Well water contamination in a rural community in ...dbain/publications/Alawattegama2015JESHA.pdf · Reports of ground water contamination in a southwestern Pennsylvania community

Well water contamination in a rural communityin southwestern Pennsylvania near unconventionalshale gas extraction

SHYAMAK ALAWATTEGAMA1 TETIANA KONDRATYUK2 RENEE KRYNOCK1MATTHEW BRICKER1 JENNIFER K RUTTER1 DANIEL J BAIN3 and JOHN F STOLZ12

1Center for Environmental Research and Education Duquesne University Pittsburgh Pennsylvania USA2Department of Biological Sciences Duquesne University Pittsburgh Pennsylvania USA3Department of Geology and Planetary Science University of Pittsburgh Pittsburgh Pennsylvania USA

Reports of ground water contamination in a southwestern Pennsylvania community coincided with unconventional shale gasextraction activities that started late 2009 Residents participated in a survey and well water samples were collected and analyzedAvailable pre-drill and post-drill water test results and legacy operations (eg gas and oil wells coal mining) were reviewed Fifty-six of the 143 respondents indicated changes in water quality or quantity while 63 respondents reported no issues Color change(brown black or orange) was the most common (27 households) Well type when known was rotary or cable tool and depthsranged from 19 to 274 m Chloride sulfate nitrate sodium calcium magnesium iron manganese and strontium were commonlyfound with 25 households exceeding the secondary maximum contaminate level (SMCL) for manganese Methane was detected in14 of the 18 houses tested The 26 wells tested for total coliforms (2 positives) and E coli (1 positive) indicated that septiccontamination was not a factor Repeated sampling of two wells in close proximity (204 m) but drawing from different depths (32 mand 54 m) revealed temporal variability Since 2009 65 horizontal wells were drilled within a 4 km (25 mile) radius of thecommunity each well was stimulated on average with 35 million gal of fluids and 32 million lbs of proppant PA DEP citedviolations included an improperly plugged well and at least one failed well casing This study underscores the need for thoroughanalyses of data documentation of legacy activity pre-drill testing and long term monitoring

Keywords Marcellus shale gas wells water quality hydraulic fracturing

Introduction

Groundwater serves as a major source of drinking waterfor many Pennsylvania residents with an estimated 3 mil-lion Pennsylvanians relying on private wells for their waterneeds[1] Pennsylvania has over 1 million water wells andaround 20000 new water wells are drilled each year[1]

Unlike public water supplies Pennsylvania does not regu-late residential groundwater use and there have been veryfew studies on general water well quality[1ndash4]

The rise in unconventional shale gas extraction usinghydraulic fracturing and the consistent claims of ground-water contamination associated with this practice hasraised concerns about groundwater quality[5ndash10]

Freshwater aquifers may be contaminated by brines orhydrocarbons from underlying formations chemicals usedin the drilling or fracturing processes waste water and thenatural gas itself through a number of pathways[11ndash18]

These include but are not limited to surface contaminationfrom leaky impoundments and spills poor well construc-tion resulting in failed cement and improper casing a pres-surized annulus as well as pre-existing faults and legacyissues from previous mining and drilling opera-tions[111219] Well construction issues involving casing andcementing failures of the gas wells account for most con-tamination incidents[19] They result in the creation of ver-tical migration pathways for fluids to migrate from thesurface downward or from deep formations under extremepressure upwards In addition hydraulic fracturing canincrease the permeability of the targeted deep formationsby creating new fractures or propagating existing fracturesthereby creating flow pathways for the upward migrationof gases and fluids[11141718]

Establishing a definitive link between hydraulic fractur-ing itself and groundwater contamination has been

Address correspondence to John F Stolz Center for Environ-mental Research and Education Duquesne University Pitts-burgh PA 15282 USA E-mail stolzduqeduColor versions of one or more of the figures in the article can befound online at wwwtandfonlinecomlesa

Journal of Environmental Science and Health Part A (2015) 50 516ndash528Copyright copy Taylor amp Francis Group LLCISSN 1093-4529 (Print) 1532-4117 (Online)DOI 101080109345292015992684

Dow

nloa

ded

by [

Uni

vers

ity O

f Pi

ttsbu

rgh]

at 0

927

18

Mar

ch 2

015

challenging The Energy Policy Act of 2005 Section 322Hydraulic Fracturing amends the Safe Drinking WaterAct to exclude ldquo(i) the underground injection of naturalgas for purposes of storage and (ii) the underground injec-tion of fluids or propping agents (other than diesel fuels)pursuant to hydraulic fracturing operations related to oilgas or geothermal production activitiesrdquo Additionallythe large distances (hundreds of meters) separating thedeep gas rich formations and the shallower groundwatersources the multitude of surface and subsurface contami-nant sources and the lack of comprehensive scientific base-line study data are all factors that contribute to thechallenge[2-420] The main objective of this study was todetermine if recent the changes in well water (quality andquantity) of a small rural community in southwesternPennsylvania could be related to the current unconven-tional shale gas extraction activities of drilling andorfracturingThe community chosen for this study comprises a devel-

opment that began as hunting camps with most eventuallyconverted to year round residences It covers a total areaof around 25 km2 (1 km pound 25 km or 062 pound 15 miles)and has about 190 households All have been dependenton wells for their source of drinking water as there is noconnection to municipal supply Although there had beenprevious oil and gas drilling and coal mining activity in thearea[2122] considerable new activity in unconventionalshale gas extraction (USGE) now surrounds the commu-nity The first reported unconventional extraction of shalegas according to the Pennsylvania Department of Environ-mental Protection (PA DEP) began in this area in early2007 and 2008 with the drilling of the Steven Lensey 1 andReedy D2 vertical wells respectively Drilling activityincreased with 20 wells spudded between March 2010 andNovember 2011Coinciding with the increased USGE activity several

residents of the community reported groundwater con-tamination with changes in water quality and quantity[23]

As drilling continued more residents filed complaints andalternative water was provided[2425] After a PA DEPinvestigation found no connection between the change inwater quality and the drilling the alternative water wasremoved and a volunteer water bank was estab-lished[2425] The current study was conducted between thefall of 2011 through the spring of 2014 The goal was todetermine how many wells in the community had beenaffected as well as those that had not and to determinetypes and possible sources of contamination An initialdoor-to-door survey of the residents was carried out inthe fall of 2011Residents were asked six questions regarding water sup-

ply well location well construction and water issues Sub-sequently well water samples were collected and analyzedand additional surveys were completed Water analysisincluded on site measurements (temperature dissolvedoxygen pH and conductivity) and lab analyses for

selected anions (chloride bromide fluoride sulfate phos-phate nitrate nitrite) using ion chromatography andselected metals using ICP-MS (EPA method 2008) A sub-set of wells was tested for light hydrocarbons and the pres-ence of coliforms and E coli Base maps of the study areawere created identifying the location of current horizontalwells legacy operations (eg gas and oil wells coal min-ing) and topography Where available pre-drill and post-drill water analyses done either by the drilling companyPennsylvania Department of Environmental Protection(PA DEP) or independent certified water testing lab werereviewed in addition to a comprehensive PA DEP filereview of permits and well completion reports

Methods and materials

Community survey

The study was conducted with full Institutional ReviewBoard (IRB) approval with the principle investigator(JFS) completing basic IRB training The survey consistedof six questions relating to well location water quality andquantity and testing

1 Do you have well water and where is your well located2 What kind of well is it (eg artesian rotary cable tool)3 Do you know how deep the well is and have you noticeda change in your well depth

4 Have you noticed any change in water quality (tastesmell color) and if so when

5 Have you noticed any change in the water flow orquantity

6 Have you had the water tested and would you be willingto share those results

Sample collection and analysis

Water samples were collected in sterile 1L French squareglass bottles and 100 ml glass bottles containing 10Mnitric acid (trace metal grade Fisher Scientific PittsburghPA USA) Samples were collected upstream of any watertreatment and filtration devices after purging the sourceusually an outside spigot for 10ndash20 min Purge times werereduced to between 2ndash5 min when running the well drywas an issue Collected samples were stored in the darkand on ice and transported back to the lab and stored at4C until analyses On site field measurements of tempera-ture dissolved oxygen pH and conductivity were madewith a YSI-Pro Plus multimeter (YSI Incorporated Yel-low Springs OH USA) Lab concentrations of selectedanions (chloride bromide fluoride sulfate phosphatenitrate nitrite) were measured using a ThermoFisher Dio-nex ICS-1100 Ion Chromatography System equipped witha conductivity cell and UVVIS detector (Dionex Sunny-vale CA USA) using EPA Method 3000[26] with

Well water contamination near unconventional shale gas extraction 517

Dow

nloa

ded

by [

Uni

vers

ity O

f Pi

ttsbu

rgh]

at 0

927

18

Mar

ch 2

015

modifications as reported in Kondratyuk et al[27] Five-point calibrations in triplicate were done with standardsfor each anion Analysis for selected cations was done witha Perkin-Elmer NexION 300X ICP-MS (Waltham MAUSA) at the University of Pittsburgh using the EPAMethod 2008[28] with modifications as reported in Kon-dratyuk et al[27]

Light hydrocarbon analyses

For light hydrocarbon analyses (methane ethane ethenepropane propylene butane) samples were collected in100-ml amber septum sealed vials without headspace Thebottles were kept in the dark on ice and hand delivered toVaporTech Analytical Laboratory and Sampling Services(Valencia PA USA) where the analyses were completed

Total coliform and E coli testing

Total coliforms were determined using EPA Method 9222B and Method 9222 G was used to detect the presence ofE coli[29] First 100-mL samples were collected in triplicateusing six 50-mL sterile polypropylene tubes The water wascollected from either an outside spigot or other inside faucetprior to any water treatment system (ie water softenersUV treatment or filtration) The samples were kept on icepacks in the dark after collection and processed within 48 h

Data mapping and file review

Base maps of the study area were created using the Arc-GIS software package (ESRI Redlands CA USA) Pre-existing base maps topography digital elevation data(LiDAR) coal mining operations and abandoned mineland were retrieved from the Pennsylvania Spatial DataAccess (PASDA) and Pennsylvania Geospatial DataClearinghouse[30] Gas field locations were obtained fromthe Pennsylvania Department of Conservation and Natu-ral Resources (PA DCNR)[31] Abandoned oil and gaswell data where available came from the PA DEP[32]Acomprehensive PA DEP file review was conducted in Janu-ary of 2013 that included permits and well completionreports for the study area Violations were accessedthrough the PA DEP website[32] In addition pre-drill andpost-drill water analyses done either by the drilling com-pany andor PA DEP and provided by study participantswere reviewed and compiled

Results

Community survey

In the initial survey a total of 143 households participatedSurvey results showed that majority of survey respondents

(61) did not know the type of well construction Whenwell type was known it was either cable tool (27 ofrespondents) or rotary (12 of respondents) The averageand median well depths were 637 m (209ft) and 543 m(178 ft) respectively with the shallowest well at a depth of198m (65 ft) and the deepest well at 2743 m (900 ft) Fifty-six (56) households (39) reported changes in water qualityor quantity since 2010 63 households (44) reported noissues and the remaining 24 households (17) were unsureOut of the 56 households reporting issues 50 indicated

changes in quality based on taste andor smell 23 house-holds had quantity issues and 18 reported both quality andquantity issues (Fig 1) Color was the most commonchange with 27 households reporting brown black ororange water Twenty-five households noticed odors and 6households a change in taste (Fig 1) Change in the waterlevel of the well was not readily discernable as 78 of therespondents (55) indicated they did not know Six house-holds stated loss of water during normal use Ninety-fivehouseholds (66) indicated that they had their watertested at some time Among these only 42 had their watertested by the drilling company as part of predrill testingAlthough most households had received their test results9 households said they had not

Water chemistry

Initially 57 water samples from 33 wells were collectedand analyzed over an 18-month period for this studyAnion analysis of these samples revealed the presence ofall analytes (eg bromide chloride fluoride nitrate phos-phate sulfate) except nitrite (Table 1) ICP-MS analysislooked at 31 analytes consisting of a combination of majorions minor ions (trace metals) inorganic chemicals andradionuclides (eg uranium) (Table 2) Only cadmiumand uranium were not detected Although their concentra-tions varied in amount from well to well the most com-mon contaminates were sodium calcium magnesium

Fig 1 Results showing the number of respondents for each ques-tion out of the 143 that participated in the survey

518 Alawattegama et al

Dow

nloa

ded

by [

Uni

vers

ity O

f Pi

ttsbu

rgh]

at 0

927

18

Mar

ch 2

015

iron manganese and strontium (Table 2) Twenty-fivehouseholds had levels of manganese above the maximumcontaminate level (MCL) (Table 2) Plotting of the ClBrmass ratios for all water samples having both anions showa broad distribution of values (Fig 2) However most liebelow the mixing line for seawater (Fig 2)Regular monitoring of the water wells from two house-

holds (Household 1 and Household 2) that were close to

each other (204 m) had similar well construction (cabletool) but well depths that differed (32 m and 54 m) inorder to determine changes in water quality over time andwhether they were affected similarly (Figs 3 and 4) Fieldanalysis showed that the water temperature and pH ofthese two wells were similar and remained steady withaverage temperatures of 11C and 14C respectively andpH of 7 in both Dissolved oxygen (DO) readings showed

Table 1 Summary of anion analysis of water samples

Anion Presence (Households ) Highest Conc (ppm) Lowest Conc (ppm) Method Detection Limits (ppm)

Chloride 100 22269 08 00033Bromide 41 139 005 00084Fluoride 71 528 003 00064Sulfate 98 13482 534 00068Phosphate 30 1076 015 00169Nitrate 88 2681 005 00073Nitrite 0 mdash mdash 00054

Table 2 Summary of cation analysis of water samples collected in this study

Cation(S)MCL(mgL)

Samples Exceeding(S) MCL

Min(mgL)

Med(mgL)

Max(mgL)

MDL(ppb)

Aluminum 005ndash02 1 0007 001 013 2571Antimony 0006 0 lt0001 lt0001 0002 0024Arsenic 001 0 lt0001 0002 0009 0239Barium 2 0 002 008 042 0521Boron 0 lt0001 002 0149 2533Cadmium 0005 0 lt0001 lt0001 lt0001 0021Calcium mdash 21 361 12624 2464Chromium 01 0 lt0001 0003 0038 0097Cobalt lt0001 0001 0008 0133Copper 13 0 lt0001 0012 0339 2272Iron 03 2 004 0135 0415 1509Lead 0015 0 lt0001 0005 0014 0028Lithium mdash 0001 0007 0021 0088Magnesium mdash 03 6562 2098 3504Manganese 005 25 0002 0067 2627 0897Molybdenum mdash lt0001 lt0001 0003 0096Nickel mdash 0001 0003 0019 0140Phosphorus mdash 0001 0027 0318 2098Potassium mdash 0041 081 267 2051Rubidium mdash lt0001 0001 0003 0002Selenium 005 0 lt0001 0002 0013 0566Silicon mdash 04 884 147 295Silver 01 0 lt0001 0002 0003 7996Sodium mdash 5279 104 21703 0527Strontium mdash 003 016 062 0100Tin mdash lt0001 lt0001 0002 0243Titanium mdash lt0001 0001 001 0171Tungsten mdash lt00001 0001 0002 0004Uranium 003 0 lt00001 lt00001 lt00001 005Vanadium mdash lt0001 lt0001 0003 2182Zinc (Zn) 5 0 lt0001 0033 0392 1202

(S)MCL ndash (Secondary) Maximum Contaminant Level (US EPA 2009)No MCL has been set MDL ndashmethod detection limit

Well water contamination near unconventional shale gas extraction 519

Dow

nloa

ded

by [

Uni

vers

ity O

f Pi

ttsbu

rgh]

at 0

927

18

Mar

ch 2

015

seasonal fluctuations increasing in the colder months withHousehold 1 exhibiting more variation in DO The specificconductance readings of the well from Household 2 mir-rored the laboratory chloride readings indicating thatchloride was the major source of dissolved solidsIn contrast Household 1 specific conductance readings

were inversely related with higher conductance readings

corresponding to lower chloride levels Here the increasein conductivity appears to be due to an increase in sulfateReview of laboratory results for Household 1 and House-hold 2 showed that except for phosphate all readings werehigher for Household 2 compared to Household 1 ICP-MS analysis for these two households showed the presenceof iron and manganese The varying results of both fieldand lab analyses indicates that while the two wells sharesimilar physical properties (ie location construction)they draw water from two different sourcesHousehold 1 and 2 (the same as above) had pre-drill

water testing data performed by the industry as well as PADEP (Tables 3 and 4) Review of this data showedincreased levels of all analytes except for barium in House-hold 1 between pre-drill and post-drill Analyses of indus-try pre-drill and three PA DEP post-drill testing data forHousehold 1 appear to indicate the increases in iron man-ganese and strontium coincided with drilling activitiesHowever the levels decrease in subsequent sampling(Table 4) Determination letters from the PA DEPreviewed in this study although negative indicate contam-ination (eg chloride iron and manganese) at levelsabove secondary MCL (Table 4) We were also able toobtain the results of the water analyses carried out by the

Fig 2 The mass ratios of Cl-Br to Cl for all well water sampleshaving both anions present collected during the study The haliteand seawater mixing curves and septic leachate field are fromKatz et al[48] The precipitation mixing curve is from Daviset al[44]

Fig 3Monthly field and lab data for household 1

Fig 4Monthly field and lab data for household 2

520 Alawattegama et al

Dow

nloa

ded

by [

Uni

vers

ity O

f Pi

ttsbu

rgh]

at 0

927

18

Mar

ch 2

015

drilling company and PA DEP for a total of five house-holds (including Households 1 and 2) Four of thesereports were pre-drill test results while the fifth was con-ducted after the drilling had commenced (Table 3) Threeof the 5 households also had their water tested by the PADEP (Households 1 2 and 5) following complaints filedby the homeowners subsequent to noticing changes to theirwell water quality (Table 4) The data show a change inwater quality from pre-drill to post-drilling

Light hydrocarbons

Eighteen wells were tested for the presence of light hydro-carbons (methane ethane ethylene propane propyleneand butane) between January 2013 and April 2014(Table 5) Those that were tested multiple times during thestudy showed that both the concentrations of methane as

well as the presence and concentration of other lighthydrocarbons varied (data not shown) Ethane wasdetected in 5 of the wells and one also had traces of pro-pane and propylene (Table 5)

Total coliforms and E coli

A total of 26 wells were tested for total coliforms and Ecoli as part of this study during 2013 Of these only onewell was positive for total coliforms and one tested posi-tive for both coliforms and E coli These results indicateno source of pathogens or that flow paths large enough toallow bacteria to be transported to the wells are absent

Base mapping

Base mapping of the study findings and data gatheredfrom public databases were used to locate the presence of

Table 3 Industry water testing results

Analyte

Household 1a

Sample date6292010

Household 2a

Sample date6302010

Household 3a

Sample date532010

Household 4a

Sample date6282010

Household 5b

Sample date6292011

Barium (mgL) 0155 0389 0227 0113 013Chloride (mgL) 24 1737 115 212 255Iron (mgL) 0149 0173 17 031 2952Manganese (mgL) 0049 0143 0353 0047 0622Spc conductance (mScm) 339 861 212 825 456Total dissolved solids (mgL) 183 581 179 420 291E coli Absent Absent Absent Absent AbsentTotal Coliform Absent Absent Absent Present PresentMethane (mgL) ND ND ND 853 266Strontium Not tested Not tested Not tested Not tested 0089aPre-drill testingbPost-drilling test

Table 4DEP water testing results

Household 1 Household 2 Household 5

AnalyteSample Date2102011

Sample Date2142011

Sample Date442011

Sample Date11142012

Sample Date8112011

Barium (mgL) 0142 0495 0143 061 0132Chloride (mgL) 39 22 25 2524 112Iron (mgL) 1175 366 0113 2059 1968Manganese (mgL) 0194 5617 0081 0287 069Strontium (mgL) 0163 0224 0189 0591 0123Spc conductance (mScm) 381 366 299 1072 300Total dissolved solids (mgL) 244 172 164 872 226E coli Absent Not tested Not tested Not tested Not testedTotal Coliform Absent Not tested Not tested Not tested Not testedToluene (mgL) ND 0000372 Not tested Not tested Not testedOther VOCrsquos ND ND Not tested Not tested Not testedMethane (mgL) Not tested Not tested Not tested 00153 00241Ethene (mgL) Not tested Not tested Not tested Not tested 00198Ethane (mgL) Not tested Not tested Not tested 00124 00198Propane (mgL) Not tested Not tested Not tested 00142 00142

Well water contamination near unconventional shale gas extraction 521

Dow

nloa

ded

by [

Uni

vers

ity O

f Pi

ttsbu

rgh]

at 0

927

18

Mar

ch 2

015

historic and current oil and gas and mining activity sur-rounding the community The Little Creek Oil Fieldwhich has had previous drilling activity lies at about1300rsquo below the community (Fig 5) The main petro-leum-bearing section is the Hundred Foot sandstone andthe Snee sandstone lies even deeper at around 1600 ft[21]

Mapping of legacy oil and gas wells abandoned andororphaned wells coal mines and abandoned mines dataconfirms the existence of previous oil and gas and miningactivity surrounding the community (Fig 6) however pin-pointing these has been difficult Lytlersquos review of oil fieldexploration in Butler County indicates 500 abandoned oilwells with 250rsquo spacing drilled into the Hundred Foot andSnee sands within Connequenessing and Forward Town-ships[33] No specific locations however were providedThe USGS Topo map obtained through the US Fish andWildlife Service indicate at least three oil wells within theconfines of the community[34] The same cannot be said forgas wells as unfortunately these same maps use the identi-cal symbol for buildings and gas wells PA DEP file reviewdata indicate at least a dozen gas wells were drilled in thearea between 1961 and 1985 however their coordinates(longitude latitude) appear to have been added later andonly plot to the general area down to minutes rather thanseconds (eg N40o50rsquoW 80o00rsquo)

PA DEP file reviews

PA DEP file reviews of permits and well completionrecords provided the locations lengths and directions ofUSGE wells in Connequenessing and Lancaster Town-ships Analysis of this data shows that by late 2012 thecommunity was surrounded by 15 well pads and a total of

65 laterals (horizontal gas wells) (Fig 7) Table 6 summa-rizes the review of unconventional shale gas drillingrecords obtained from the PA DEP Each well was stimu-lated with varying amounts of fluids and proppant onaverage 35 million gal and 32 million lbs respectivelyA list of violations for oil and gas activity in the town-

ship from September 2010 through September 2012 wasobtained from the PA DEP website (Table 7)[32] Viola-tions ranged from simple administrative issues such as fail-ure to post proper documentation to more directenvironmental impacts such as improper discharges andcompromised well casings and inadequate plugging of awell

Discussion

The results of the survey indicate that a significant num-ber of families (56 out of 143) saw noticeable changes inquality andor quantity to their private well water sup-ply since 2009 (Fig 1) Although the recent USGEactivities began in 2009 the respondents began noticingthe changes at different times after that This delaymight be attributed the location within the communityof each household relative to the well pads the substan-tial difference in depth of the water wells (that varied indepth from 19 to 274 m) as well as the form of contami-nation Most respondents saw changes in secondarywater standards namely color taste or smell It is notsurprising that no one surveyed had equipment to rou-tinely monitor or test their well water quality The studyby the Center for Rural Pennsylvania indicated thatabout half of the participants in their study had not hadproper water testing done thus many did not know ifthey had issues[1] In contrast more than 66 of thehouseholds in this study (95 out of the 143) indicatedthey had had their water tested at some time Howeveronly 42 households had had their water tested by thegas drilling company as part of predrill testing The sur-vey results suggest that more comprehensive (eggreater distances from the well pad) pre-drill testingshould be required as recommended by the CitizensMarcellus Shale Commission[35]

Typical contaminants seen in Pennsylvania domesticwells include bacteria from septic or runoff chloride fromroad salt nitrate and phosphate from agriculture iron andsulfate from mining brines from oil and gas wells andmethane[1ndash3] The brine and produced water associatedwith unconventional gas extraction in the Marcellus Shaleis reported to contain high levels of total dissolved solidshalides (eg chloride bromide) strontium barium andnaturally occurring radionuclides[36ndash41] Water analysesconducted during this study indicated elevated levels ofchloride manganese iron and specific conductance ascompared to good quality groundwater[1] Chloride wasfound in all of the wells sampled however none exceeded

Table 5 Light hydrocarbons (in mgL)

Household Methane Ethane Ethylene Propane Propylene Butane

LV1 4232 058 ND ND ND NDBS1 ND ND ND ND ND NDRA1 037 ND ND ND ND NDPH1 056 ND ND ND ND NDHK1 993 ND 003 ND ND 008CH2 ND ND ND ND ND NDWO1 033 ND ND ND ND NDCO1 ND ND ND ND ND NDCB1 109 002 ND ND ND NDCB2 137 ND 002 ND ND NDSV1 1557 143 ND 005 007 NDSV2 ND ND ND ND ND NDHT1 421 ND ND ND ND NDHT2 042 001 ND ND ND NDHT3 055 ND ND ND ND NDHT4 052 ND ND ND ND NDHT5 183 028 ND 002 005 NDHT6 36 ND 004 ND ND ND

522 Alawattegama et al

Dow

nloa

ded

by [

Uni

vers

ity O

f Pi

ttsbu

rgh]

at 0

927

18

Mar

ch 2

015

Fig 5 Survey results plotted over the Little Creek Oil Field that underlies the community

Fig 6 Current and historic oil and gas and mining activity surrounding the community as could be determined from state sources(eg PASDA PA DEP)

Well water contamination near unconventional shale gas extraction 523

Dow

nloa

ded

by [

Uni

vers

ity O

f Pi

ttsbu

rgh]

at 0

927

18

Mar

ch 2

015

the secondary drinking water limit of 250 mg Liexcl1

(Table 1) The presence of chloride in Pennsylvaniagroundwater is common however concentrations are typ-ically less than 25 mg Liexcl1[142]

The results from the total coliform and E coli testing aswell as the mass ratio analyses for chloride and bromideindicate that septic leachate may be ruled out as a source[43]

Other potential sources of contaminants include precipita-tion road salt (eg halite) and brines associated with coaloil and gas fields Here the use of ClBr mass ratios can beused to assess potential sources of chloride[44ndash48] Reportedvalues for ClBr mass ratios range from 50ndash150 for

precipitation 100ndash200 for shallow groundwater 300ndash600for septic leachate and 1000ndash10000 for halite[4448] Pro-duced water from unconventional shale gas extraction mayhave concentrations of chloride in excess of 100000 mgLiexcl1 but the ClBr mass ratios are indicative of evaporatedseawater[49] Our analyses revealed that the majority of thewell samples fell below the primarymixing line for seawater(Fig 2) This suggests that brine may be the primary sourceof elevated chloride mixing with groundwater[4849]

Similar to chloride iron and manganese have limitsset under the secondary standards for drinking water[50]

These two analytes were elevated in a number of

Table 6 Summary of DEP file reviewa findings showing well name dates of stimulation (eg fracking) volume of fluids amount ofproppant and length of the lateral

Well Name Dates Fracked Volume of Fluid (gal) Proppant (lb) Length Fracked (ft)

Steven Lesney 1 2122007 11839 4154 118Reedy 2 6102008 1066700 1004920 62Shannon 1H 1052010 ndash 1092010 3572208 3766100 3000Shannon 2H 12102010 ndash 12192010 3179278 2461050 2660Voll 1H 172011 ndash 1212011 3159184 2634500 3240Voll 2H 172011 ndash 1222011 4215810 3800400 4200Ragan 11 4202011 37500 25000 216Grosick 1H 1142011 ndash 11122011 4744446 No data 3575Grosick 2H 1102012 ndash 1132012 3924774 No data 2965aData gathered from DEP file reviews of well record and completion reports

Fig 7 Locations lengths and directions of lateral wells surrounding the community Data compiled from DEP file review of loca-tion plats

524 Alawattegama et al

Dow

nloa

ded

by [

Uni

vers

ity O

f Pi

ttsbu

rgh]

at 0

927

18

Mar

ch 2

015

samples with iron found above the 03 mg Liexcl1 standardlimit in 35 of samples and manganese was above the005 mg Liexcl1 standard limit in 44 of samples (Table 2)The ubiquitous presence of manganese in these wellswith 25 households above the maximum contaminatelevel (MCL) is a serious health concern (Table 2)[50]

Manganese is an essential element required for manybiological enzymes however it is also a known neuro-toxin[51] Long-term exposure can lead to manganismand Parkinsonrsquos Disease[52] Recent studies have shownthat manganese concentrations in drinking waterbetween 300-400 ppb can lead to lower IQ in chil-dren[5354] The presence of iron and manganese is acommon occurrence in western Pennsylvania watersespecially around coal mining areas[1] Although the PADEP base map for abandoned coal mines does not showthe presence of such mines underneath the confines ofthe community strip mines are in close proximity(Fig 6) and the watershed is under a TMDL monitoringprogram[55] Thus mine drainage may be another impor-tant source of contamination to the wellsMethane was found in 14 of the 18 wells tested and

while most were in low microgram amounts at least onewell had mg Liexcl1 quantities (Table 5) Methane can beeither biogenic or thermogenic origin Methanogensmicroorganisms belonging to the domain Archaea gener-ate methane from carbon dioxide acetate or simpleorganic compounds (eg formate)[56] Thermogenic

methane on the other hand is formed through the ldquocrack-ingrdquo or catagensis of fossil organic carbon (eg kerogen)and is usually found with other light hydrocarbons[1417]

In addition biogenic methane is depleted in 13C and hasd13C values ranging from iexcl60 to iexcl120 ooo[17] Thus bio-genic sources may be distinguished from thermogenicsources through isotopic (eg d13C-CH4 and d2H-CH4)and geochemical analyses (eg propanemethaneratios)[9141756-58] The presence of associated hydrocar-bons (eg ethane propane propylene butane) may thenbe an indication that the source of methane is thermogenicin at least six wells (Table 5) Definitive determination ofthe thermogenicity however will require isotopic analy-ses but may not indicate the exact source (ie both LittleCreek Field and deeper deposits are thermogenic)[5758]

Mapping of survey results shows the spatial extent ofgroundwater issues experienced by the residents No pat-tern of contamination is apparent as the households indi-cating changes in water quality are not clustered together(Fig 7) Overlay of the survey results with underlying oiland gas reserves shows that the affected homes sit atop theLittle Creek Oil Field (Fig 5) Both the Hundred Footsandstone and the Snee sandstone of the Little Creek weredeveloped in the late 1800rsquos and early 1900rsquos with at leastone well reaching over 4000 ft[2133] The PA DEP filereview also included permits for at least a dozen conven-tional wells in the area that were drilled between 1961 and1985 with several on current USGE sites (eg Graham

Table 7 PA DEP reported USGE violations for the township issued between September 2010 to 2012

Site Name Violation ID Violation Date Violation Description

Edward Gilliland 0 OGWell 594808 982010 Failure to plug a well upon abandonmentVoll Unit 1H OGWell 595298 9142010 Failure to maintain 2rsquo freeboard in an

impoundmentVoll Unit 1H OGWell 595299 9142010 Failure to report defective insufficient or

improperly cemented casing win 24 h orsubmit plan to correct win 30 days

Voll Unit 3H OGWell 599948 11162010 Stream discharge of IW includes drill cuttingsoil brine andor silt

Gilliland Unit 4H OGWell 599859 11202010 Failure to properly store transport process ordispose of a residual waste

Bricker Unit 1H OGWell 619173 8312011 Failure to notify DEP landowner politicalsubdivision or coal owner 24 h prior tocommencement of drilling

Bricker Unit 1H OGWell 619174 8312011 Failure to post permit number operator nameaddress telephone number in a conspicuousmanner at the site during drilling

Grosick Gilliland Carson Pipeline ESX 629497 1172012 Discharge of industrial waste to waters ofCommonwealth without a permit

Grosick Gilliland Carson Pipeline ESX 630037 1232012 Discharge of industrial waste to waters ofCommonwealth without a permit

Bricker Pipeline ESX 641921 6182012 Discharge of industrial waste to waters ofCommonwealth without a permit

Patton Unit 1H OGWell 650294 9282012 Conservation well located less than 330rsquo fromlease or unit line without waiver

Well water contamination near unconventional shale gas extraction 525

Dow

nloa

ded

by [

Uni

vers

ity O

f Pi

ttsbu

rgh]

at 0

927

18

Mar

ch 2

015

Gilliland pads) Thus the region has had significant legacyoil gas and mining operations (Fig 6)More telling is the location of the new unconventional

wells as there are 15 well pads with 65 horizontal wellswithin a -m radius of the community (Fig 7) The lateralsfollow a north-north west or south-south-east drillingdirection to take advantage of the enhanced permeabilityof the J1 joint of the Marcellus shale maximizing yields[59]

The large volumes of fluids and proppant used in the slickwater hydraulic fracturing of the new wells (on average32 million gal and 3 million lbs respectively) (Table 6)could also affect the local hydrology and contribute to sub-surface disturbance resulting in well water contamina-tion[11] While the depth of the Marcellus in this part of thestate is around 5000 ft the violations cited in late 2010(Table 7) especially the failed casingcement job couldprovide conductive pathways for the migration of legacymining as well as deep formations fluids[1119] Althoughthe PA DEP eventually determined that there was no con-nection between the new drilling and the changes in waterquality their own data indicate that at least for Household1 there was a dramatic and contemporaneous increase inconcentrations of barium strontium iron and manganese(Table 4) as well as contamination in other wells (egchloride iron and manganese) at levels above secondaryMCL (Table 4)

Conclusions

We have used a variety of methods to determine whetherthere is a correlation between the changes in well waterquality in water wells with surrounding USGE in this com-munity in Southwest Pennsylvania The survey resultsindicate that there has been an increase in well water issuesin the community since 2010 Water chemistry resultsshow elevated cations and anions including manganeseiron bromide and chloride Different wells had differentcontaminants although the majority had manganese abovethe MCL Light hydrocarbon analyses suggested a ther-mogenic source for the methane in some wells Analysis ofmapping results revealed the community lies over the Lit-tle Creek Oil Field and locations of previous mining andoil and gas activitiesDEP file review indicates several violations that could

result in groundwater contamination The proximity andlocation of USGE well sites to the community provideshorter pathways for the transport of surface and subsur-face contamination The number of lateral wells (65)within 4 km of the community could have contributed tosubsurface disturbance ultimately resulting in well watercontamination Further in-depth study of the local geologyand hydrology in addition to access to all pre-drill testsand well completion records would allow for a more defin-itive assessment This study demonstrates the challengesfaced in making a positive determination (eg relating

contamination to drilling activity) and the need for thor-ough investigation including legacy activity pre-drill test-ing and long term monitoring

Acknowledgments

The authors would like to thank the community for theirwillingness to participate in the study We also thank SMayes C Nolan C Brown and T Umstead for addi-tional sample collection and analyses as well as WMGriffin RS Oremland and A Ingraffea for helpfulcomments

Funding

This work was supported in part by the Heinz Endow-ments and Colcom Foundation

References

[1] Swistock BR Clemens S Sharpe WE Drinking water qualityin rural Pennsylvania and the effect of management practices TheCenter for Rural Pennsylvania Harrisburg PA 2009

[2] Swistock BR Gas well drilling and your private water supplyPenn State College of Agricultural Sciences Cooperative Exten-sion Harrisburg PA 2010

[3] Boyer EW Swistock BR Clark J Madden M Rizzo DEThe impact of Marcellus gas drilling on rural drinking water sup-plies The Center for Rural Pennsylvania Harrisburg PA 2012

[4] Sloto RA Baseline groundwater quality from 20 domestic wells inSullivan County Pennsylvania 2012 US Geological Survey Sci-entific Investigations Report 2013-5085 Reston VA 2013

[5] Groundwater Protection Council (GWPC) Modern Shale GasDevelopment in the United States A Primer prepared for the USDepartment of Energy National Energy Technology Laboratory(NETL) 2009

[6] US Environmental Protection Agency Study of the potentialimpacts of hydraulic fracturing on drinking water resources prog-ress report EPA601R-12011 2012

[7] Gregory K Vidic R Dzombak D Water management chal-lenges associated with the production of shale gas by hydraulicfracturing Elements 2011 7 181ndash186

[8] Vidic RD Brantley SL Vandenbossche JM Yoxtheimer DAbad JD Impact of shale gas development on regional waterquality Science 2013 340 1235009

[9] Vengosh A Warner N Jackson R and Darrah T The effectsof shale gas exploration and hydraulic fracturing on the quality ofwater resources in the United States Procedia Earth and PlanetaryScience 3013 7 863ndash866

[10] Vengosh A Jackson RB Warner N Darrah TH and Kon-dash A A critical review of the risks to water resources fromunconventional shale gas development and hydraulic fracturing inthe United States Environ Sci Tech 2014 48(15) 8334ndash8348

[11] Harrison S Evaluating system for ground-water contaminationhazards due to gas- well drilling on the glaciated Appalachian Pla-teau Ground Water 1983 21 689ndash700

[12] Harrison S Contamination of aquifers by over pressuring theannulus of oil and gas wells Ground Water 1985 23 317ndash324

[13] Osborn SG Vengosh A Warner N R and Jackson RB Meth-ane contamination of drinking water accompanying gas-well

526 Alawattegama et al

Dow

nloa

ded

by [

Uni

vers

ity O

f Pi

ttsbu

rgh]

at 0

927

18

Mar

ch 2

015

drilling and hydraulic fracturing P Nat Acad Sci 2011 1088172ndash8176

[14] Warner NR Jackson RB Darrah TH Osborn SG DownA Zhao K White A and Vengosh A Geochemical evidence forpossible natural migration of Marcellus formation brine to shallowaquifers in Pennsylvania P Nat Acad Sci 2012 109 11961ndash11966

[15] Myers T Potential contamination pathways from hydraulicallyfractured shale to aquifers Ground Water 2012 50 872ndash882

[16] Saiers JE and Barth E Potential contaminant pathways fromhydraulically fractured shale aquifers Ground Water 2012 50826ndash826

[17] Jackson RB Vengosh A Darrah TH Warner NR DownA Poreda RJ Osborn SG Zhao K and Karr JD Increasedstray gas abundance in a subset of drinking water wells near Marcel-lus shale gas extraction P Nat Acad Sci 2013 110 11250ndash11255

[18] Kang M CO2 Methane and Brine Leakage Through SubsurfacePathways Exploring Modeling Measurement and Policy OptionsPhD Dissertation Princeton University 2014 131 p

[19] Ingraffea AR Wells MT Santoro RL and Shonkoff SBCAssessment and risk analysis of casing and cement impairment inoil and gas wells in Pennsylvania 2000ndash2012 P Nat Acad Sci2014 111(30) 10955ndash10960

[20] Chapman MJ Gurley LN and Fitzgerald SA Baseline wellinventory and groundwater-quality data from a potential shale gasresource area in parts of Lee and Chatham Counties North Caro-lina October 2011- August 2012 US Department of Interior andUS Geological Survey Data Series 861 2014

[21] Richardson GB Geology and Mineral Resources of the Butlerand Zelienople Quadrangles Pennsylvania Geological Survey Bul-letin 873 United States Department of Interior United StatesPrinting Office 1936

[22] History of Butler County Pennsylvania 1895 RC Brown Cohttpwwwrootswebancestrycom~pabutler1895 (accessedNov 2014)

[23] Associated Press Connoquenessing WESA Pittsburgh NPR Sta-tion 2012 Available at httpwesafmpostconnoquenessing-water-ok (accessed Nov 2014)

[24] Khan N 2014 A day in the life of a water bank Available athttppublicsourceorginvestigationsday-life-of-water-bankVGy5h4elpL0 (accessed May 2014)

[25] Frazier R 2014 Fracking and groundwater contamination Itrsquoscomplicated Allegheny Front Available at httpwwwalleghenyfrontorgstoryfracking-and-groundwater-contamination-its-complicated (accessed Jul 2014)

[26] US Environmental Protection Agency Method 3000 Determina-tion of inorganic anions by ion chromatography Available at httpwaterepagovscitechmethodscwabioindicatorsupload2007_07_10_methods_method_300_0pdf 2007 (accessed Jun 2014)

[27] Kondratyuk T Eastham JL Rutter JK Bain DJ Basu PStolz JF Application of anion analysis for chemical characteriza-tion of fluids associated with Marcellus Shale gas extraction ApplGeochem 2015 in press

[28] US Environmental Protection Agency Method 2008 Determina-tion of trace elements in waters and wastes by inductively coupledplasma ndashmass spectrometry httpwaterepagovscitechmethodscwabioindicatorsupload2007_07_10_methods_method_200_8pdf 2007 (accessed Jun 2014)

[29] US Environmental ProtectionAgency Protocol forDeveloping Path-ogen TMDLs EPA 841-R-00-002 Office of Water (4503F) UnitedStates Environmental ProtectionAgencyWashingtonDC 2001

[30] Pennsylvania Spatial Data Access (PASDA) and PennsylvaniaGeospatial Data Clearinghouse Available at httpwwwpasdapsuedu (accessed Jun 2014)

[31] Pennsylvania Department of Conservation and Natural Resources(PA DCNR) Available at httpwwwdcnrstatepaus (accessedJun 2014)

[32] Pennsylvania Department of Environmental Protection (PA DEP)Available at httpwwwdepwebstatepausportalserverptcommunitydep_home5968 (accessed Jun 2014)

[33] Lytle WS Oil fields of the greater Pittsburgh region MineralResource Report 70 Pennsylvania Geological Survey HarrisburgPA 1976

[34] US Fish and Wildlife Service National Wetlands Inventory Wet-lands mapper Available at httpwwwfwsgovwetlandsdatamapperHTML (accessed Oct 2014)

[35] Citizen Marcellus Shale Commission Marcellus Shale A CitizensView Pennsylvania Budget and Policy Center Harrisburg PA2011 69 p

[36] Blauch M Myers R Moore T and Houston N MarcellusShale post-frac flowback waters - where is all the salt coming fromand what are the implications Society of Petroleum EngineersInternational 2009 SPE 125740

[37] PalmertonGroupPADEPFracflow-backwater study presence of inor-ganics Available at httpwwwpalmertongroupcompdfPADEP20Frac20Flow_Back20Water20Study_20Presence20of20Ino-rganicspdf ed 2010 (accessed Jun 2014)

[38] Chapman EC Capo RC Stewart BW Kirby CS HammackRW Schroeder KT and EdenbornHMGeochemical and stron-tium isotope characterization of producedwaters fromMarcellus shalenatural gas extraction Envir Sci Tech 2011 46 3545ndash3553

[39] Balaba RS and Smart RB Total arsenic and selenium analysisin Marcellus shale high-salinity water and hydrofracture flowbackwastewater Chemosphere 2012 89 1437ndash1442

[40] Rowan EL Engle MA Kirby CS and Kraemer TFRadium content in oil-and gas-field produced waters in the North-ern Appalachian Basin (USA) summary and discussion of dataUS Geological Survey Scientific Investigations Report 2011-5135Reston VA 2011

[41] Brown VJ Radionuclides in fracking waste water managing atoxic blend Environ Health Perspect 2014 122 A50ndash55

[42] Swistock BR Sharpe WE Clark J A Water tests What dothe numbers mean Penn State Cooperative Extension College ofAgricultural Sciences University Park The Pennsylvania StateUniversity 2003

[43] Zimmerman TM Zimmerman ML and Lindsey BD Rela-tion between selected well-construction characteristics and occur-rence of bacteria in private household-supplly wells Southcentraland Southeastern Pennsylvania US Geological Survey WaterResources Investigations Report 01-4206 2001

[44] Davis SN Whittemore DO Fabryka-Martin J Uses of Chlo-rideBromide ratios in studies of potable water Groundwater1998 36(2) 338ndash350

[45] Cartwright I Weaver TR Fifield LK ClBr ratios and envi-ronmental isotopes as indicators of recharge variability andgroundwater flow An example from the southeast Murray BasinAustralia Chem Geol 2006 231 38ndash56

[46] Leybourne MI Goodfellow WD BrCl ratios and O H C andB isotope constraints on the origin of saline waters from easternCanada Geochem Cosmochem Acta 2007 71 2209ndash2223

[47] Alcala FJ and Custodio E Using the ClBr ratio as a tracer toidentify the origin of salinity in aquifers in Spain and Portugal JHydrol 2008 359 189ndash207

[48] Katz BG Eberts SM and Kauffman LJ Using ClBr ratiosand other indicators to assess potential impacts of groundwaterquality from septic systems A review and examples from principalaquifers in the United States J Hydrol 2011 39 151ndash166

[49] Haluszczak LO Rose AW and Kump LR Geochemicalevaluation of flowback brine from Marcellus gas wells in Pennsyl-vania USA App Geochem 2013 28 55ndash61

[50] US Environmental Protection Agency National Primary Drink-ing Water Regulations EPA 816-F-09-004 US EPA WashingtonDC 2009

Well water contamination near unconventional shale gas extraction 527

Dow

nloa

ded

by [

Uni

vers

ity O

f Pi

ttsbu

rgh]

at 0

927

18

Mar

ch 2

015

[51] Crossgrove J and Zheng W Manganese toxicity upon overexpo-sure NMR Biomed 2014 17 544ndash553

[52] Aschner M Erikson KM Herrero Hernandez E and Tjalk-ens R Manganese and its role in Parkinsonrsquos disease From trans-port to neuropathology Neuromol Med 2009 11(4) 252ndash266

[53] Wasserman GA Liu X Parvez F Ahsan H Levy D Fac-tor-Litvak P Kline J van Geen A Slavkovich V LolaconoNJ Cheng Z Zheng Y and Graziano JH Water manganeseexposure and childrenrsquos intellectual function in Araihazar Ban-gladesh Environ Health Persp 2006 114(1) 124ndash129

[54] Khan KWassermanGA Liu X Ahmed E Parvez F Slavko-vich V Levy D Mey J van Geen A Graziano JH and Fac-tor-Litvak P Manganese exposure from drinking water andchildrenrsquos academic achievement NeuroToxicology 2012 33 91ndash97

[55] PA DEP Little Connoquenessing Creek watershed TMDL ButlerCounty Technical Report PA DEP Harrisburg PA 2009

[56] Oremland RS Biogeochemistry of methanogenic bacteria InBiology of Anaerobic Microorganisms Zehnder AJB Ed JohnWiley and Sons Hoboken NJ 1988 641ndash690

[57] Molofsky LJ Conner JA Farhat SK Wylie AS WagnerT Methane in Pennsylvania water wells unrelated to MarcellusShale fracturing Oil Gas J 2011 109 54ndash67

[58] Molofsky LJ Conner JA Wylie AS Wagner T Farhat SK Evaluation of methane sources in groundwater in NortheasternPennsylvania Groundwater 2013 51 333ndash349

[59] Engelder T Lash GG Uzcategui RS Joint sets that enhancethe production of Middle and Upper Devonian gas shales of theAppalachian Basin AAPG Bull 2009 95 1399ndash1422

528 Alawattegama et al

Dow

nloa

ded

by [

Uni

vers

ity O

f Pi

ttsbu

rgh]

at 0

927

18

Mar

ch 2

015