 0.2 TC1 7/19/2016 Anna Wendt 41.6805 -76.6769 4.9 4.9 The Branch 127 4/10/2018 ANF Snapshot Day 2 41.5422 -79.2440 0.9 0.9 THMIRU002 10/30/2018 ANF Snapshot Day 3 41.7802 -78.6011 3.5 3.5 THMIRU003 10/30/2018 ANF Snapshot Day 3 41.5360 -78.8857 1.5 1.5 THUNRU001 10/24/2017 ANF Snapshot Day 1 41.7998 -78.7677 0.4 0.4 THUNRU002 4/10/2018 ANF Snapshot Day 2 41.7821 -78.7797 1.6 1.6

240

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane Toe001 7/19/2016 Anna Wendt 41.6574 -76.7905 3.0 3.0 Tomjack Creek at Burlington, PA Anna Wendt 41.7800 -76.6064 1.1 1.1 Towanda Creek 7/6/2016 SRBC 41.6925 -76.6291 19.4 -53.3 Towanda Creek 10/20/2016 SRBC 41.6925 -76.6291 15.7 17.5 -47.2 TRHOLL001 10/24/2017 ANF Snapshot Day 1 41.9667 -78.8419 0.3 0.3 Trib 36989 to Little Chartiers Creek 10/26/2015 Wendt et al., 2018 40.1780 -80.1662 2.2 2.2 Trib 39657 to Pigeon Creek 10/26/2015 Wendt et al., 2018 40.1778 -79.9790 3.1 3.1 Trib 39670 to Pigeon Creek 10/26/2015 Wendt et al., 2018 40.1630 -80.0090 0.4 0.4 Trib 5.5 5/21/2013 41.2479 -76.6681 0.2 0.2 Trib Black Moshannon Lake 7/18/2015 Wendt et al., 2018 40.8908 -78.0423 1.4 1.4 Trib Black Moshannon Lake 7/18/2015 Wendt et al., 2018 40.8943 -78.0434 4.3 4.3 Trib to Plum Run 1/14/2016 Wendt et al., 2018 40.2576 -80.2180 2.1 2.1 Trib_to_CouncilRun 8/10/2015 Wendt et al., 2018 41.0906 -77.8189 0.2 Trib_to_CouncilRun 5/5/2016 Wendt et al., 2018 41.0906 -77.8189 0.0 Tributary to Council Run 10/12/2015 Wendt et al., 2018 41.0906 -77.8189 0.3 LHU_Trib_to_CouncilRun 11/9/2015 Wendt et al., 2018 41.0906 -77.8189 0.2 0.2 TROURU003 6/24/2017 Pine Creek Snapshot Day 41.4000 -77.4618 0.5 0.5 TURNRU001 10/24/2017 ANF Snapshot Day 1 41.8022 -78.7472 0.9 0.9 TURNRU002 4/10/2018 ANF Snapshot Day 2 41.7823 -78.7651 0.2 0.2 Tuscarora Creek 7/5/2016 SRBC 41.6532 -76.1353 1.2 Tuscarora Creek 10/19/2016 SRBC 41.6532 -76.1353 1.1 1.1 TWEKRY001 10/1/2016 West Virginia Snapshot Day 39.2288 -79.5637 0.4 0.4 Twenty mile creek 2/18/2017 Volunteer 42.2610 -79.7790 0.2 0.2 Twenty mile creek 2 2/18/2017 Volunteer 42.2640 -79.7780 0.2 0.2 Twin Run 9/16/2015 Wendt et al., 2018 41.1075 -77.6936 0.8 Twin Run 10/12/2015 Wendt et al., 2018 41.1075 -77.6936 0.3 Twin Run 4/11/2016 Wendt et al., 2018 41.1075 -77.6936 0.1

241

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane TwinRun 8/10/2015 Wendt et al., 2018 41.1075 -77.6936 0.3 0.4 TWLIRU001 10/24/2017 ANF Snapshot Day 1 41.4919 -78.8639 2.1 TWLIRU001 4/10/2018 ANF Snapshot Day 2 41.4919 -78.8639 9.9 -40.6 TWLIRU001 10/30/2018 ANF Snapshot Day 3 41.4919 -78.8639 5.4 5.8 TWMIRU01 10/30/2018 ANF Snapshot Day 3 41.7203 -78.9241 4.4 4.4 -42.1 Two Rock Run 3/1/2016 Wendt et al., 2018 41.1306 -77.8044 0.4 0.4 Unnamed Tributary 8/6/2015 Wendt et al., 2018 40.1780 -80.1750 1.0 1.0 Unnamed Tributary 8/6/2015 Wendt et al., 2018 40.1831 -80.1326 1.3 1.3 Unnamed Tributary 8/6/2015 Wendt et al., 2018 40.1996 -80.1314 3.4 3.4 Unnamed Tributary 8/6/2015 Wendt et al., 2018 40.2166 -80.1527 3.4 3.4 Unnamed Tributary 8/6/2015 Wendt et al., 2018 40.2170 -80.1407 1.4 1.4 Unnamed Tributary 1 6/22/2015 Wendt et al., 2018 40.2230 -80.1350 1.6 Unnamed Tributary 1 8/6/2015 Wendt et al., 2018 40.2230 -80.1350 3.7 2.6 Unnamed Tributary 2 8/6/2015 Wendt et al., 2018 40.2290 -80.1260 2.6 Unnamed Tributary 2 6/22/2015 Wendt et al., 2018 40.2290 -80.1260 1.0 1.8 UNT to Rose Valley Lake 7/29/2015 Wendt et al., 2018 41.3836 -76.9786 6.3 6.3 UPBORU001 6/24/2017 Pine Creek Snapshot Day 41.3185 -77.4143 0.2 0.2 Upper East Fork Sinnemahoning 9/26/2015 Wendt et al., 2018 41.6284 -77.8604 0.7 0.7 Upper Hunts Run 9/26/2015 Wendt et al., 2018 41.5029 -78.1252 0.9 0.9 UPSHRU001 10/24/2017 ANF Snapshot Day 1 41.6525 -79.1038 1.6 UPSHRU001 4/10/2018 ANF Snapshot Day 2 41.6525 -79.1038 0.5 1.0 UTBECR002 10/24/2017 ANF Snapshot Day 1 41.5161 -78.8197 0.8 UTBECR002 4/10/2018 ANF Snapshot Day 2 41.5161 -78.8197 0.2 0.5 UTEBTC01 10/30/2018 ANF Snapshot Day 3 41.6404 -78.8539 1.4 1.4 UTLIRI001 10/1/2016 West Virginia Snapshot Day 38.5069 -79.7130 0.3 0.3 UTSPRU001 10/1/2016 West Virginia Snapshot Day 38.7882 -79.5557 0.1 0.1 UTTRRU001 6/24/2017 Pine Creek Snapshot Day 41.3993 -77.4769 1.8 1.8

242

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane Wendt et al., 2018 - Test W of Van Buren point 7/3/2016 15:30 Dataset 42.4455 -79.4197 1.3 1.3 WAGNRU001 4/10/2018 ANF Snapshot Day 2 41.5322 -78.9376 0.3 0.3 Walker 1/14/2016 Wendt et al., 2018 40.5702 -80.190 0.5 Walker 1/14/2016 Wendt et al., 2018 40.5702 -80.190 0.4 0.4 Wallis Run 7/29/2015 Wendt et al., 2018 41.3794 -76.9226 0.7 0.7 WARNRU001 10/1/2016 West Virginia Snapshot Day 38.7175 -79.6292 0.3 0.3 WBPICR001 6/24/2017 Pine Creek Snapshot Day 41.7328 -77.6503 1.0 1.0 WBPICR002 6/24/2017 Pine Creek Snapshot Day 41.6821 -77.7745 2.3 2.3 WBTC 1 7/18/2016 Anna Wendt 41.8071 -76.6323 4.5 4.5 West Branch Freeman Run 9/26/2015 Wendt et al., 2018 41.6342 -78.1031 0.5 0.5 West Branch of Cowley Run 9/26/2015 Wendt et al., 2018 41.5986 -78.1857 0.5 0.5 West Branch Towanda Creek at West Franklin, PA Anna Wendt 41.6971 -76.6235 1.5 1.5 West Branch-Big Run 3/1/2016 Wendt et al., 2018 41.1479 -77.7809 0.3 West Branch-Big Run 5/5/2016 Wendt et al., 2018 41.1479 -77.7809 0.0 West Branch of Big Run 9/16/2015 Wendt et al., 2018 41.1479 -77.7809 0.3 0.2 WestRu01 10/30/2018 ANF Snapshot Day 3 41.6587 -78.8152 12.5 12.5 -61.4 WestRu02 10/30/2018 ANF Snapshot Day 3 41.6474 -78.8588 5.2 5.2 -44.9 WFGRRI001 10/1/2016 West Virginia Snapshot Day 38.6010 -79.8216 1.4 1.4 WHITRU002 10/1/2016 West Virginia Snapshot Day 38.6634 -79.8642 0.3 0.3 Wildboy Run 9/26/2015 Wendt et al., 2018 41.6101 -77.8909 1.0 1.0 WINDRU001 4/10/2018 ANF Snapshot Day 2 41.7515 -78.7223 0.3 0.3 WindRu01 10/30/2018 ANF Snapshot Day 3 41.7618 -78.7359 0.3 0.3 Wolf Run 10/12/2015 Wendt et al., 2018 41.1112 -77.8965 0.4 Wolf Run 4/11/2016 Wendt et al., 2018 41.1112 -77.8965 0.1 0.3 LHU_WolfRun_Panther 8/10/2015 Wendt et al., 2018 41.0899 -77.8684 0.2 Wolf Run – Panther Rd. 10/12/2015 Wendt et al., 2018 41.0899 -77.8684 0.2

243

CH4 Mean δ13C- Site Sample Date Sampler Lat Long (µg/L) CH4 CH4 Ethane Wolf Run – Panther Rd. 4/11/2016 Wendt et al., 2018 41.0899 -77.8684 0.1 0.2 Wolf Run - State Line 8/10/2015 Wendt et al., 2018 41.1112 -77.8965 1.2 1.2 WOLFRU003 10/24/2017 ANF Snapshot Day 1 41.5728 -78.7429 2.5 WOLFRU003 4/10/2018 ANF Snapshot Day 2 41.5728 -78.7429 0.9 WOLFRU003 10/30/2018 ANF Snapshot Day 3 41.5728 -78.7429 0.2 1.2 WOLFRU004 10/30/2018 ANF Snapshot Day 3 41.5987 -78.8814 5.7 5.7 -51.4 WOLFRU05 10/30/2018 ANF Snapshot Day 3 41.7502 -79.0613 0.2 0.2 WR 1 7/10/2018 PSU 41.6474 -78.8591 2.6 2.6 WRUNT001 4/10/2018 ANF Snapshot Day 2 41.9260 -78.8406 0.1 0.1 YOAKRU002 10/1/2016 West Virginia Snapshot Day 38.6932 -79.8491 0.3 0.3 7/19/2016 Anna Wendt 41.7624 -76.5500 0.3 0.3

244

Table E18. Sites with site-aggregated mean [CH4] > 4 µg/L and observations from follow up investigations

Site-aggregated mean methane Field and additional Site (µg/L) for site Category1 sampling comments Nothing out of ordinary BEARCR006 8.4 O+G observed during follow up Nothing out of ordinary BEARCR007 5.8 O+G observed during follow up Nothing out of ordinary BEARCR008 6.8 O+G observed during follow up Unrecorded abandoned BIMICR001 4.6 O+G wells discovered in followup Black Moshannon State Park - Site 1 6.7 B Known wetland Known wetland and (Wendt Black Moshannon State Park (near outflow to Dam) 25.6 B et al., 2018) Detectable ethane and isotopes suggest BloodR001 23.3 O+G thermogenic CH #2 (20 feet upstream) 14.0 C AMD Discharge observed CH #3 30.3 C AMD Discharge observed Chartiers Main stem 13.6 C AMD Discharge observed Chartiers Run #1 4.3 O+G+C AMD discharge observed C Run001 23.9 C upstream AMD discharge observed C Run002 19.8 C upstream Lake nearby and shale gas Cross Creek 4.3 O+G wells, not revisited in person DLCK DS1 10.1 UNS DLCK US 3 10.4 O+G Downstream of a beaver EAHICR002b 6.27 B dam East Van Buren point 11.6

245

Site-aggregated mean methane Field and additional Site (µg/L) for site Category1 sampling comments Proximity to other sites containing elevated thermogenic methane and EBTC 2 22.5 O+G thermogenic isotopes Documented ethane, thermogenic isotopes, and a EBTC 4 36.3 O+G leaking well upstream Proximity to other sites containing thermogenic methane and thermogenic EBTC04 9.18 O+G isotopes EBTC3 11.5 O+G Thermogenic isotopes ELKHRU01 10.05 B+O+G proximity to wetland, discussion SRBC, seasonal thermogenic or biogenic Elk Run 9.8 O+G+B isotopes Thermogenic Isotopes, close to gas well but also close to Erie #1 (Walnut Creek) 7.3 O+O+G wetlands and near a landfill Erie #2 (Trib. 1 Walnut Creek) 20.0 O+G Gregs Run (repeat sample) 6.4 O+G Woda et al., 2018 Horton Run 4.2 O+G Observable wetlands and oil and gas wells along stream, ambiguous isotopes (Table HUNTCR005 4.0 B+O+G E17) Observable wetlands and oil and gas wells along stream, ambiguous isotopes (Table HUNTCR006 5.5 B+O+G E17) Observable wetlands and oil HUNTCR009 5.8 B+O+G and gas wells along stream,

246

Site-aggregated mean methane Field and additional Site (µg/L) for site Category1 sampling comments ambiguous isotopes (Table E17) Isotopes are close to biogenic, but many abandoned wells observed in online maps, not visited in KINZHW001 12.9 O+G person Laurel Run Downstream from many 7.3 O+G suspected GLD springs Downstream from leaking pipe (within 5 feet) appearing to drain a well pad. Methane observed coming out of pipe with Gas Rover. There is known coal Laurel Run 3 10.7 O+G mining within 1.5 km Little Chartiers Creek 5.4 B+O+G+C Little Muncy Creek downstream (near Frenchtown, PA) 5.0 O+G Meshoppen 0 15.3 B Meshoppen Bog stream 46.8 B+O+G Meshoppen DS 1 13.5 B Meshoppen DS 1.5 5.1 UNS Meshoppen DS 2 4.9 O+G Meshoppen US 1 12.0 B Meshoppen US 1.5 10.8 B+O+G Meshoppen US 2 21.3 B+O+G Meshoppen US 3 15.6 B Top of abandoned mine Miller Run Trib 1 22.0 C drainage waterfall

247

Site-aggregated mean methane Field and additional Site (µg/L) for site Category1 sampling comments Abandoned well located about 100 feet from site and vents observed in a nearby Moss 2 8.5 O+G parking lot Area of shallow gas wells as Muddy Creek 6.1 B+O+G+C reported by Fred Zelt Nearby Paradise road in NBSR 1 (North Branch Sugar Run) 4.2 UNS Bradford, PA Sample was collected in very PANTRU001 10.7 B marshy area Parks Creek near Rome, PA 7.0 O+G PINECR005 4.1 UNS SAH-13-13 9 Partners 5.3 UNS SC2 4.9 B SLATRU003 5.1 B+O+G South Fork Cross Creek 4.1 O+G Very close to wetland SR 1.1 6.4 O+G See Woda et al., 2018 SR 1.2 8.8 O+G See Woda et al., 2018 SR 1.4 13.0 O+G See Woda et al., 2018 SR 1.45 18.0 O+G See Woda et al., 2018 SR 1.5 22.9 O+G See Woda et al., 2018 SR 1.55 17.6 O+G See Woda et al., 2018 SR 1.6 7.7 O+G See Woda et al., 2018 SR 1.8 6.2 O+G See Woda et al., 2018 SR 2 8.4 O+G See Woda et al., 2018 SSR4 10.6 O+G TC1 4.9 O+G Discussion SRBC concluded Towanda Creek 17.5 O+G+B that area is very boggy.

248

Site-aggregated mean methane Field and additional Site (µg/L) for site Category1 sampling comments Hydrogen and carbon isotopes are ambiguous. Gas wells nearby. Trib Black Moshannon Lake 4.3 B TWLIRU001 5.8 O+G <1km downstream of TWMIRU01 4.4 O+G Ludlow seepage UNT to Rose Valley Lake 6.3 O+G Very close to wetland WBTC 1 4.5 B+O+G WestRu01 12.5 B Biogenic isotopes WestRu02 5.19 B+O+G WOLFRU004 5.65 B+O+G 1Category (or bin) determined as described in Ch. 3: O+G= oil and gas, B = biogenic, UNS= Unsure, C= Coal