Microbial production and modification of gases in sedimentary basins A geochemical case study from a

AnnaM.Martini,LynnM.Walter,TimC.W.Ku,

JoyceM.Budai,JenniferC.McIntosh,andMartinSchoell

ABSTRACT

AnexpandeddatasetforgasesproducedfromtheAntrimShale,aDevonianblackshaleintheMichiganbasin,UnitedStates,hasallowedforadetailedexaminationoftherelatedchemicalandiso-topiccompositionalchangesinthesolid-gas-liquidsystemsthatdiscriminatebetweenmicrobialandthermogenicgasorigin.IntheAntrimShale,economicmicrobialgasdepositsarelocatednearthebasinmarginswheretheshalehasarelativelylowthermalmaturityandfreshwaterinfiltratesthepermeablefracturenetwork.Themostcompellingevidenceformicrobialgenerationisthecorrela-tionbetweendeuteriuminmethaneandcoproducedwater.Alongthebasinmargins,thereisalsoasystematicenrichmentin13Cofethaneandpropanewithdecreasingconcentrationsthatsuggestsmicrobialoxidationofthesethermogenicgascomponents.Micro-bialoxidationaccountsnotonlyfortheshiftind13Cvaluesforethane,butalso,inpart,forthegeographictrendingascomposi-tionasethaneandhigherchainhydrocarbonsarepreferentiallyre-moved.Thisoxidationislikelyananaerobicprocessinvolvingasyntrophicrelationshipbetweenmethanogensandsulfate-reducingbacteria.

Theresultsofthisstudyareintegratedintoapredictivemodelformicrobialgasexplorationbasedonkeygeochemicalindicatorsthatarepresentinbothgasandcoproducedwater.OneunequivocalsignatureofmicrobialmethanogenesisistheextremelypositivecarbonisotopevaluesforboththedissolvedinorganiccarboninthewaterandthecoproducedCO2gas.Incontrast,thed13Cvalueofmethaneisoflimiteduseinthesereservoirsasthevaluestypically

Copyright#2003.TheAmericanAssociationofPetroleumGeologists.Allrightsreserved.

ManuscriptreceivedFebruary7,2002;provisionalacceptanceJuly17,2002;revisedmanuscriptreceivedJanuary6,2003;finalacceptanceMarch19,2003.

AAPGBulletin,v.87,no.8(August2003),pp.1355–13751355

AUTHORS

AnnaM.Martini$DepartmentofGeology,AmherstCollege,Box2038,Pratt,Amherst,Massa-chussetts,01002-5000;ammartini@amherst.edu

AnnaM.MartinireceivedherB.A.degreeingeologyfromColgateUniversity,herM.S.degreefromSyracuseUniversity(1992),andherPh.D.fromtheUniversityofMichigan(1998).SheiscurrentlyanassistantprofessorofgeologyatAmherstCollege.Herresearchinterestsincludeunconventionalnaturalgasplays,isotopictracingofmicrobialinterac-tions,andthegeochemistryofsalinefluids.LynnM.Walter$DepartmentofGeologicalSciences,UniversityofMichigan,AnnArbor,Michigan,48109-1603;lmwalter@umich.eduLynnM.WalterreceivedherM.S.degreefromLouisianaStateUniversity(1978)andherPh.D.fromtheUniversityofMiami(1983).ShewasanassistantprofessoratWashingtonUniversityinSt.Louisuntil1988.ShethenjoinedtheUniversityofMichigan,wheresheisnowaprofessorofgeologicalsciencesanddirectoroftheExperimentalandAnalyticalGeochemistryLaboratory.Herresearchinterestsfocusonthehydrogeochemistryofnear-surfaceanddeeperbasinenvironments,withanemphasisoncarbontransformationsandmineralmasstransport.

TimC.W.Ku$DepartmentofGeologicalSciences,UniversityofMichigan,AnnArbor,Michigan,48109-1603;presentaddress:Depart-mentofEarthandEnvironmentalSciences,WesleyanUniversity,265ChurchStreet,Middle-town,Connecticut,059-1603;tcwku@umich.eduTimKureceivedhisB.S.degreeingeologicalsciencesfromtheUniversityofRochesterandhisM.S.degreeandPh.DingeologyfromtheUni-versityofMichigan.HeiscurrentlyanassistantprofessorintheDepartmentofEarthandEnvi-ronmentalSciencesatWesleyanUniversity.Hisresearchfocusesonbiogeochemicalprocessesinterrestrialsoilsandmarinesediments.

JoyceM.Budai$DepartmentofGeologicalSciences,UniversityofMichigan,AnnArbor,Michigan,48109-1603;presentaddress:GreatLakesCollegesAssociation,535W.William,AnnArbor,Michigan,48103;budai@glca.orgJoyceBudaireceivedaB.A.degreeinEnglishattheUniversityofKansas,anM.S.degreein

geologyatRiceUniversity,andaPh.D.ingeologyattheUniversityofMichigan.WhilearesearchscientistattheUniversityofMichigan,herinterestsincludedthesedimentaryhistoryoftheMichiganbasinandtheroleoffluidflowduringdeformationoftheWyomingoverthrustbelt.JoyceisnowworkingoncollaborativeprogramsforfacultyatprivateliberalartscollegesinMichigan,Indiana,andOhio.

JenniferC.McIntosh$DepartmentofGeo-logicalSciences,UniversityofMichigan,AnnArbor,Michigan,48109-1603;jmcintos@umich.edu

JenniferMcIntoshiscurrentlyworkingonherPh.D.ingeologyattheUniversityofMichigan.ShereceivedherB.A.degreeingeology-chemistryatWhitmanCollegein1998andherM.S.degreeingeologyattheUniversityofMichiganin2000.Herinterestsincludethehydrogeochemistryofsedimentarybasins,theimpactofPleistoceneglaciationonregionalflowsystems,andmodi-ficationofbrinesviamicrobialmethanogenesis.MartinSchoell$GasConsultInternational

fallbetweenthecommonlyacceptedfieldsforthermogenicandmicrobialgas.Inaddition,theconfoundingisotopicandcomposi-tionaloverprintofmicrobialoxidation,increasingthe13CC1;C2;C3valuestotypicallythermogenicvalues,mayobscurethedistinctionbetweenmethanogenicandthermogenicgas.

INTRODUCTION

Unconventionalgasdeposits,suchasthoseproducedfromcoalbedsandshales,areanimportanthydrocarbonresource.Coalbedmeth-aneaccountsfor7%ofthetotalUnitedStatesnaturalgaspro-duction.Theorganic-richshalesoftheeasternUnitedStateshavealonghistoryofproductionandrecentdevelopmentandaccountfornearly2%ofUnitedStatesnaturalgasproduction.Since1988,shalegasproductionhasincreasedbyasmuchas60%resultingprimarilyfromasinglenewblack-shaleplay,theAntrimShaleintheMichi-ganbasin(Figure1)(HillandNelson,2000).Manyoftheseun-conventionaldepositscontainamixofthermogenicandmicrobialgas.

Naturalgasisgeneratedbybacterialmethanogenesis,thermalcrackingofkerogenorcoal,orbysecondarycrackingofoil.Bacterialmethanecomprisesmorethan20%oftheworld’sgasresources(Rice,1993).Themostcommonindicatorofbacterialgasisthepresenceofmethanewithlowd13Cvalues(<À55x).Inaddition,becausesomeintermediarymicrobialprocessesproduceCO2asaby-product,itspresenceandisotopiccompositioncansuggestaspecificbacterialprocess.Overallgaschemistryisalsoindicativeofitsoriginbecausemicrobialprocessesproduceonlymethaneinsignificantquantities(>1molvol.%),sothathigherchainhydro-carbonsareattributabletothermogenicgeneration.Differentiatinggasoriginiscomplicatedbythefactthatdifferentgenerationmechanismscanproducesimilarisotopicandcompositionalvalues.Secondaryprocesses,suchasmigration,bacterialoxidation,andmixing,furthercomplicateidentificationofgas-generationmecha-nismsbyalteringtheprimarydiagnosticsignatures.

Inunconventionalplays,theoriginofgasisfundamentaltoassessinganaturalgasreservoirandinguidingexplorationstrategies.Generally,athermogenicgasplayisexpectedtobemostproductiveinthedeepersectionsofabasinwheretheorganicmaterialhasex-periencedmorethermalcracking.However,inanunconventionalplaywhereinthegasispredominantlymicrobial,themarginsofthebasinwheretheorganicmatterislessmatureandhydrologicflowsystemsareactivewouldbethetargetofexploration.Theserela-tivelyshallowsettingsarecommontobothshaleandcoalbedgasdeposits.Inbothofthesecases,gasisabsorbedintotheshaleorcoalmatrixandisproducedbyloweringboreholepressureandallowingthegastodesorb.Thepresenceofmicrobialgasincoalbedreservoirshasbeenestablishedusingthed13CisotopiccompositionofCH4andCO2.FordepositsintheGreatArtesianbasin,Australia(SmithandPallasser,1996),andintheSanJuanbasin,UnitedStates(Scott

LLC,693St.GeorgeRd.,Danville,California,94526;gasconsult@hotmail.com

MartinSchoellisaninternationallyknowngeo-chemistwhospecializedduringhis30-yearcareeronisotopegeochemistryofoilandnaturalgas.Martinistheprincipalauthorandcoauthorofmanyfundamentalpapersongasgeochemistryanditsapplicationingasexplorationandpro-duction.From1984to2001,hewasaseniorscientistatChevronandworkedmostlyininter-nationalgasexploration.Henowworksasanindependentconsultant.

ACKNOWLEDGEMENTS

SupportforthisprojectwasgeneratedbythePetroleumResearchFund,administeredbytheAmericanChemicalSociety(PRFGrantNo.35927toL.M.W.andPRFGrantNo.36133toA.M.M.)andtheGasResearchInstituteundercontractnumber5094.WealsothankthenumerousgasoperatorsintheAntrimShalewhogavegenerouslyoftheirtimeandexpertiseinfieldsampling.Finally,thismanuscriptwasgreatlyaidedbythecommentsofMariaKopicki.

1356MicrobialProductionandModificationofGasesinSedimentaryBasins

Figure1.(a)DevonianshalegasresourcesintheeasternUnitedStateswithtotalgas-in-placeestimates(Walteretal.,1997).NPC=NationalPetroleumCouncil;USGS/DOE=U.S.GeologicalSurvey/DepartmentofEnergy.(b)TotalannualgasproductionforDevonianshalereservoirs(modifiedfromHillandNelson,2000).TheNewAlbany,Antrim,andOhioshalesarelocatedintheIllinois,Michigan,andAppalachianbasins,respectively.

etal.,1994;Schoelletal.,2001),microbialactivitywasidentifiedusingtherelativeconcentrationsandd13Cisotopiccompositionofmethane,CO2,anddissolvedinorganiccarbon(DIC)intheassociatedwater.

Givenitsimportanceasoneofthemostproductiveandrecentblack-shalegasplays,theAntrimShale,Michiganbasin,providesanidealsettinginwhichtostudymicrobialmethanogenesisandmodificationofthermogenicgas.TheMichiganbasinhasalonghistoryofthermogenicgasproductionfromdeepDevonian–Cambrianstrata.Itisonlysincethemid-1980sthatshallowblackshalesbecameatargetofrapiddevel-opment,beginningwith100wellsin1985toasmuchas12,000wellstoday(HillandNelson,2000).Earlierstudies(Martinietal.,1996,1998)focusedonthegasplaylocatedinthenorthernlowerpeninsulaofMichi-gan(Figure2a).Inthistrend,bothmicrobialandther-

mogenicgaswasidentified,althoughthebacterialgasdominated.

Thisarticlepresentsdatafromnewareasofde-velopmentinthebasinanddiscussesinparticularthehigherchainhydrocarbons.Anintegratedgeochemicalapproachisusedtounravelprimaryandsecondarygasproductionandthemultiplepostgenerativeprocessesthataffectboththecompositionandisotopicvaluesofthegases.Finally,thecasewillbemadethatdetermin-ingtheactualoriginofgasesbygeochemicalmethodsfacilitatesexplorationinemergingunconventionalgasreservoirs.

OVERVIEWOFNATURALGASCOMPOSITIONANDSTABLEISOTOPECHEMISTRY

Gascompositionaldifferences,particularlyintherel-ativeamountsofmethane,ethane,andpropane(C1,C2,andC3,respectively),havecommonlybeenusedtodistinguishbacterialfromthermogenicgasbecausemicrobesdonotproducesignificantquantitiesofhy-drocarbonsotherthanmethane(Bernardetal.,1978;Oremland,1981;RoweandMuehlenbachs,1999).However,puremethanecanbeproducedunderther-mogenicregimesand,moreimportantly,bymigrationofgasesthroughanorganic-richshaleorcoalbed.A‘‘chromatographic’’effectcanbeproducedduringmi-grationthroughthesebeds,leavingthehigherchainhydrocarbonsbehind(Schoell,1984;Jenden,1985;PriceandSchoell,1995).

Thed13Cofmethaneproducedviathermogenicprocessesiscontrolledbythed13Coftheprecursororganicmatterandthetemperatureofgasgeneration.Highertemperaturesproducemorepositived13CCH4values.Therelationshipbetweenthermalmaturityandd13Cofmethanehasbeendevelopedempirically(Gali-mov,1968;Stahl,1977;Schoell,1984;Jenden,1985)andmathematically(Berneretal.,1985;Clayton,1991;Rooneyetal.,1995;Tangetal.,2000).Post-generativeprocessescomplicatethesimplerelation-shipsbasedontheprimarykineticfractionationsforthermogenicgas.Theseprocessesincludemixingandsecondarymicrobialoxidation.Suchoxidativereac-tionsoccurinbothaerobic(Colemanetal.,1981)andanaerobicenvironments(Boetiusetal.,2000;ValentineandReeburgh,2000).

Thed13Cofbacterialmethaneisalsohighlyvar-iable,ranginginvaluefromÀ100toÀ31x.Itis

affectedbythed13Cofthemethaneprecursorandthe

Martinietal.

1357

Figure2.LocationofAntrimShaledevelopmentintheMichiganbasin.TheAntrimShalesubcropisshowninblack,andthenumberofwellssampledinthisstudyineachareaisgiven(n).Althoughninenewwellsarepresentedinthisarticleforthenorthernmargin,previousdatafromMartinietal.(1998)includemorethan180wells.

Inreaction1,thesurroundingH2Oprovidesonlyoneoffourhydrogenmoleculestotheproductmethane.Inreaction2,allthemethanehydrogenisrelatedtothedDofthesurroundingwater(Schoell,1980).Thishasbeendemonstratedexperimentally(Danielsetal.,1980;Balabaneetal.,1987)andstronglysuggestedinthenorthernproducingtrend(northernmargin)oftheAntrimShale(Martinietal.,1996,1998).Inreservoirswheregasandwaterhaveremainedincommunication,thehydrogenisotopecompositionofthewatercanbecorrelatedtobacteriallyproducedgas.Aconfoundingfactoristhat,inmanyenvironments,bothmetabolicpathwaysoccurdependingonthesubstratesavailable.

Microbialactivityiscommonlyidentifiedusingtheby-productsofthesemetabolicprocesses.Wherewateriscoproduced,anomalouslyhighDICcontentcoupledwithextremelyhigh(>10x)d13Cvaluesindicatesmicrobialactivity(Scottetal.,1994;Martinietal.,1998).Evenwherewaterisnotproduced,thequantityandisotopiccompositionofCO2canindicatemicro-bialactivity.Carbondioxideisaby-productofnumer-ousmetabolicpathways,includingacetatefermenta-tion.Infact,ahostofanaerobicreactionsformCO2,and,eveninareservoirproducingmethaneviaCO2re-duction,significantCO2gascanbegeneratedfrominter-mediarymicrobialreactions(Jenden,1985;Martinietal.,1998).

predominantmetabolicpathway(CO2reductionvs.acetatefermentation).Variableisotopefractionationcanalsoresultfromenvironmentalstressesbecauseoftemperature,nutrients,H+source,etc.(Whiticaretal.,1986;SugimotoandWada,1993,1995).Finally,thesamepostgenerativeprocessesthataffectthermogenicgaseswillalsomodifymicrobialgascompositions.

Thehydrogenisotopecompositionofthegascanalsodistinguishthermogenicfrombiogenicgassourceswhencoproducedwaterisavailable.Forthermogenicgases,controlsonthedDisotopicvalueofCH4aretemperature-dependentfractionationandthedDoftheprecursororganicmatter(Whiticar,1999).Thedeu-teriumconcentrationinmicrobialmethaneiscon-trolledbythedominantmetabolicpathway.Thetwopathwaysare

acetatefermentation:

CH3COOH!CH4þCO2

ðreaction1Þ

GEOLOGICSETTINGOFTHEANTRIMSHALETheMichiganbasinisanelliptical,intracratonicbasinwithastratigraphicrecordofintermittentsubsidencethroughoutthePaleozoic(HowellandvanderPluijm,1990).Basinmarginsroughlycoincidewiththeshore-linesofLakeMichiganandLakeHuron,whichsur-roundMichiganonthewest,north,andeast(Figure2).Likemanyothercratonicbasins,theMichiganbasincontainsathicksequenceofUpperDevoniansedi-ments,includingtheorganic-richAntrimShale.TheAntrimShaleisstratigraphicallycorrelativetosimilarblack,organic-richshalesofLateDevonianage,suchastheNewAlbanyShaleoftheIllinoisbasinandtheOhioShaleoftheAppalachianbasin,bothofwhichareknownhydrocarbonsources.

TheAntrimShaleconsistsoffinelylaminated,silty,pyritic,andorganic-richblackshalesinterbeddedwithgrayandgreenshalesandcarbonateunits.Itisdividedintofourmembers:theNorwood,Paxton,Lachine,andupperAntrim(GutschickandSandberg,

CO2reduction:CO2þ4H2!CH4þ2H2O

ðreaction2Þ

1358MicrobialProductionandModificationofGasesinSedimentaryBasins

1991).TheNorwoodandLachinemembershavethehighestorganiccontents(0.5–24wt.%totalorganiccarbon)andarethemaintargetsofgasexploration(Walteretal.,1996).

Kerogen(macromolecularinsolubleorganicmat-ter)andbitumen(solubleorganicmatter)havebeenevaluatedinAntrimShalesamplesbytheRock-Evalpyrolysismethodtodeterminedepositionalenviron-ment,oil-gassourcecapacities,andthermalmaturityofthesedimentaryorganicmatter(Rullko¨tteretal.,1992).TheNorwoodandLachinemembersoftheAntrimShalecontaindominantlytype1material.Thetype1materialisprobablyderivedfromTasmanites,aplanktonicalgaecommonintherestricted,epeiricseasthatdepositedtheAntrimShale(Dellapenna,1991).

ThethermalmaturityoftheAntrimShaleinthenorthernmarginoftheMichiganbasin,asdeterminedbyvitrinitereflectance(Ro),rangesfrom0.4to0.6%(Rullko¨tteretal.,1992),indicatingarelativelylowlevelofthermalmaturation(preoilgeneration)(Waples,1985).However,intheAntrimShaleofthecentralbasin,samplesreachanRoof1.0%,whichissufficienttogenerateoil(Rullko¨tteretal.,1992).However,suchvitrinitereflectancedatashouldbeviewedwithsomecaution,asithasbeenshownthatthepresenceofma-rinealgalmaterialmaydepressthereflectanceparam-eter(PriceandBarker,1985).

TheAntrimShaleisnaturallyfracturedacrossthestudyarea(Figure3a).Thefracturesareregionalandprobablyformedduringdeeperburial.Twodominantconjugatefracturesetsarepresent;onetrendstothenortheastandtheothertrendstothenorthwest.Bothsetsarenearlyvertical.Thefracturesetshavebeenob-servedthroughouttheDevoniansectionandacrossthenorthernmargin,andsimilarfracturesoccuraroundtheMichiganbasin.ForadetaileddiscussionoffracturingintheAntrimShale,refertoHolditchandAssociates(1996)orWalteretal.(1996)andreferencestherein.

ThefracturenetworkpresentintheAntrimShaleactsasareservoirandconduitforfluids,allowingmigrationandmeteoricrechargenearthesubcrop.Glacialmorainesinthenorthernandwesternmarginsformregionaltopographichighsthatprovidehydraulicheadformeteoricrecharge.ThehydrologicrechargetotheshaleoccursthroughtheglacialdriftthatblanketsMichiganandflowsintothesubcrop(Vugrinovich,1988;Martinietal.,1996,1998).Inaddition,theshalemayberechargedverticallyfromtheunderlyingper-meableTraverselimestone(Walteretal.,1996).Inthesouthernandeasterntrends,thereisalowerregionaltopographichighthatprobablyreducesmodernre-

charge.OxygenandhydrogenstableisotopicevidenceindicatesthatdilutewatersintheAntrimShaleandconfineddriftwellswererechargedundercooler-than-presentconditions,likelyduringthelastglacialinterval.Icesheetsmayhaveprovidedsignificanthydraulicheadanddilationoffracturesforpulsesofdilute,glacialrechargetoentertheshalealongthemarginsofthebasin(Martinietal.,1998).Glacialmeltwaterrechargesignificantlyenhancedmicrobialmethanogenesisintheage-equivalentNewAlbanyShalealongtheeasternmarginoftheIllinoisbasin.SimilargeochemicaltrendsinformationwaterandgaschemistrycanbeobservedintheNewAlbanyShalecomparedtotheAntrimShale(McIntoshetal.,2002).

CONTROLSONGASPRODUCTIONFROMTHEANTRIMSHALE

Gasesthataresampledatthewellheadareaffectedbythein-situconditionsofgasstorageandtransport.Thesefactorsmay,inturn,altertheisotopicvaluesoftheproducedgasandcertainlyaffectthegascomposi-tion.Productionpracticesmayalsoaffectthegascom-position,forexample,nitrogenfracturing,thepressuremaintainedattheborehole,thedrillingfluidused,andthedurationofproductionforagivenwell.

IntheAntrimShale,gasisgeneratedandstoredinsitu.IthasbeenestimatedthatfortheAntrimShale,roughly70–75%oftheproducedgasisdesorbedfromorganicmatterandclaywithintheshale,theremaindercomingfromfractureswithintheshale(Frantz,1996).Thestrongabsorptionofthevariousgascomponentsintheshalesuggeststhefollowing:

1.Thegasmustbeproducedbyloweringofpressureattheborehole(Figure3b)

2.Thegasislesslikelytohavemigratedbecauseofthestrengthofthebondsformed

3.In-situreservoirconditionsmaycontrolmicrobialproductionandconsumptionofhydrocarbons

4.Allgasesareunlikelytobeabsorbedequally;infact,CO2issorbedapproximatelythreetimesmorestronglythanCH4(Figure3c)Althoughmoststudiesongasabsorptionhaveconcentratedoncoalbedreservoirs(McLennanetal.,1995),organic-richshalesbehaveinasimilarmanner(Scottetal.,1994).InastudybyArrietal.(1992),experimentalresultsofsorptionbyCH4andCO2Martinietal.

1359

demonstratedthatgasesdidnotsorbindependently,butrathercompetedforthesamesites.BecauseCO2wasmorestronglysorbedinitially,aspressurescon-tinuallydropwithproduction,agreaterpercentageofCO2isdesorbed.Thiseffectbecomesverysignificantforbottom-holepressuresbelow300psi(Figure3c).Inareasofmicrobialmethanogenesis,whereCO2isaby-product,thisbecomesaproblemforproducersforwhomCO2isawasteproductthatcorrodespipesandmustberemovedviastripping.

Figure3.(a)PhotographofAntrimShaleatthePaxtonQuarry,AlpenaCounty,Michigan.Closelyspacedverticalfracturesetsactasconduitsforbothwaterandgas.(b)TypicaldrawdowncurveforanAntrimShalewellfield.Gasisproducedasthepressureattheboreholeisloweredbydewatering.(c)Variationsintherelativeconcentrations(molvol.%)ofCH4andCO2intheproducedgasaspressureisreducedattheborehole.Carbondioxideismorestronglyadsorbedintotheshaleandthusitsrelativeconcentrationincreaseswithdecreasingpressure.

1360MicrobialProductionandModificationofGasesinSedimentaryBasins

METHODSFieldSampling

Formationwaterandgassampleswerecollectedatthewellheadfrom72Antrimgaswells.TemperatureandpHmeasurementsweremadeimmediatelyinthefield.WatersamplesforDICandd13CDICanalysiswerefil-tered(0.45mmfilter),bottled,preservedwithzephi-ranchloridetopreventbiodegradation,andrefrigeratedimmediately.Samplesforalkalinityandanionanalysiswerefilteredimmediatelyandstoredonice.Samplesintendedforcationmeasurementswerefiltered,thenacidifiedtopH<2withultrapureHNO3.GeochemicalAnalysis

Cationanalysesweremadebyaninductivelycoupledplasma-atomicemissionspectrometrytechniqueusingaLeemanLabsPlasmaSpecIIIinstrument.Precisionwas±3%forNa,Ca,Mg,Sr,andSiand±5%forK.AnionsweremeasuredonaDionex4000iseriesionchromatographusingAS2,AS5,andAS10columnswithsuppressedconductivitydetectionataprecisionlevelof±2%usinggravimetricstandards.Alkalinitywasmeasuredinthelaboratorybyelectrometricend-pointtitrationusingaTIM90Radiometer-automatedtitrationsystem(precisionwas±2%).Totalalkalinitylargely(99%)consistedofcarbonatealkalinity(HCO3ÀandCO32À);organicacidandH2Scontributionswererelativelyinsignificant.

ThepH($6.5–7.5)andchlorinity(asmuchas4.5M)ofwaterintheAntrimShaleindicatethatHCO3ÀisthedominantcomponentofDIC,withlessthan10%contributiontoDICfromundissociatedcar-bonicacid(H2CO3).Tocorroboratethisassumption,manywatersampleswereanalyzedfortotalDICbytitrationcoulometryusinganUIC5011carboncoulom-eter.ThealkalinityandDICanalysesagreedwell,withintheanalyticalerrorof±2.0%.Becauseallwaterswereanalyzedfortotalalkalinity(bytitration)andthisalka-linityisoverwhelminglybicarbonate,thetermAlkcisusedinterchangeablywithDICthroughoutthisarticle.IsotopeAnalysis

Oxygenandhydrogenisotopeanalysesofwaterwereperformedonasubsetof26samples.OxygenisotoperatiosweredeterminedbytheCO2equilibrationmethod(EpsteinandMayeda,1953)usingamodifiedversionofthemethoddescribedbySockietal.(1992).

Aliquots(2mL)ford18OanalyseswereinjectedintoaCO2-chargedBectonDickinsonVacutainer,placedhorizontallyinawaterbathat25jC,andshaken.Equilibrationtimesrangedfromlessthan12hrfordilutegroundwaterstomorethan24hrforbrines.HydrogenisotoperatiosweremeasuredbyaZn-reductionmethoddescribedbyVennemanandO’Neil(1993).BothCO2andH2wereanalyzedonaFinniganMat-DeltaSmassspectrometer.Isotopicanalysesaregiveninthestandarddnotationrelativetostandardmeanoceanwater(SMOW)withprecisionsof±0.1and±2xford18OanddD,respectively.

ThecarbonisotopecompositionofDICwasdeterminedonasubsetof60samplesthatwerepre-servedwithZephiranchlorideandstoredinserumvialscappedwithaTeflonseptum.Off-lineCO2extractionusedanN2-purgedsyringewithsubsequentadditionofphosphoricacidinvacuo.Thed13CanalysesweremadeonaFinniganMat-DeltaSmassspectrometerwithananalyticaluncertaintyof±0.1x.Carbonisotopecom-positionsaregiveninthestandarddnotationrelativetoPeedeebelemnite(PDB).

Gascompositionsandisotopevaluesweredeter-minedonasubsetofsamplesfrom35wellsbyIsotechLaboratories,ChevronPetroleumTechnologyCo.,andShellWesternExplorationandProductionInc.Routinemethodsofgaschromatographywereappliedtode-terminetheconcentrationofC1,C2,C3,andC4hy-drocarbons.Theaccuracyofisotopemeasurementswas±0.1and±2.0xford13CanddD,respectively.ForgaseswithC2+concentrationsoflessthan1%,on-linecompoundspecificisotopeanalysiswasperformed.Thereproducibilityfortheseanalyseswere±1xformethaned13Cvaluesand±1.5xforethaneandpro-paned13Cvalues.

RESULTSSalinitySources

Salinitygradients,representedbyClÀconcentrations(Table1),indicatethatfreshwaterrechargeoccursnearthesubcropoftheAntrimShale(Figure4).Valuesrangefromlessthan0.1MClÀatthemarginsofthebasintoahighof5.9MClÀinthecentralbasin.Thechloridegradientissharpalongthebasinmarginswithmoregradualchangenearthebasincenter.Whereasformationwatersalinityislargelyirrelevantinthede-velopmentofthermogenicgas,salinitiesgreaterthan$4.0MClÀhavebeenshowntoinhibitmethanogenic

Martinietal.

1361

1362

Range6.26.417.815.5<0.050<0.050À4.57À6.33À8.42À12.50À13.20À28.3À43.1À54.2StatepHTemperature(jC)25.2927.8223.963.0028.39Density(g/cm3)Na(M)Cl(M)SO4(mM)HCO3(meq/kg)Acetate(mM)d13OH2O(SMOW)dDH2O(SMOW)d13CDIC(PDB)5E6E6E8E1E1E1E1E1E1E1E1E2E1E6W7.27.28.27.88.18.18.17.58.27.26.88.1À12.90À13.9018.514.216.015.915.118.020.018.020.019.025.911.90.0070.0080.0410.022<0.0020.0060.003<0.002<0.0020.002<0.0500.025<0.0080.0000.0400.0000.0180.000<0.0080.0000.0000.000<0.0800.012MichiganMichiganMichiganMichiganMichiganMichiganMichiganMichiganMichiganMichiganMichiganMichiganMichiganMichiganMichigan1.1571.1081.0711.0071.0011.0041.0031.0061.0051.0071.0051.0191.0181.1681.0003.5932.3921.90.1120.0120.00.0430.1010.0750.1390.0920.3880.3833.6020.0144.1462.7081.7180.1530.0020.0320.0180.0710.0490.1130.0720.3780.3754.4570.0034.98.917.65.59.836.527.040.431.841.128.7.539.04.110.326.999.5013W6.13.913W12.514.1<0.008À5.8027.5830.28<0.008À11.10À10.51À9.78À14.65À12.25À72.3À67.7À65.2À101.7À84.414W14W14W15W15W15W15W15W15W15W13.816.216.115.115.813.715.517.0MichiganMichiganMichiganMichiganMichiganMichiganMichiganMichiganMichiganMichiganMichiganMichiganMichiganMichiganMichigan6.66.46.26.46.66.76.36.46.56.46.91.1471.0711.0831.0781.0861.0361.0631.0621.0661.01.0571.0072.7711.4571.9142.5921.4091.5921.5571.7090.5051.1441.2091.2661.1311.0790.0754.1271.6812.5983.6101.7292.1412.0872.1860.7701.6281.5911.7091.4981.5030.0830.1070.0620.044<0.0210.0580.0360.0580.06813.325<0.05<0.05<0.050.2060.7030.0043.019.219.014.313.915.822.122.019.824.420.99.4<0.0080.082À12.41À84.2Table1.ElementalandIsotopicCompositionsofFormationWatersfromtheAntrimShale,MichiganBasinWellSectionTownshipNorthernMargin123456710111213141532321453229333332313241813128N28N28N28N32N32N32N32N32N31N31N31N31N28N32NMicrobialProductionandModificationofGasesinSedimentaryBasins

3425NWesternMargin161718192021222324252627282930339192311111212111325N23N23N24N23N23N23N23N23N23N23N15.137.67À69.01À537558026729380010001.......0000000<<4271049.......88574121121222666666622222220000000.......0000000<<<<<<<0855639145169347810.......11110119545770410380013798.......1111000075280065673550000000.......11111110972559.......77884111111190601.......6766677nnnnnnnaaaaaaagggggggiiiiiiihhhhhhhccccccciiiiiiiMMMMMMMWWWWWWW55555551111111NNNnNNN443334222232227709447221212345673333333000.2DDBB209601..00039013..55030767..22290022..1140..122279..55nnaaggiihhcciiMMEW11NN23982331598..7593..2143ÀÀ066209..425.ÀÀ2À707007303365100135884921374355..........3511020450DD6570712116..........BB1101011000800008003006555595595500000000100...........00000000000<<<<<<<<86822697396304477434499242839029............453555554554832958540839881778711527030336937466............222222222222350358169519004008108808221221221121............1111111111110020067787...023......61922D.822111N97807029332............554566665566nnnnnnnnnnnnaaaaaaaaaaaaggggggggggggiiiiiiiiiiiihhhhhhhhhhhhcccccccccccciiiiiiiiiiiiMMMMMMMMMMMMEEEE2231NNNN44324859212nigraMnrehtroN01234567014444444444559955...09932228....84844444ÀÀÀÀ78355322....8888ÀÀÀÀ46728288..........6344443511111111119203243081965240420001000000..........0000000000<7660259501002801184667556785..........1111111111486984590474601437478001011011..........0111111111310403448556666666660000000000..........11111111112212221222..........99999999999009122334..........5665666666aaaaaaaaaannnnnnnnnnaaaaaaaaaaiiiiiiiiiiddddddddddnnnnnnnnnnIIIIIIIIIInigraMnrehtuo23456701S5555555566Martinietal.

1363

CentralBasind13CDIC(PDB)dDH2O(SMOW)À37.4À100.0activity(Zinder,1993).TheAntrimShalehasbeenshowntoproducecopiousmicrobialgasfromareaswithsalinitieslessthan$4MClÀ(Martinietal.,1998).

DissolvedCarbonSpecies

ConcentrationsandisotopiccompositionofdissolvedcarbonspeciesarehighlyvariableintheAntrimShaleandareclearindicatorsofmicrobialactivity.Bothor-ganic(DOC)andinorganic(DIC)dissolvedcarbonspeciesarelinkedthroughmicrobialactivity.Specif-ically,DOCisbothanenergysourceforandameta-bolicproductofthemicrobialcommunity.Dissolvedinorganiccarbonisalsoaby-productofnumerousmeta-bolicpathwaysand,inthepresenceofhydrogen,canbeusedtogenerateCH4viaCO2-reducingmicrobes.

Oneoftheearliestrecognizedindicatorsofwide-spreadmicrobialactivityinthenorthernmarginoftheAntrimShalewasexceptionallyhighDICvalues(Mar-tinietal.,1998).GroundwatersintheglacialdriftanddeepMichiganbasinbrinesbothhaveapproximately0–5meq/LDIC(Figure5a).WatersintheAntrimShalealongthenorthern,western,andsouthernmar-ginsallhaveDICconcentrationsofmorethan10meq/L,whichcannotresultfromsimplemixingofdriftwaterandbasinalbrine.Instead,thesehighDICval-uesmustbearesultofmicrobialmetabolicpathwaysoccurringwithintheAntrimShale,ahypothesisthatisprovencorrectbytheveryhighd13Cvaluespres-ent(Figure5a).Fracture-fillingcalciteintheAntrimShalealongthenorthernbasinmarginalsocontainshighd13CDICvalues,indicatingthatcalciteprecipita-tioncoincidedwithmicrobialmethanogenesis(Budaietal.,2002).

Acetate(CH3COOÀ)isafavoredmetaboliteforfermentingbacterialconsortia.Itcanbeproducedbymicrobialmetabolismofmorecomplexorganicmole-cules,bylow-temperaturediagenesisoforganicmatter,andbythermalalterationoforganicmatterattem-peratureshigherthan80jC(CarothersandKharaka,1978).IntheAntrimShale,acetateconcentrationsarecommonlynegligiblewheretherearehighlevelsofmicrobialactivityasindicatedbyhighDICvalues.Inthenorthernandwesternmargins,concentrationsofacetateareloworundetectable(Figure5b),whereitislikelybeingconsumedasrapidlyasitisbeingproduced.Inthecentralbasin,acetateconcentrationsaremuchhigher,reflectingeitherthermalproductionorseverelyretardedconsumption.Thetwosamplesanalyzedfromthesouthernmarginshowwidelydivergentresults,

d13OH2O(SMOW)31.11HCO3(meq/kg)Acetate(mM)Density(g/cm3)Temperature(jC)RangeTownshipSectionTable1.ContinuedWell13

62636566676869MicrobialProductionandModificationofGasesinSedimentaryBasins

28198S7N2E5EIndianaIndianaIndianaIndianaIndianaIndianaMichiganOhioState6.46.46.46.46.56.55.86.9pH9.19.29.29.29.19.211.318.01.0681.0591.0621.0591.0621.0631.1261.0141.2271.0571.1961.0401.1221.1442.0790.261Na(M)1.8191.5711.5881.4701.42.1923.2470.307Cl(M)0.0460.0540.0460.0370.066<0.0210.5040.005SO4(mM)13.817.916.317.816.316.720.50.058À5.31À14.2027.84Figure4.Chloridevaluesincreasebasinward,indicativeoffreshwaterrechargearoundthemarginofthebasin(veryminorsmoothingofcontourlineswasperformed).M=mol/L.correspondingtotheirlocationwithrespecttothesubcrop.ThesamplewithlowacetateandhighDICconcentrationsislocatedinOhio,atshallowdepthsnearthesubcropoftheAntrimShale.Thesecondwell,withhigheracetateandnoDIC,isinMichigan,fartherfromthesubcropandcorrespondinglydeeper,withhighersalinities(i.e.,morelikebasinalbrine).

TherelationshipbetweenDOCandDICinAn-trimShalewatersiscomplex.Foragivencarbonsub-strate,decreasingconcentrationresultingfrommicro-bialconsumptionisreflectedinasystematicincreasein13Cconcentrationaslightercarbonspeciesareprefer-entiallyremoved.Bacterialconsumptionof1molofacetateproduces1moleachofCH4andCO2.Bacterialremovalofacetateproducesasignificantkineticiso-topeeffectof$10x,suchthattheremainingacetate

becomesenrichedin13C(Whiticar,1999).Asfurtheracetatemetabolismproceeds,theproducts(CO2andCH4)alsobecomeincreasinglyenrichedin13C.PropertiesandOriginsofProducedGases

Todeterminewhethermicrobialactivityisthedom-inantsourceofthemethanegeneratedfromtheAntrim

Shale,onemustexaminethechemistryofthepro-ducedgas.Methanecarbonisotopesandgascomposi-tionarecommonlyusedtoidentifygasorigins.Mich-iganbasingasesdifferfromconventionalmicrobialandthermogenicgasesinthattheyplotisotopicallyandcompositionallybetweenthegenerallyacceptedvaria-tionsofmicrobialandthermogenicgases(Figure6).Methaneinthethermogenicgasesfromthecentralbasinisisotopically13Cdepletedcomparedtomostotherthermogenicgases.Incontrast,methaneinwest-ernmargingasesisisotopicallysimilartothermogenicgases,yetdepletedinC2+compounds.Mixingof

Figure5.(a)Highconcentrationsofdissolvedinorganiccarbon(DIC)andhighd13CDICvaluessuggestwidespreadmicrobialactivityforthenorthern(Martinietal.,1998),southern,andwesternmargins.Incontrast,centralbasinbrinesandgroundwatershaverelativelylowDICandd13CDICvalues.PDB=Peedeebelemnite.(b)Acetate(meq/L)vs.DIC(meq/L)forwatersfromthewesternandsouthernmarginsandcentralbasin.Thenorthernmargindataareshowninthepatternedrectangle(Walteretal.,1997).

Martinietal.

1365

generaltrendwillbetoamorebiogenicsignaturecompositionally(James,1983).Inaddition,thebac-terialoxidationofhydrocarbonsleadsto13Cenrich-mentintheresidualcomponent.Thechangesinthecarbonpoolanditsisotopiccompositioncanbede-scribedbyaclosed-systemRayleighfunction:

d13CC;t¼d13CC;iþalnð1ÀFÞ

ðequation1Þ

Figure6.ModifiedBernardplotshowingdatafromtheAntrimShale.Notethatthelowestvalueford13CCH4isfoundinthecentralbasin,representativeofimmaturethermogenicgas.DatafortheAntrimShalefallbetweenthefieldscommonlysuggestedforthermogenicandbiogenicsources.Thisistheresultofthepresenceofanimmaturethermogenicgasbeingdilutedbyvariableamountsofmicrobialgas.PDB=Peedeebelemnite.

thermogenicandmicrobialgasescontainingsimilard13CCH4valuesisonepossibleexplanationforthein-termediarycompositionofMichiganbasingases.However,thiswouldbeahighlyunlikelymixingtrend.Inaddition,manyofthemorethermogenicgases(i.e.,highC2andC3concentration)actuallyhavelowerd13Cvaluesthanthepresumedmicrobialgases.Meth-aned13Cvaluesvaryconsiderably.Inshort,standardinterpretiveschemesdonotapplytoMichiganbasingases.

Bacterialconsumptionofhydrocarbongasesmodi-fiesboththeisotopiccompositionofthegasfractionsandthecompositionofthegasitself.Thisoxidativeprocesshasbeenshowntooccurbothanaerobicallyandaerobicallyandcouldaccountfortheanomalouspat-tern.Inanaerobicenvironments,oxidationoccursatthebaseofthesulfatereductionzone(IversenandJørgensen,1985;BlairandAller,1995;Fossingetal.,2000).Inshallowhydrologicsystems,bacterialoxida-tionoccurswherehydrologicconditionspermitdis-solvedO2toreachthehydrocarbonsstoredinthereservoir.Althoughbacteriaconsumeallthegaseoushydrocarbons,ithasbeensuggestedthatmethaneoxidationoccursmostrapidly(Whiticar,1999).Inthisscenario,asoxidationcontinues,theC1/[C2+C3]ratiodecreases,movingtowardamorecompositionallyther-mogenicsignature.However,ifhigherchainhydro-carbons(C2andC3)aremorereadilyoxidized,the1366

whered13CC;iistheisotopiccompositionofthehy-drocarbonpriortooxidation,d13CC,tisthecomposi-tionatsometime(t),Fisthefractionofhydrocarbonremaininginthepool,andaisthefractionationfactorforbacterialoxidation(typicallybetween5and30xandadjustedtofitthedataobserved)(Whiticar,1999).

Thed13Cofethaneandpropanehavebeenplottedvs.themolevolumepercentofeachspeciesinFigure7.Forthesetwocomponents,theprocessofoxidationisreadilyapparent,incontrasttomethane(seeFigure6).Methaneoxidationisobscuredfortwopossiblereas-ons.Becausemethaneisthedominantgasintheres-ervoir,comprisingbetween88and99molvol.%inthenorthernandwesternmargins,bacterialoxidationcon-sumesarelativelysmallproportionofthislargepool.Inaddition,microbialmethanogenesiswithintheres-ervoirmaskstheRayleightrendbycontributingtothemethanepool.Incontrast,therelativelylowconcen-trationsofC2andC3($0–11molvol.%)coupledwithasinglesource(thermogenicproduction)allowtheoxidationtrendtobeclearlyseenforthesehydro-carbons.

Thewellslocatedinthewesternmarginshowthegreatestdegreeofoxidationbasedsolelyonthehighd13CvaluesforC2andC3.Gasesfromthecentralbasinhavelowd13CC2andd13CC3valuesandcontainthegreatestconcentrationofthesegases,indicativeofanimmature,thermogenicorigin(James,1983;Coleman,1999;RoweandMuehlenbachs,1999;Whiticar,1999).ThestrongoxidationeffectsseeninC2andC3frac-tionscanbecorrelatedtotheinfluxoffreshwateratthesubcropbyusingthed18OH2O(xSMOW)asaproxy.ThisrechargemaytransportdissolvedoxygenorSO42ÀtotheAntrimShalereservoirinitiatingCH4oxidation.Theverylowd18OH2OvaluescorrespondtotheveryhighC2andC3d13Cvaluesasexpected(Fig-ure8).Hereagain,thed13CCH4valuesremainnearlyinvariantovertheentireAntrimShale.IntheAntrimShale,microbialoxidationofC2+gasesleadstoapro-gressiveincreaseintheC1/[C2+C3]parameter,makingthegasappearmorebacterialinorigin.Oxi-dationofC2+hydrocarbonshasalsobeenseeninthe

MicrobialProductionandModificationofGasesinSedimentaryBasins

Figure7.(a)d13Cvs.concentrationforethane;(b)d13Cvs.concentrationforpropane.BothindicateaRayleighdistillationreactionbetweenconcentrationandisotopicvalues.Thissuggeststhatoxidationisoccurringforthesehigherchainhydrocarbons.PDB=Peedeebelemnite.

NewAlbanyShaleandtheFruitlandcoalbedgases(Schoelletal.,2001;Walteretal.,2001;McIntoshetal.,2002).

MicrobialoxidationisclearlyindicatedwhenC2andC3d13Cvaluesarecrossplotted(Figure9).Becausethesegasesarethermogenicinorigin,covarianceofC2andC3isexpectedasaresultofthematurationoforgan-icmatterfollowingthegeneralempiricalrelationshipd13CC3¼0:767Âd13CC2À4:1ðJenden;1985Þ

ðequation2Þ

Thiscorrelationrelatestothethermalmaturityofthesourcematerial;thehigherthematurity,thehigherthed13Cofthegases.However,fortheAntrimShale,thiscorrelationdoesnotfitwithmaturityoftheshale,butinsteadisaby-productofmicrobialoxidation.Thethermalmaturityoftheshalearoundthebasinisfairlywellconstrained(Rullko¨tteretal.,1992).Forthecen-tralbasinwellswithanRo%1.0andaveragevalueofd13CC2¼À45,ad13CC3valueofÀ37isobtained.Usingequation2,thesenumbersfittheexpectedcorrelationbetweenC2andC3foundinthermogenicgases.Ontheotherextremearewellslocatedinthewesternmarginwhere,basedonthehydrogeochemi-caldata,ahydrologicallyopensystemhasexperiencedglacialrecharge.Here,valuesforC2andC3areap-proximatelyÀ36andÀ15x,respectively.Whencomparedtoareasinthenorthernmarginwiththesamethermalmaturity(Ro%0.5),thesevaluesareenrichednearly15xforethaneand20xforpro-paneandsuggestthatoxidationhasoccurred.BecauseallC2andC3isassumedtobeproducedthermally,itisexpectedthatnorthernmarginwellsnotaffectedbyoxidationshouldhavelowd13Cvaluesforbothethaneandpropanecommensuratetotheirlowmaturity.Thedataclearlyfollowthistrendwithrespecttoethane,forwhichvalueslowerthanÀ50xarerecordedinsomeofthedeepest,mostsalinewellsinthenortherntrend(Figure9).

LinksbetweenWaterandGas

Duringmicrobialmethanogenesis,adirectlinkexistsbetweenthehydrogenisotopecompositionofH2OandCH4.ThestrongregionalgradientofhydrogenisotopecompositionsofwatersintheAntrimShalepermitschemicaltracingofhydrogensourcedfromH2O(Martinietal.,1998).Formostofthewestern,northern,andsouthernmargins,thedDCH4H2Orela-tionshipindicatesthedominanceofbacterialCO2reduction,fittingtheequation

dDCH4¼dDHÞ2Oþ160ðÆ10xÞðSchoell;1980ÞðFigure10ðequation3ÞThepredicteddDofCH4producedbyathermo-genicmechanismdependsontheorganicsourcematerialandonthedegreeoforganicmaturity.Gen-erally,thedDofCH4inthermogenicgasesismorepositivethaningasformedbymicrobialpathways

Martinietal.

1367

Figure8.d13CofC1,C2,andC3vs.d18OH2O.Thed18OH2OvalueisusedasaproxytoindicatetherechargeofdilutewatersintotheAntrimShale.Theoxida-tionofthehigherchainhydrocarbonsislinkedtotheinfluxinthesedilutewaters.PDB=Peedeebelemnite;SMOW=stan-dardmeanoceanwater.

(Schoell,1980).Regardlessoftheinitialvalue,how-ever,theisotopiccompositionofthermogenicmeth-aneshouldbedistinctfromthecoproducedwater,unlessisotopicequilibriumbetweenCH4andH2Oisestablished.Thisprocessisnotoriouslyslowattemper-aturescommoninmostcratonicsedimentarybasins;evenat200jC,CH4shouldstillhaveadDvalue65xlowerthanthatofthecoproducedwater(Schoell,1980).Forlowertemperatures,suchasthosefoundintheAntrimShale,thepredicteddDCH4wouldbeeven

morepositive(Schoell,1980).IntheAntrimreservoir,isotopicequilibriumwouldleadtohighdDvaluesthatarenearly100xgreaterthanthoserecorded.MethanefromthecentralbasinhasdDvaluesconsistentwithathermogenicoriginandnoevidenceofequilibriumwithassociatedwaters.

AnotherlinkbetweentheproducedgasandwaterliesinthecarbonisotopecompositionsofCO2(g)andDIC.TheisotopiccompositionofCO2gasiscontrolledbycarbonateequilibriuminthewaterandbyfrac-tionationresultingfrommicrobialactivity.Usingme-dianvaluesofd13CDIC(28x)andtemperature(15jC)

13intheAntrimShale,thedCoftheCO2inequilib-riumwithDICshouldbeapproximately19x(Whi-ticar,1999).Forthenorthern,southern,andwesternmargins,thesystemisinequilibriumasindicatedbyanapproximately8xfractionationbetweenthetwocarbonreservoirs(seeTables1,2).Inareservoirun-dergoingCO2reduction,thepredictiverelationshipbe-tweencarbonisotopecompositionofCO2andCH4isasfollows:

aC¼ðd13CCO2þ1000Þ=ðd13CCH4þ1000Þ

ðequation4Þ

Forthenorthern,western,andsouthernmargins,thefractionationobserved(aC$1.070)isconsistentwithmicrobialmethanogenesisviaCO2reductionunderclosedconditions.Onesamplefromthesouthernmarginfallsalongafractionationtrendof1.060andisfromadeepwellinMichigan,wheremicrobialactivityhasbeeninhibitedbyhighsalinity(seeTables1,2).ThefactthattheCO2presentasanabsorbedgasintheshale

Figure9.Ethaneandpropaned13Cvaluescompared.Thegraytrendrepresentstheexpectedpathforthermogenicgasesasmaturityincreases.PDB=Peedeebelemnite.1368

MicrobialProductionandModificationofGasesinSedimentaryBasins

Figure10.Hydrogenisotopecompositionsfromcoproducedmethaneandwater.ThecovariancebetweenthedDofmethaneandwaterisindicativeofmethanogenesisviatheCO2-reductionpathway.ThetrendlinerepresentsthetheoreticallypredictedmicrobialCO2-reductionpath(Schoell,1980).SMOW=standardmeanoceanwater.

isinequilibriumwiththeDICinthewatersuggestsrelativelyrapidexchangebetweenthetworeservoirsofcarbon.Inaddition,theclearcorrelationforCO2-reductionfractionationbetweenmethaneandcarbondioxidelinksthesetworeservoirsformostAntrimShalewellslocatednearthemarginofthebasin.Interestingly,intheseareas,thed13Cvaluesofethaneandpropaneprovideevidenceofintensemicrobialox-idation.Aby-productofthisoxidationisCO2,whoseisotopiccompositionissimilartothed13Cofthepre-cursormethane,ethane,orpropane.

DISCUSSION

HistoryofAntrimShaleGasGeneration

PrimarygenerationofthermogenicgasoccurredduringthePermian(CerconeandPollack,1991).Thegasgeneratedvariedincompositiondependingonthethermalmaturityinagivenlocation.Thisvariationisshowninacomparisonofthreewells,oneeachfromthecentralbasinandthesouthernandnorthernmargins(showninFigure11).Withdecreasingthermalmaturity,recordedbyvitrinitereflectance(Ro),the13Cconcentrationintheethanedecreases.Thisisconsistentwithresultsfromgas-generationmodels(James,1983;RoweandMuehlenbachs,1999).Al-thoughthenorthernmargind13CC2valueofÀ51xisunusualformostthermogenicgasfields,similarvalues

havebeennotedincasesofextremelyimmatureor-ganicmatter(Mattevellietal.,1992;Coleman,1999;RoweandMuehlenbachs,1999).Asexpected,theamountofC2gasdecreaseswithdecreasingthermalmaturity.Becausethenorthernmarginwellisfromthesouthernedgeoftheproducingarea(i.e.,farthestfromthesubcrop),nosignificantoxidationhaslikelyoc-curred.Increasingd13CCO2valuessignifymicrobialactivitythatisresponsiblefortheshiftindatapointstotherightinFigure11.

Thefollowinggas-generationscenarioismostconsistentwithourdata.Initialthermogenicgasatthebasinmarginscontainssomefractionofethanewithanoriginald13CofapproximatelyÀ51x.Severalepisodesofmeteoricrechargeintheseareasfollowed,perhapsmostsignificantlyduringperiodsofrecentglaciationwherehydraulicheadwasprovidedby1-to2-km-thickicesheets.DissolvedO2mayhavepene-tratedtheAntrimShalecausingrapidoxidationofC1,C2,andC3andleavingbehindCO2withlowd13Cvalues.Morelikely,thedissolvedO2waslimitedandaperiodofsulfatereductionoccurredwithintheaquifer,theSO42ÀbeingcarriedwiththegroundwaterthroughthesubcroporflowingupfromtheTraverselimestonebelowtheAntrimShale.WepresumethatthetrendsseeninFigure8areproducednotbytheinfluxofoxygenateddilutewater,butbydilutewatercontainingSO42À.

Recently,theprocessofanaerobichydrocarbonoxidationhasbeenverifiedinhydratesedimentsontheseafloor(Sassenetal.,1998;Hinrichsetal.,1999;Boetiusetal.,2000).Boetiusetal.(2000),usingspecific1GSrRNA-targetedoligonucleotideprobes,identifiedacolonyofArchaeaencasedinalayerofsulfate-reducingbacteria(SRB)inthemethanehydratesedimentsoffthecoastofOregon.Interestingly,al-thoughisotopicindicationofCH4oxidationwasfound(extremelylowd13CinHCO3Àandcalcitecement),noaerobicoxidizerswereidentified.

MethanogensandSRBworkinginasyntrophicrelationshipcreateathermodynamicallyfavoredseriesofreactions(ValentineandReeburgh,2000):CH4þ2H2O!CO2þ4H2methanogens

ðreaction3Þ

SO42Àþ4H2þHþ!HSÀþ4H2Osulfatereducingbacteria

ðreaction4Þ

SO42ÀþCH4!HCO3ÀþHSÀþH2Onetreactionðreaction5Þ

Martinietal.

1369

1370

State77.3490.8186.46À233.219.63.961.400.150.920.300.053.272.095.7614.335.327.58À54.2À55.0À53.7À54.1À51.9À50.2À34.2À37.67.516.1À46.2À48.0À36.3N2d13CCH4d13CC2H6d13CC3H8dDCH4d13CCO2logMethaneEthanePropaneCO2(PDB)(PDB)(SMOW)(PDB)C1/(C2+C3)(mol%)(mol%)(mol%)(mol%)(mol%)(PDB)1.201.732.96.3696.750.000.001.102.150.040.001.002.603.36.5696.67À47.20.010.030.000.003.161.020.282.28À49.4À52.3À34.316.73.983.51À37.23.865E6E6E8E1E1E1E1E1E1E1E1E2E1E6W79.600.010.010.01À53.4À57.4MichiganMichiganMichiganMichiganMichiganMichiganMichiganMichiganMichiganMichiganMichiganMichiganMichiganMichiganMichigan92.4192.4886.6685.7786.4166.0160.370.110.060.171.760.240.331.250.120.110.160.120.071.150.607.076.154.945.834.833.693.42À51.2À246.119.81.726.255.556.7322.4928.84À49.9À49.0À49.7À49.8À50.3À49.6À37.1À38.0À36.9À35.5À37.7À15.0À14.5À16.0À17.4À14.7À239.9À254.3À256.1À260.7À259.5À256.417.521.722.821.922.421.11.492.382.292.512.572.4013WMichiganMichiganMichigan13WMichiganMichigan14WMichigan14WMichigan14WMichigan15WMichigan15WMichigan15WMichigan15WMichigan15WMichigan15WMichigan15WMichiganTable2.CompositionandIsotopicValuesforGasProducedfromAntrimShaleWellsWellSectionTownshipRangeNorthernMargin123456710111213141532321453229333332313241813128N28N28N28N32N32N32N32N32N31N31N31N31N28N32NMicrobialProductionandModificationofGasesinSedimentaryBasins

3425NWesternMargin161718192021222324252627282930339192311111212111325N23N23N24N23N23N23N23N23N23N23N740381...2125.229.752À4.24À1.05À10855134....000171084123....779111...000472215...01013056132....901998nnnnnnnaaaaaaagggggggiiiiiiihhhhhhhccccccciiiiiiiMMMMMMMWWWWWWW55555551111111NNNnNNN4433334222222277094472212123456733333336936785978.....0000047..477722ÀÀ140...787333ÀÀÀ740...475444ÀÀÀ091...474555ÀÀÀ79108150...26.02131472011...0003079969843.....241336240249985.....6148710906639841.....71166578nnnnnnnnnnnnnnaaaaaaaaaaaaaaggggggggggggggiiiiiiiiiiiiiihhhhhhhhhhhhhhcccccccccccccciiiiiiiiiiiiiiMMMMMMMMMMMMMMEWEEEE112231NNNNNN234432984859221201234567013344444444445562.15.226.44À8.05À27.029.804.092.473.58aaaaaaaaaaaannnnnnnnnnnnaaaaaaaaaaaaiiiiiiiiiiiiddddddddddddnnnnnnnnnnnnIIIIIIIIIIII2345670123555555556666Martinietal.

1371

CentralBasinSouthernMarginN2d13CCH4d13CC2H6d13CC3H8dDCH4d13CCO2logMethaneEthanePropaneCO2(PDB)(PDB)(SMOW)(PDB)C1/(C2+C3)(mol%)(mol%)(mol%)(mol%)(mol%)(PDB)Inthisprocess,methanogens‘‘reverse’’theirmetabo-lismanduseH2Oasaterminalelectronacceptor.The

H2istransferredtotheSRBforthereductionofSO4.Methanogenshavebeenshowntomethanotrophduringgrowth;however,therateofmethanogenesisisordersofmagnitudegreaterthantherateofmeth-aneoxidation(ZehnderandBrock,1979).Analter-nativemechanismforanaerobichydrocarbonoxida-tioninvolvestheformationofaceticacidandH2from2CH4molecules(ValentineandReeburgh,2000):2CH4þ2H2O!CH3COOHþ4H2ðreaction6ÞValentineandReeburgh(2000)haveshownthatthepreviousreactionpathways(reactions3,4,and5)foranaerobicmethaneoxidationareonlymarginallythermodynamicallyfavored.The‘‘excess’’energyiscreatedviatheSRB’suseofH2providedbythemethano-gens.Inreturn,energyistransferredbacktotheArchaeafromtheSRBs.MoreenergymaybeavailableviatheuseofC2orC3asafoodsource.ThismaybeamoretenablehypothesisfortheAntrimShale,becausethepresenceofsignificantdissolvedO2intheaquiferisunlikely.SupportcomesfromthefactthattheareaswiththegreatestamountofC2andC3oxidation(i.e.,thewesternmargin)alsoproduceH2SatthewellheadandhavesomeofthehighestSO42ÀconcentrationsrelativetosalinityfoundintheAntrimShaleformationwaters(Walteretal.,1997).Also,theTraverselime-stone,whichisincontactwiththelowermember(Norwood)oftheAntrimShale,isrichinSO42ÀandH2S.Finally,theAntrimisapyriticblackshaleinwhichdissolvedoxygenwouldbeunabletopenetratedeeply(Petschetal.,2001).

Eventually,asthesulfatereducersexhaustedtheavailableSO42À,themicrobialcommunitywithintheaquiferconvertedtomethanogenicactivity,likelyusingbothacetateandCO2-4H+assubstrates.However,theacetateconcentrationswerelimitedsothatCO2re-ductionbecamethedominantprocess.Thismostre-centperiodofintensemicrobialmethanogenesishaslikelylastedforatleastthepast$14,000years,afterthemostactivehydrologicconditionsduringglaciationterminated.Earlyoninthisprocess,thelowd13CCO2poolfromoxidationwouldhavebeenusedtoproduceextremelylowd13CCH4,asshownbelow:

CO2þ4H2!CH4þ2H2O

withd13C

À50x

À120x

ðreaction7Þ

0.781.162.96À43.0À54.9À55.3À42.9À46.8À46.1À37.2À238.6À235.724.25.319.10.7943.5228.002.0.174.462.97.271.800.255.784.760.071.0045.4673.1762.2427.18SectionTownshipRangeTable2.Continued28193623348S7N7N36N36N2E5E3E13E13E1372MicrobialProductionandModificationofGasesinSedimentaryBasins

EasternMargin73Well6566676869707172MichiganIndianaIndianaIndianaIndianaMichiganOhioOhioIndianaIndianaState84.626.133.451.651.226.031.04Figure11.Carbonisotopevaluesofethanevs.carbondioxide.Thermalmatu-rityofthreesamplesfromdifferingpartsofthebasinareindicatedbyvitrinitereflectancedata(Ro)(Rullko¨tteretal.,1992).Theisotopiccompositionofthegasesiscontrolledbyoriginalthermalmaturityandissubsequentlymodifiedbysecondaryprocesses,suchasmethanogen-esisandanaerobicoxidation.

Someofthemethanehaslikelydegassedfromthereservoirintotheoverlyingglacialdrift,aprocessthatisoccurringconstantly.Overtimeinthisrelativelyclosedsystem,themicrobialbreakdownofkerogenhasledtoa13C-enrichedpoolofcarbonspecies.

highDICcontent(>10mM)andd13CDICvalues(>10x)offormationwaters

󰀋lowsulfateconcentrations(<0.05mM)

󰀋increasedreservoirpermeabilityviahighdensityoffracturenetworksinshalesorcleatsincoalbeds󰀋highCO2(g)concentrations(>2%)andhighd13CCO2values(>10x)

󰀋

SignificanceforExploration

Inimmature,organic-richshalesaswellasincoalbeds,

thepresenceofbacterialmethanemayincreasetheireconomicpotential.Identifyingzoneswherecondi-tionsarefavorableforactivemicrobialmethanogenesiscanbeaccomplishedviageochemicalanalysisofgasandcoproducedwater.Amulticomponentanalysisiscriticalforunravelingthesometimes-complexgeo-chemicalhistorypresent.

Althoughoptimalconditionsformicrobialgasplaysaredifficulttodelineate,aswellashavinglikelychangedduringthegeologichistoryofagivenregion,thefollowingaresuggestiveofmicrobialcolonization:

󰀋

diluteformationwaters(<3MClÀ)coproducedwiththegas

NotethathighCO2(g)concentrationsarefoundinmanygasfieldsofthermogenicorigin,soitiscriticaltodeterminethed13CvaluefortheCO2.Suchcommonindicatorsasthed13Cofmethaneandgaswetnesscanbemisleadingbecauseofthevariousmicrobialmodi-ficationsthatmayoccur.

Thisstudyhasshownthatisotopicandcomposi-tionaldatafrommixedmicrobial/thermogenicgasfieldsmustbecarefullyexamined.Compositionalparametersoncethoughttouniquelyidentifythegas-generationprocesscanoverlapandbeobfuscatedbysecondarymicrobialeffects.Inaddition,theprimaryrangesofimmaturethermogenicd13CvaluesforC1andC2gasesarequitelow($À55xformethaneandÀ48xforethane)andoverlapd13Cvaluesformicrobialproduc-tionofmethaneinmanyenvironments.

Martinietal.

1373

REFERENCESCITED

Arri,L.E.,D.Yee,W.D.Morgan,andM.W.Jeansomme,1992,

Modelingcoalbedmethaneproductionwithbinarygassorption:SocietyofPetroleumEngineers,SPEPaper24363,p.459–472.

Balabane,M.,E.Galimov,M.Hermann,andR.Letolle,1987,

Hydrogenandcarbonisotopefractionationduringexperi-mentalproductionofbacterialmethane:OrganicGeochem-istry,v.11,p.115–119.

Bernard,B.B.,J.M.Brooks,andW.M.Sackett,1978,Light

hydrocarbonsinrecentTexascontinentalshelfandslopesediments:JournalofGeophysicalResearch,v.83,p.4053–4061.

Berner,U.,E.Faber,G.Scheeder,andD.Panten,1985,Primary

crackingofalgalandlandplantkerogens:kineticmodelsofisotopevariationsinmethane,ethaneandpropane:ChemicalGeology,v.126,p.233–245.

Blair,N.E.,andR.C.Aller,1995,Anaerobicmethaneoxidationon

theAmazonshelf:GeochimicaetCosmochimicaActa,v.59,no.21,p.3707–3715.

Boetius,A.,K.Ravenschlag,C.J.Schubert,D.Rickert,F.Widdel,

A.Giesek,R.Amann,B.B.Jørgensen,U.Witte,andO.Pfannkuche,2000,Amarinemicrobialconsortiumapparentlymediatinganaerobicoxidationofmethane:Nature,v.407,p.623–626.

Budai,J.M.,A.M.Martini,L.M.Walter,andT.C.W.Ku,2002,

Fracture-fillcalciteasarecordofmicrobialmethanogenesisandfluidmigration:acasestudyfromtheDevonianAntrimShale,Michiganbasin:Geofluids,v.2,p.163–183.

Carothers,W.W.,andY.K.Kharaka,1978,Aliphaticacidanionsin

oil-fieldwaters—implicationsfororiginofnaturalgas:AAPGBulletin,v.62,p.2441–2453.

Cercone,K.R.,andH.N.Pollack,1991,Thermalmaturityofthe

Michiganbasin,inP.A.CatacosinosandP.A.DanielsJr.,eds.,EarlysedimentaryevolutionoftheMichiganbasin:GeologicalSocietyofAmericaSpecialPaper256,p.1–11.

Clayton,C.,1991,Carbonisotopefractionationduringnaturalgas

generationfromkerogen:MarineandPetroleumGeology,v.8,p.232–240.

Coleman,D.D.,1999,Theuseofgeochemicalfingerprintingto

identifymigratedgasattheEppsundergroundgasstoragefield:SocietyofPetroleumEngineers,SPEReprintSeries50,p.66–75.

Coleman,D.D.,J.B.Risatti,andM.Schoell,1981,Fractionationof

carbonandhydrogenisotopesbymethane-oxidizingbacteria:GeochimicaetCosmochimica,v.45,p.1033–1037.

Daniels,L.,G.Fulton,R.W.Spencer,andW.H.Orme-Johnson,

1980,OriginofhydrogeninmethaneproducedbyMethano-bacteriumthermoautotrophicum:JournalofBacteriology,v.141,p.694–698.

Dellapenna,T.M.,1991,Sedimentological,structural,andorganic

geochemicalcontrolsonnaturalgasoccurrenceintheAntrimformationinOtsegoCounty,Michigan:Master’sthesis,WesternMichiganUniversity,Kalamazoo,Michigan,147p.Epstein,S.,andT.Mayeda,1953,Thevariationin18Ocontentof

waterfromnaturalsources:GeochimicaetCosmochimicaActa,v.4,p.213–224.

Fossing,H.,T.G.Ferdelman,andP.Berg,2000,Sulfatereduction

andmethaneoxidationincontinentalmarginsedimentsinfluencedbyirrigation(south-eastAtlanticoffNamibia):GeochimicaetCosmochimicaActa,v.,no.5,p.7–910.Frantz,J.,1996,TechnologyapplicationsimproveAntrimShale

wellrecoveriesandeconomics:GasTIPS,v.2,p.5–11.

Galimov,E.M.,1968,Isotopiccompositionofcarboningasesof

thecrust:InternationalGeologyReview,v.11,no.10,p.1092–1104.

Gutschick,R.C.,andC.A.Sandberg,1991,LateDevonianhistory

ofMichiganbasin,inP.A.CatacosinosandP.A.Daniels,eds.,EarlysedimentaryevolutionoftheMichiganbasin:GeologicalSocietyofAmericaSpecialPaper256,p.181–202.

Hill,D.G.,andC.R.Nelson,2000,Gasproductivefractured

shales:anoverviewandupdate:GasTIPS,v.6,p.4–13.

Hinrichs,K.,J.M.Hayes,S.P.Sylva,P.G.Brewer,andE.F.

DeLong,1999,Methane-consumingarchaebacteriainmarinesediments:Nature,v.398,p.802–805.

Holditch,S.A.,andAssociates,1996,Antrimshaledevelopment

technologyprojectfinalreport:GasResearchInstitute,GRI-96/03,149p.

Howell,P.D.,andB.A.vanderPluijm,1990,Earlyhistoryofthe

Michiganbasin:subsidenceandAppalachiantectonics:Geo-logy,v.18,no.12,p.1195–1198.

Iversen,N.,andB.B.Jørgensen,1985,Anaerobicmethane

oxidationratesatthesulfate-methanetransitioninmarinesedimentsfromKattegatandSkagerrak(Denmark):AmericanSocietyofLimnologyandOceanography,v.30,p.944–955.James,A.T.,1983,Correlationofnaturalgasbyuseofcarbon

isotopicdistributionbetweenhydrocarboncomponents:AAPGBulletin,v.67,p.1176–1191.

Jenden,P.D.,1985,AnalysisofgasesintheEarth’scrust:Gas

ResearchInstituteReport,GRI-85/0106,222p.

Martini,A.M.,J.M.Budai,L.M.Walter,andM.Schoell,1996,

Microbialgenerationofeconomicaccumulationsofmethanewithinashalloworganic-richshale:Nature,v.383,p.155–157.Martini,A.M.,L.M.Walter,J.M.Budai,T.C.W.Ku,C.J.Kaiser,

andM.Schoell,1998,Geneticandtemporalrelationsbetweenformationwatersandbiogenicmethane:UpperDevonianAntrimShale,Michiganbasin,USA:GeochimicaetCosmo-chimicaActa,v.62,p.1699–1720.

Mattevelli,L.,T.Ricchiuto,andC.Martinenghi,1992,Deep

isotopiclightmethaneinnorthernItaly,inR.Vially,ed.,Bacterialgas:Paris,EditionsTechnip,p.121–132.

McIntosh,J.C.,L.M.Walter,andA.M.Martini,2002,Pleistocene

rechargetomid-continentbasins:effectsonsalinitystructureandmicrobialgasgeneration:GeochimicaetCosmochimicaActa,v.66,no.10,p.1681–1700.

McLennan,J.D.,P.S.Schafer,andT.J.Pratt,1995,Aguideto

determininggascoalbedcontent:GasResearchInstitute,GRI-94/0396,191p.

Oremland,R.S.,1981,Microbialformationofmethaneinanoxic

estuarinesediments:AppliedEnvironmentalMicrobiology,v.42,p.122–129.

Petsch,S.T.,R.A.Berner,andT.I.Eglinton,2001,Afieldstudyof

thechemicalweatheringofancientsedimentaryorganicmatter:OrganicGeochemistry,v.31,p.475–487.

Price,L.C.,andC.E.Barker,1985,Suppressionofvitrinite

reflectanceinamorphousrichkerogen—amajorunrecog-nizedproblem:JournalofPetroleumGeology,v.8,no.1,p.59–84.

Price,L.C.,andM.Schoell,1995,Constraintsontheoriginsof

hydrocarbongasfromcompositionsofgasesattheirsiteoforigin:Nature,v.378,p.368–371.

Rice,D.D.,1993,Controls,habitat,andresourcepotentialof

ancientbacterialgas,inR.Vially,ed.,Bacterialgas:Paris,EditionsTechnip,p.91–118.

Rooney,M.A.,G.E.Claypool,andH.M.Chung,1995,Modeling

thermogenicgasgenerationusingcarbonisotoperatiosofnaturalgashydrocarbons:ChemicalGeology,v.126,p.219–232.

Rowe,D.,andK.Muehlenbachs,1999,Low-temperaturethermal

1374MicrobialProductionandModificationofGasesinSedimentaryBasins

generationofhydrocarbongasesinshallowshales:Nature,v.398,p.61–63.Rullko¨tterJ.,R.Marzi,andP.A.Meyers,1992,Biologicalmarkers

inPaleozoicsedimentaryrocksandcrudeoilsfromtheMichiganbasin:reassessmentofsourcesandthermalhistoryoforganicmatter,inM.Schidlowski,ed.,Earlyorganicevolution:implicationsformineralandenergyresources:Berlin,Springer-Verlag,p.324–335.

Sassen,R.,I.R.MacDonald,N.L.GiunassoJr.,S.Joye,A.G.

Requejo,S.T.Sweet,J.Alcala-Herrera,D.A.DeFreitas,andD.R.Schink,1998,Bacterialmethaneoxidationinsea-floorgashydrate:significancetolifeinextremeenvironments:Geology,v.26,p.851–854.

Schoell,M.,1980,Thehydrogenandcarbonisotopiccomposition

ofmethanefromnaturalgasesofvariousorigins:GeochimicaetCosmochimicaActa,v.44,p.9–661.

Schoell,M.,1984,Isotopetechniquesfortracingmigrationofgases

insedimentarybasins:JournaloftheGeologicalSocietyofLondon,v.140,p.415–422.

Schoell,M.,K.Muehlenbachs,D.D.Coleman,S.Thibodeaux,L.

Walters,andA.Martini,2001,NaturalsitesofbioconversionofCO2andhydrocarbonsinthesubsurface:SanJuanbasinandMichiganbasin(abs):AAPGAnnualMeetingProgram,v.10,p.A180.

Scott,A.R.,W.R.Kaiser,andW.B.AyersJr.,1994,Thermogenic

andsecondarybiogenicgases,SanJuanbasin,ColoradoandNewMexico—implicationsforcoalbedgasproducibility:AAPGBulletin,v.78,p.1186–1209.

Smith,J.W.,andR.J.Pallasser,1996,MicrobialoriginofAus-traliancoalbedmethane:AAPGBulletin,v.80,p.1–7.Socki,R.A.,H.R.Karlsson,andG.K.EverettJr.,1992,Extraction

techniqueforthedeterminationofoxygen-18inwaterusingpre-evacuationglassvials:AnalyticalChemistry,v.,p.829–831.

Stahl,W.,1977,Carbonandnitrogenisotopesinhydrocarbonre-searchandexploration:ChemicalGeology,v.20,p.121–149.Sugimoto,A.,andE.Wada,1993,Carbonisotopiccompositionof

bacterialmethaneinasoilincubationexperiment:contribu-tionsofacetateandCO2/H2:GeochimicaetCosmochimicaActa,v.57,p.4015–4027.

Sugimoto,A.,andE.Wada,1995,Hydrogenisotopiccomposition

ofbacterialmethane:CO2/H2reductionandacetatefermen-tation:GeochimicaetCosmochimicaActa,v.59,p.1329–1337.

Tang,Y.,J.K.Perry,P.D.Jenden,andM.Schoell,2000,

Mathematicalmodelingofstablecarbonisotoperatiosinnaturalgases:GeochimicaetCosmochimicaActa,v.,p.2673–2687.

Valentine,D.L.,andW.S.Reeburgh,2000,Newperspectiveson

anaerobicmethaneoxidation:EnvironmentalMicrobiology,v.2,p.477–484.

Venneman,T.W.,andJ.R.O’Neil,1993,Asimpleand

inexpensivemethodofhydrogenisotopeandwateranalysesofmineralsandrocksbasedonzincreagent:ChemicalGeology,v.103,p.227–234.

Vugrinovich,R.,1988,Relationshipbetweenregionalhydrogeology

andhydrocarbonoccurrencesinMichigan,U.S.A.:JournalofPetroleumGeology,v.11,p.429–442.

Walter,L.M.,J.M.Budai,L.M.Abriola,C.H.Stearns,A.M.

Martini,andT.C.W.Ku,1996,HydrogeochemistryoftheAntrimShale,northernMichiganbasin:AnnualReport,GasResearchInstitute,GRI-95/0251,173p.

Walter,L.M.,J.M.Budai,A.M.Martini,andT.C.W.Ku,1997,

HydrogeochemistryoftheAntrimShaleintheMichiganbasin:GasResearchInstitute,GRI-97/0127,95p.

Walter,L.M.,J.McIntosh,A.M.Martini,andJ.M.Budai,2001,

HydrogeochemistryoftheNewAlbanyShale:GasResearchInstitute,GRI-00,0158,58p.

Waples,D.W.,1985,Geochemistryinpetroleumexploration:

Boston,IHRDC(InternationalHumanResourcesDevel-opmentCorporation),248p.

Whiticar,M.J.,1999,Carbonandhydrogensystematicsofbacterial

formationandoxidationofmethane:ChemicalGeology,v.161,p.291–314.

Whiticar,M.J.,E.Faber,andM.Schoell,1986,Biogenicmethane

formationinmarineandfreshwaterenvironments:CO2reductionvs.acetatefermentation—isotopicevidence:Geo-chimicaetCosmochimicaActa,v.50,p.693–709.

Zehnder,A.J.B.,andT.D.Brock,1979,Methaneformationand

methaneoxidationbymethanogenicbacteria:JournalofBacteriology,v.137,p.420.

Zinder,S.H.,1993,Physiologicalecologyofmethanogens,inJ.G.

Ferry,ed.,Methanogenesisecology,physiology,biochemistry,andgenetics:London,Chapman&Hall,p.128–206.

Martinietal.1375