MammalfaunalresponsetothePaleogenehyperthermalsETM2AEChew - PDF document

CPD11,1371–1405,2015 MammalfaunalresponsetothePaleogenehyperthermalsETM2A.E.Chew TitlePage Abstract Introduction Conclusions References Tables Figures J I J I Back Close FullScreen/Esc Printer-friendlyVersion InteractiveDiscussion DiscussionPaper|DiscussionPaper|DiscussionPaper|DiscussionPaper| AbstractScientistsareincreasinglyturningtodeep-timefossilrecordstodecipherthelong-termconsequencesofclimatechangeintheracetopreservemodernbiotasfromanthro-pogenicallydrivenglobalwarming.“Hyperthermals”arepastintervalsofgeologicallyrapidglobalwarmingthatprovidetheopportunitytostudytheeectsofclimatechange5onexistingfaunasoverthousandsofyears.AserieshyperthermalsisknownfromtheearlyEocene(56–54millionyearsago),includingthePaleocene-EoceneThermalMaximum(PETM)andtwosubsequenthyperthermals,EoceneThermalMaximum2(ETM2)andH2.ThelaterhyperthermalsoccurredfollowingtheonsetofwarmingattheEarlyEoceneClimaticOptimum(EECO),thehottestsustainedperiodoftheCenozoic.10ThePETMhasbeencomprehensivelystudiedinmarinea

ndterrestrialsettings,buttheterrestrialbioticeectsofETM2andH2areunknown.Theirgeochemicalsigna-tureshavebeenlocatedinthenorthernpartoftheBighornBasin,WY,USA,andtheirlevelscanbeextrapolatedtoanextraordinarilydense,well-studiedterrestrialmammalfossilrecordinthesouth-centralpartofthebasin.High-resolution,multi-parameterpa-15leoecologicalanalysisrevealssignicantpeaksinspeciesdiversityandturnoverandchangesinabundanceandrelativebodysizeatthelevelsofETM2andH2inthesouth-centralBighornBasinrecord.IncontrastwiththePETM,faunalchangeatthelaterhyperthermalsislessextreme,doesnotincludeimmigrationandinvolvesapro-liferationofbodysizes,althoughabundanceshiftstendtofavorsmallercongeners.20FaunalresponseatETM2andH2isdistinctiveinitshighproportionofspecieslossespotentiallyrelatedtoheightenedspeciesvulnerabilityinresponsetothechangesal-readyunderwayatthebeginningoftheEECO.FaunalresponseatETM2andH2isalsodistinctiveinhighproportionsofbetarichness,suggestiveofincreasedgeographicdispersalrelatedtotransientincreasesinhabitat(

;oral)complexityand/orprecipitation25orseasonalityofprecipitation.Theseresultssuggestthatrapidecologicalchanges,in-creasedheterogeneityinspeciesincidence,andheightenedspeciesvulnerabilityand1372 CPD11,1371–1405,2015 MammalfaunalresponsetothePaleogenehyperthermalsETM2A.E.Chew TitlePage Abstract Introduction Conclusions References Tables Figures J I J I Back Close FullScreen/Esc Printer-friendlyVersion InteractiveDiscussion DiscussionPaper|DiscussionPaper|DiscussionPaper|DiscussionPaper| lossmaybeexpectedacrossmostofNorthAmericainthenearfutureinresponsetoanthropogenically-drivenclimatechange.1IntroductionContemporaryscienticprioritiesincludethestudyofpastgeobiologicalsystemstopredictEarthsystemresponsetoclimateforcing(NationalResearchCouncil,2011).5TheearlyEocene(56–52Ma)isparticularlyrelevantforunderstandingmodernan-thropogenicwarmingasitwitnessedglobaltemperatureuctuationincludingseveralhyperthermals(intervalsofgeologicallyrapidglobalwarming)intheapproachtotheEarlyEoceneClimaticOptimum(EE

CO),thehottestsustainedperiodoftheCenozoic(53–51Ma,Zachosetal.,2008).Thelargestandbestknownofthehyperthermalsis10thePaleocene-EoceneThermalMaximum(PETM)atthebaseoftheEocene(KennettandStott,1991;Zachosetal.,1993).Excursionsinmultiplecarbonisotoperecords(carbonisotopeexcursions,CIEs)atthePETMindicatethatseveralthousandpeta-gramsofreducedcarbonwerereleasedintotheocean–atmospheresystemin20ka(reviewinMcInerneyandWing,2011).Thisinitiateda100kaperiodofelevated15globaltemperature(5–7Cwarmer)andperturbationsinEarth'scarboncycling,oceanchemistryandplanktoncommunities(Bowenetal.,2006;Gingerich,2006;McInerneyandWing,2011).Onland,bioticresponsetothePETMisbestknownfromthefossilrecordoftheBighornBasininnorthwesternWyoming,whichdocumentsmajorintra-andintercontinentalimmigration,widespreadtemporarydwarng,andchangesinthe20diversity,trophicstructureandphysiologyoforasandfaunas(Curranoetal.,2008;Gingerich,1989;GingerichandSmith,2006;Roseetal.,2012;Secordetal.,2012;Smithetal.,2009;Wingetal.,2005;Yan

setal.,2006).ThePETMhasbeendescribedasthebestdeep-timeanalogueforanthropogenicclimatewarming(Bowenetal.,2006;Gingerich,2006;McInerneyandWing,2011).25Amajoradvantageofdeep-timerecordsisthepotentialfordocumentationofmulti-pleevents,providingtheopportunitytocharacterizefaunalresponsetoclimatechange1373 CPD11,1371–1405,2015 MammalfaunalresponsetothePaleogenehyperthermalsETM2A.E.Chew TitlePage Abstract Introduction Conclusions References Tables Figures J I J I Back Close FullScreen/Esc Printer-friendlyVersion InteractiveDiscussion DiscussionPaper|DiscussionPaper|DiscussionPaper|DiscussionPaper| ofvaryingrateandmagnitudeagainstdierentbackgroundconditions.Consistenciesinfaunalrepsonseunderspecicconditionsstrengthenthecaseforcausalityandcanbeusedforpredictivepurposes.TwoadditionalearlyEocenehyperthermals,EoceneThermalMaximum2(ETM2=H1)andH2(Crameretal.,2003;Lourensetal.,2005),occurred2maafterthePETM,constitutingwhatiseectivelyasetofrepeatednat-5uralexperimentsinclimatechange.TheCIEsofETM2andH2are

similarbutonehalftoonethirdthemagnitudeofthePETMCIE(Lourensetal.,2005;Sextonetal.,2011;Stapetal.,2010).TheyoccurredwhentheEarthwaswarmerandmayhavepushedhigh-latitudetemperaturestogreaterextremesthanthePETM(Sluijsetal.,2009).ChangesinplanktonatETM2andH2weresimilartothoseatthePETMwith10thedegreeofresponseproportionatetothemagnitudeoftheCIEs(Fosteretal.,2013;Gibbsetal.,2012;Sluijsetal.,2009;Stassenetal.,2012).However,instarkcontrastwiththewell-studiedPETM,terrestrialbioticresponsetoETM2andH2iscurrentlyunknown.TheETM2andH2CIEshavebeendocumentedinthenorthernpartoftheBighornBasin(Abelsetal.,2012)andfromoneotherterrestrialsequenceinIndia15(Clementzetal.,2011),butneitherrecordincludessucientfossilstopermittestingoffaunalresponse.Thedense,highly-resolved,well-documentedmammalrecordfromtheFifteenmileCreek(FC)inthesouth-centralpartoftheBighornBasin(Fig.1)chroniclesalmosttheentireearlyEocenefromthePETMtotheEECO(Bownetal.,1994b).Thelargest20sampleofPETMmammalshasbeenstudiedanddescribedfromtheFCrecord(Roseetal.,2012)alo

ngwithotherfaunaleventsor“biohorizons”,thelargestofwhichafterthePETMisBiohorizonB(Chew,2009a;Schankler,1980).BiohorizonBmarksamajorturningpointinfaunaldiversity(ChewandOheim,2013)thathasbeencorrelatedwithpaleoecologicalchangeacrossNorthAmericaattributedtotheonsetofwarmingat25theEECO(Woodburneetal.,2009).InthenorthernBighornBasinisotoperecord,theCIEsofETM2andH2occur60–80kaafterbiostratigraphiceventsatthebeginningofBiohorizonB(Abelsetal.,2012)butthereisnoobviouslycorrelatedfaunalchangeafterBiohorizonBintheFCrecord(Chew,2009a,b;ChewandOheim,2009).Thislack1374 CPD11,1371–1405,2015 MammalfaunalresponsetothePaleogenehyperthermalsETM2A.E.Chew TitlePage Abstract Introduction Conclusions References Tables Figures J I J I Back Close FullScreen/Esc Printer-friendlyVersion InteractiveDiscussion DiscussionPaper|DiscussionPaper|DiscussionPaper|DiscussionPaper| wasinterpretedasbioticinsensitivitytoETM2andH2(Abelsetal.,2012).However,nopreviousanalysisoftheFCrecordachievedsucientresolutiontodetectf

aunalperturbationatthescaleofthehyperthermals(40ka).Thisreportdescribesthersthigh-resolution,multi-parameterpaleoecologicalanalysisoftheexceptionalFCrecordtotestmammalfaunalresponsetoETM2andH2.52Methodsandmaterials2.1CollectionsThelowangleofdipandwideareaofexposurealongtheFifteenmileCreek(FC)inthesouth-centralpartoftheBighornBasin(Fig.1)allowsWillwoodFormation(earlyEocene)fossillocalitiestobetiedbymeterleveltoacompositestratigraphicsectionof10700m(Bownetal.,1994b).ThebaseoftheFCsection(0m)restsonadistinctiveredbedthatmarksthebeginningofthePETMatSandCreekDivideontheeasternedgeofthestudyarea(Roseetal.,2012).TheC24r-C24ngeomagneticpolarityshifthasbeenlocatednearthemiddleofthesection(455m,Clydeetal.,2007).Nearthetopofthesection(634m),the40Ar=39Ardateofavolcanicashindicatesthattheupperlevels15arewithintheEECO(Smithetal.,2004).Numericalages(56.33,53.57,and52.9Ma,respectively)areassignedtothesethreetiepointsfollowingtherecentregionalrecal-ibrationofTsukuiandClyde(2012).Averagesedimentaccumulationrat

esbetweenthetiepointsincreasefrom0.165to0.267mka�1abovetheC24r-C24ngeomagneticpolarityshift,whichisinbroadagreementwithpreviousanalysisofdepositionalrates20basedonpaleosols(BownandKraus,1993).TheseratessuggestthatonemeterofFCsectionthicknessrepresents6kainthelowerlevelsand4kaabovetheC24r-C24ngeomagneticpolarityshift.Previousanalysisofpaleosolcarbonateswasnotsucientlyresolvedtodemon-stratetheCIEsofETM2andH2intheFCsection(Fig.2,Kochetal.,2003)buttheir25levelscanbeextrapolatedfromisotopicworkintheMcCulloughPeaksofthenorth-1375 CPD11,1371–1405,2015 MammalfaunalresponsetothePaleogenehyperthermalsETM2A.E.Chew TitlePage Abstract Introduction Conclusions References Tables Figures J I J I Back Close FullScreen/Esc Printer-friendlyVersion InteractiveDiscussion DiscussionPaper|DiscussionPaper|DiscussionPaper|DiscussionPaper| ernBighornBasin.Abelsetal.(2012)identiedtheCIEsofETM2andH2withinanintervalofmixedgeomagneticpolaritybelowtheshiftfromtheC24reversedtoC24normalgeomagneticzones(Fig.2).Biostra

tigraphiceventsatthebeginningofBiohori-zonBarealsolooselytiedtotheMcCulloughPeaksisotopesections,includingthelastappearanceofthecondylarthHaplomylusspeirianusandtherstappearanceof5theartiodactylBunophorusetsagicus.Thesespeciesco-occuratasinglelocality(MP122,5kmwestofthenearestisotopesection)thatwastracedtonearthemiddleofa35mthickgapbetweenthemintheisotopesections(Fig.2).TheC24r-C24nge-omagneticshiftandthenearlysimultaneousBiohorizonBstratigraphiceventsbrackettheETM2andH2CIEsandarealsoknownat455m(Clydeetal.,2007)and381m10(thisproject)intheFCsection.Betweenthesetiepoints,theMcCulloughPeakssedi-mentsareroughly42%thickerthantheFCsediments.ScalingtheMcCulloughPeakssectionsby0.68allowstheextrapolationofETM2andH2tothe410–420and430–440mlevels,respectively,oftheFCsection.TheseareroughpredictionsduetotheuncertaintyassociatedwiththelevelofthebiostratigraphiceventsintheMcCullough15Peaksandtovariationinsedimentaccumulationratesovertime,especiallyaroundBiohorizonB(BownandKraus,1993;Clyde,2001

).Allspecimensincludedinthisprojectwerecollectedfrom410fossillocalitiesspan-ning290–510mintheFCsection.NeighboringlocalitiesalongtheElkCreek(Fig.1)havealsobeentiedtotheFCsectionbutareexcludedfromthisanalysisbecauseof20dierencesinsectionthickness(upto70m,Bownetal.,1994b)thatwouldcompromiseresolution.Thisexclusionresultsincomparativelylimitedsamplesizesbelow370m(Fig.2).Morethan32000specimensareincludedinthisstudy(TableS1),represent-ing103lineagesandspecies(TableS2,68genera,27families,16orders).Ofthese,�1100arerecentlycollectedspecimens(2004–2011eldseasons)notincludedin25previouspaleoecologicalanalyses(Chew,2009a,b;ChewandOheim,2009,2013).Specimensareidentiedtospecieslevel(asinChew,2009a).Singletontaxaandstratigraphicoutliersareexcludedtoavoidinationofpaleoecologicalparametersandlossofresolution.Specieswithsingleoccurrencesinthisdatasetthatarenotexcluded1376 CPD11,1371–1405,2015 MammalfaunalresponsetothePaleogenehyperthermalsETM2A.E.Chew TitlePage Abstract Introduction C

onclusions References Tables Figures J I J I Back Close FullScreen/Esc Printer-friendlyVersion InteractiveDiscussion DiscussionPaper|DiscussionPaper|DiscussionPaper|DiscussionPaper| (TableS2)areknowntohaveexistedbelow290mand/orabove510m.Astratigraphicoutlierisdenedasawell-documented,clearlyidentiableindividualrecovered50-100moutsideofthestratigraphicrangeofthespecies.Sevenstratigraphicoutlierswereidentiedandexcluded(Anacodonursidens–Condylarthra,Apatemysrodens–Apatotheria,BunophorusetsagicusandBunophorusgrangeri–Artiodactyla,Lamb-5dotherium–Perissodactyla,Pachyaenaossifraga–Mesonychia,Palaeictopsbicuspis–Leptictida).2.2SpecimendatabinningThespecimendataarebinnedbymeterlevel,providingthemaximumpossibleresolu-tion(4–6ka).Atthisresolution,stratigraphicgapsconstitute40%oftherecord10andtherearelargedisparitiesinsamplesize(0–3000specimensm�1)andatrendofincreasingsamplesizeovertime(Spearman's=0.19,p0.05),allofwhichcompli-catethec

alculationandinterpretationofpaleoecologicalparameters.Althoughlongerdatabinsdecreaseresolution,theyeliminategapsandallowextensivesamplesizestandardization,permittingthecalculationofmultiple,complimentaryandunbiasedpa-15leoecologicalparameters.FivemetersistheminimumbinthicknessthateliminatesallgapsinthezoneinwhichETM2andH2mustoccur(370–455m)intheFCsection.However,eachve-meterbinrepresents30ka,whichapproachesthelengthofthehyperthermalsunderinvestigationandmakesitimpossibletoconstructasinglebinningseriesthatdividesthesectionappropriatelytocaptureeachevent.Onealternativeis20toapproximatemeter-levelresolutionthroughthecombinationofaseriesofrandomlyoverlappingbinsofdierentlengths.Fourseriesofequal-timedatabinsarecreatedthroughanexhaustivesearchtoeliminategapsandmaximizesamplesizesatve-,six-,seven-andeight-meterbinlengths(TableS3).(Toaccommodateincreasingsedi-mentaccumulationrateabove455m,thebinsineachseriesarelengthenedaccord-25ingly;5–7,6–8,7–10,and8–11m).Collectively

,thebinningseriesprovidecontinuouscoverageandsamplesizes&#x]TJ/;&#x 10.;邑&#x Tf ;.4;# 0;&#x Td ;&#x[000;100specimensfrom376–505mintheFCsection.Pale-oecologicalparametersarecalculatedforeachseries.Parametervaluesareassigned1377 CPD11,1371–1405,2015 MammalfaunalresponsetothePaleogenehyperthermalsETM2A.E.Chew TitlePage Abstract Introduction Conclusions References Tables Figures J I J I Back Close FullScreen/Esc Printer-friendlyVersion InteractiveDiscussion DiscussionPaper|DiscussionPaper|DiscussionPaper|DiscussionPaper| thestandardizedappearancesandrelativeabundancesprovidedbythealgorithmictreatmentofthebinneddata.2.3.1RichnessRichnessisthenumberofspeciespresentinasampleandishighlydependentonsamplesize.Wheresamplesallow(�100specimens,continuouslydistributed),rar-5efactionisusedtoproducestandardizedestimatesofrichness(Colwell,2013;Hol-land,2003).Alpha(average,within-sample)richnessisestimatedusingconventional,individual-basedrarefaction(IR,Fig.3),whichplotsthenumberofspeciesfoundthrou

ghtheaccumulationofindividuals(Hurlburt,1971;Sanders,1968).Pointesti-matesofalpharichnessatasamplesizeof100specimensaredirectlycomparable10betweensamples.Toestimatebeta(dierentiationbetweensample)richness,sample-basedrarefaction(SR,Fig.3)isused,whichplotsthenumberofspeciesthatarefoundthroughtheaccumulationofsamples(Chiaruccietal.,2008;Colwelletal.,2004;GotelliandColwell,2001).SRisdependentuponthespatialdistributionofspecimens.Inthepresenceofdistributionalheterogeneity,SRrichnessestimatesarelowerthanIR15richnessestimates,asIRassumesarandomdistributionofindividualsandproducesacurveofmaximal,theoreticalrichness(Colwelletal.,2004;Foote,1992;GotelliandColwell,2001;Olszewski,2004).ThedierencebetweenIRandSRcurvesreectsbetarichness(CristandVeech,2006;GotelliandColwell,2001;Olszewski,2004)andislargestnearthebaseoftheSRcurve(Fig.3).ComparableIRandSRpointrichness20estimatesfromthebaseofeachSRcurveareusedtoestimatebetarichness(asinChewandOheim,2013).Gamma(totallandscape)richnessisthesumofalphaandbetarichness.

2.3.2EvennessAspectsofevennessareindependentofsamplesize,butevennessisdiculttochar-25acterize(Magurran,2004).Twoindicesareusedhere,bothcalculatedfromstandard-1379 CPD11,1371–1405,2015 MammalfaunalresponsetothePaleogenehyperthermalsETM2A.E.Chew TitlePage Abstract Introduction Conclusions References Tables Figures J I J I Back Close FullScreen/Esc Printer-friendlyVersion InteractiveDiscussion DiscussionPaper|DiscussionPaper|DiscussionPaper|DiscussionPaper| izedproprotionalrelativeabundances.Therstisthewell-knownProbabilityofInter-specicEncounter,PIE,index(Hurlburt,1971),whichistheinverseofSimpson'sdom-inanceindexstandardizedfornitecollectionsize.PIE=1�"sXi=1ni(ni�1)=N(N�1)#,(1)whereniisthenumberofspecimensofspecies“i”andNisthetotalnumberof5specimensinasample.Thoughwidelyemployedasadescriptorofthe“evenness”ofspeciesabundancedistributions,PIEisstronglycorrelatedwiththeproportionalrel-ativeabundanceofthetwomostcommonspeciesinthesedata(mainlyequidandhyopsodontidspeci

es;Spearman's=�0.49to�0.84,p=0.00).Toavoidconfusion,itisreferredtohereasanindexof“inversedominance”.Thesecondindexisamod-10icationofrank-abundanceanalysis(Fig.3),inwhichspeciesarerankedfrommosttoleastabundantandtheirnatural-logtransformedrelativeabundancesareplottedagainstranks(Magurran,2004).Rank-abundancecurvesprovideavisualrepresenta-tionofanabundancedistributionthatisshapedbythemajorityofthespeciespresentinasample.Theslopesofexponentialtrendlinesttedtothecurvesaredirectlycompa-15rablebetweensamples(Fig.3,asinCaronandJackson,2008).Thetofthetrendlinesforthesedataishigh(R2�0.75)andtheslopesofthetrendlinesareshallowandneg-ative(�0.1).Thereciprocaloftheabsolutevalueoftheslopesisusedtotransformthemintoanindexof“inclusiveabundance”.Thetwoindiceshavevaluesbetweenzeroandone.Highervaluesofinversedominanceindicatehigherevennessthrough20decreaseddominanceofthesamplebyafewtaxa.Highervaluesofinclusiveabun-danceindicatehigherevennessthroughamoreequaldistributionoftheabu

ndancesofthemajorityofthespeciesinthesample.Thetwoindicesaresummedasanindexofevenness.1380 CPD11,1371–1405,2015 MammalfaunalresponsetothePaleogenehyperthermalsETM2A.E.Chew TitlePage Abstract Introduction Conclusions References Tables Figures J I J I Back Close FullScreen/Esc Printer-friendlyVersion InteractiveDiscussion DiscussionPaper|DiscussionPaper|DiscussionPaper|DiscussionPaper| andaresimilarinallaspectsoffaunalchangedescribedhere.Thesimplestexplana-tionfortheirsimilarityisacomparabletrigger,andETM2andH2areakin(Abelsetal.,2012;Sextonetal.,2011;Stapetal.,2010).ChangeatfaunaleventsB-1andB-2issuperciallysimilartothatdescribedattheonlyotherwell-knownearlyEocenehyper-thermal,thePETM(Gingerich,1989;Roseetal.,2012;Secordetal.,2012),including5increasesindiversityandturnoverandageneralshifttowardssmallerbodysize.Inaddition,theincreasesin(alpha)richnessandturnoverarelesspronouncedatfaunaleventsB-1andB-2thanatthePETM(Table2),whichisalsothecaseinmarineplank-tonacrossthehyperthermals(Fosteretal.,2013;Gibbsetal

.,2012;Stassenetal.,2012)andconformswiththeexpectationthatETM2andH2weresmallerevents.Itis10assumedherethatthereisacausalrelationshipbetweenETM2andH2andfaunaleventsB-1andB-2.4.1ComparisonwiththePETMTurnoverandchangesinbodysizeatthePETMaredramaticcomparedwithfau-naleventsB-1andB-2(Table2).PronouncedturnoveratthePETMwasrecognized15longbeforethehyperthermalwasknownbytheplacementoftherstmajorboundary(Clarkforkian/Wasatchian)intheNorthAmericanLandMammalAgesequence(Wood,1941).NearlyhalfoftheBighornBasinmammalgeneraand80%ofthespeciesthatexistedduringthePETMarenew(Roseetal.,2012;Woodburneetal.,2009).Thisturnoverwasfueledbyimmigration;upto40%ofnewgeneraatthePETM20wereimmigrants(Roseetal.,2012;Woodburneetal.,2009)fromthesouthernpartofthecontinent(Burger,2012;Gingerich,2001)andfromtheHolarcticcontinentsvianorthernlandbridges(Bowenetal.,2002;Gingerich,2006;Roseetal.,2011).Incom-parison,10%ofgeneraatfaunaleventsB-1andB-2arenew(Table2)andnoneofthesearedocumentedimmigrants(Woodburneetal.,2009).Decreasesinbodysize

25atthePETMarewidespread(e.g.,Smithetal.,2009)including40%ofallmammalgenera(Secordetal.,2012).Thesedecreasesoccurredthroughtemporarydwarngoflineagesandspeciesviametaboliceects,orthroughtheimmigrationofclosely1384 CPD11,1371–1405,2015 MammalfaunalresponsetothePaleogenehyperthermalsETM2A.E.Chew TitlePage Abstract Introduction Conclusions References Tables Figures J I J I Back Close FullScreen/Esc Printer-friendlyVersion InteractiveDiscussion DiscussionPaper|DiscussionPaper|DiscussionPaper|DiscussionPaper| abundanceofdominantspecies,seeChew,2009a,b;ChewandOheim,2013)andcarriedthroughfaunaleventsB-1andB-2.However,faunaleventsB-1andB-2areuniqueintheirhigherproportionsofspeciesloss,whichnearlyequaltheproportionsofnewspeciesateachevent(Table2).NearlyhalfoftheturnoveratfaunaleventsB-1andB-2occurswithinlineages,withcorrespondinglysmallproportions(6%)of5genericevents.Incontrast,andinspiteoftheirwidelydierentmechanisms,boththePETMandBiohorizonBarecharacterizedbyburstsofnewspecies,includingmany

newgenera,andcomparativelyfewlosses.ThePETMwasatransientepisodeofeco-logicalchange,includingimmigrationandbodysizeadjustment,whereasBiohorizonBinvolvedmarkedevolutionarychange(Woodburneetal.,2009).Botheventswere10initiatedbysignicantclimaticandenvironmentaldisturbancethatended1maperi-odsofrelativelystaticconditions;warmandmoistbeforethePETMandcoolanddrybeforeBiohorizonB(Krausetal.,2013;KrausandRiggins,2007;Snelletal.,2013;Wilf,2000;Wingetal.,2000).Incontrast,therapidwarmingofETM2andH2occurredsoonaftertheonsetoftheclimaticandenvironmentaldisturbancerelatedtotheEECO15andBiohorizonB.Faunalstructuremayhavebeencomparativelyunstableascommu-nitieswereadjustingtochangingconditions,perhapsleavingmorespeciesvulnerabletofurtherchange.TheturnoverwithinlineagesatfaunaleventsB-1andB-2suggeststhatmorespecieswerelostthroughevolutionarytransitionsatETM2andH2.FaunaleventsB-1andB-2arealsouniqueintheirhighproportionsofbetarichness20(Table2).Betarichnessreectsheterogeneityinspeciesincidencepatterns(CollinsandSimberlo&#

19;,2009;Colwelletal.,2004;GotelliandColwell,2001)thatmayoccurthroughspecializationfordispersedmicrohabitatsand/orheightenedecologicalinter-actionsthatpromptspeciestoseekoutoravoideachother(e.g.,competition,preda-tion).Previousanalysisidentiedariseinbetarichnessinthe2maafterBiohorizon25Btowhichbothmechanismsmayhavecontributed(Fig.3,ChewandOheim,2013;Woodburneetal.,2009).Acoincidentlong-termincreaseinalpharichness(Fig.3,ChewandOheim,2013;Woodburneetal.,2009)impliesthatthereweremorespeciespackedintotheavailablespaceofthelandscape,increasingthepotentialforecologi-1387 CPD11,1371–1405,2015 MammalfaunalresponsetothePaleogenehyperthermalsETM2A.E.Chew TitlePage Abstract Introduction Conclusions References Tables Figures J I J I Back Close FullScreen/Esc Printer-friendlyVersion InteractiveDiscussion DiscussionPaper|DiscussionPaper|DiscussionPaper|DiscussionPaper| calinteractions.Increasedhabitatcomplexityassubtropicalandtropicalorasbecamemoreestablishedimpliesmoreopportunitiesformicrohabitatspecializa

tion(ChewandOheim,2013;Wingetal.,2000;Woodburneetal.,2009).TheproportionsofnewspeciesandalpharichnessarenotparticularlyhighatfaunaleventsB-1andB-2,sug-gestingthatthetemporarypeaksinbetarichnessatETM2andH2areprobablynot5relatedtoanaccelerationofspeciespackingandheightenedecologicalinteractions.Instead,theymayrepresentincreasedmicrohabitatspecializationinresponsetotran-sientincreasesoralcomplexity,perhapsheightenedbythemoreseasonal,possiblymoreintenseandepisodic,precipitationsuggestedbytransientlithologicalchanges(Abelsetal.,2012).104.3ImplicationsformodernanthropogenicchangeAspectsoffaunalchangeintheBighornBasinrecordoftheearlyEocenearerele-vantforpredictingmodernanthropogeniceects.ThePETM,ETM2andH2raisedMATintheBighornBasintonearlythesameabsolutevalue(20Cgiventhepro-portionalityofCIEandtemperature,andlong-termtemperaturetrendsatETM2and15H2,Abelsetal.,2012;FrickeandWing,2004;Wingetal.,2000).Extrapolatingfromcurrentandprojectedregionalratesofchange,Wyoming'sMAT(8CaccordingtoUSclimated

ata)willapproachthisvaluein300yearsevenifemissionsarestabi-lizedbeforethen,giventhetimescaleofclimateprocessesandfeedbacks(PachauriandReisinger,2007).Thisrateofwarmingfarexceedsthoseofthepast,implyingthat20species-specic,rapidecologicaladjustments(e.g.,geographicrangeandbodysizechanges)willprobablyoccurinthenearfutureastheydidatthePETM,theintervalwiththehighestrateofwarming.RiverrunoandwateravailabilityareexpectedtodecreaseinthedryareasofwesternNorthAmericawithongoingclimatechangebutprecipitia-tionandthefrequencyofheavyprecipitationeventsareexpectedtoincreaseacross25therestofthecontinentwiththecontractionoftheGreenlandicesheet(PachauriandReisinger,2007).ThelatterchangesaremoreconsistentwiththeBighornBasinrecordofthebeginningoftheEECO.Inaddition,humanactivitiessuchasurbanization,habi-1388 CPD11,1371–1405,2015 MammalfaunalresponsetothePaleogenehyperthermalsETM2A.E.Chew TitlePage Abstract Introduction Conclusions References Tables Figures J I J I Back Close FullScreen/Esc Printer-friendlyVersion

InteractiveDiscussion DiscussionPaper|DiscussionPaper|DiscussionPaper|DiscussionPaper| Clementz,M.,Bajpai,S.,Ravikant,V.,Thewissen,J.G.M.,Saravanan,N.,Singh,I.B.,andPrasad,V.:EarlyEocenewarmingeventsandthetimingofterrestrialfaunalexchangebe-tweenIndiaandAsia,Geology,39,15–18,2011.Clyde,W.C.:MammalianbiostratigraphyoftheMcCulloughPeaksareainthenorthernBighornBasin,in:Paleocene-EoceneStratigraphyandBioticChangeintheBighornAndClarksFork5Basins,Wyoming,editedby:Gingerich,P.D.,UniversityofMichiganPapersonPaleontologyVolume33,AnnArbor,Michigan,109–126,2001.Clyde,W.C.,Hamzi,W.,Finarelli,J.A.,Wing,S.L.,Schankler,D.,andChew,A.:Basin-widemagnetostratigraphicframeworkforthebighornbasin,Wyoming,Geol.Soc.Am.Bull.,119,848–859,2007.10Collins,M.D.andSimberlo,D.:Rarefactionandnonrandomspatialdispersionpatterns,Envi-ron.Ecol.Stat.,16,89–103,2009.Collinson,M.E.,Hooker,J.J.,andGrocke,D.R.:Cobhamlignitebedandpenecontemporane-ousmacroorasofsouthernEngland:arecordofvegetationandreacrossthePaleocene-E

oceneThermalMaximum,in:CausesandConsequencesofGloballyWarmClimatesin15theEarlyPaleogene,editedby:Wing,S.L.,Gingerich,P.D.,Schmidtz,B.,andThomas,E.,Geol.Soc.Am.SpecialPaper369,Boulder,CO,333–350,2003.Colwell,R.K.:EstimateS9.1:StatisticalEstimationofSpeciesRichnessandSharedSpeciesfromSamples(SoftwareandUser'sGuide),availableat:http://purl.oclc.org/estimates,2013.Colwell,R.K.,Mao,C.X.,andChang,J.:Interpolating,extrapolating,andcomparingincidence-20basedspeciesaccumulationcurves,Ecology,85,2717–2727,2004.Cramer,B.S.,Wright,J.D.,Kent,D.V.,andAubry,M.-P.:Orbitalclimateforcingofdelta13CexcursionsinthelatePaleocene-earlyEocene(chronsC24n-C25n),Paleoceanography,18,21-1–21-25,2003.Crist,T.O.andVeech,J.A.:Additivepartitioningofrarefactioncurvesandspecies–areare-25lationships:unifyingalpha-,beta-andgammadiversitywithsamplesizeandhabitatarea,Ecol.Lett.,9,923–932,2006.Currano,E.D.,Wilf,P.,Wing,S.L.,Labandeira,C.C.,Lovelock,E.C.,andRoyer,D.L.:SharplyincreasedinsectherbivoryduringthePaleocene-Eocenethe

rmalmaximum,P.Natl.Acad.Sci.USA,105,1960–1964,2008.30Foden,W.,Midgley,G.F.,Hughes,G.,Bond,W.J.,Thuiller,W.,Homan,M.T.,Kaleme,P.,Underhill,L.G.,Rebelo,A.,andHannah,L.:Achangingclimateiserodingthegeographical1392 CPD11,1371–1405,2015 MammalfaunalresponsetothePaleogenehyperthermalsETM2A.E.Chew TitlePage Abstract Introduction Conclusions References Tables Figures J I J I Back Close FullScreen/Esc Printer-friendlyVersion InteractiveDiscussion DiscussionPaper|DiscussionPaper|DiscussionPaper|DiscussionPaper| rangeoftheNamibDeserttreeAloethroughpopulationdeclinesanddispersallags,Divers.Distrib.,13,645–653,2007.Foote,M.:Rarefactionanalysisofmorphologicalandtaxonomicdiversity,Paleobiology,18,1–16,1992.Foote,M.:Originationandextinctioncomponentsoftaxonomicdiversity:generalproblems,5Paleobiology,26(Supplementtono.4),74–102,2000.Foreman,B.Z.:Climate-drivengenerationofauvialsheetsandbodyatthePaleocene-Eoceneboundaryinnorth-westWyoming(USA),BasinRes.,26,225–241,2014.Foreman,B.Z.,Heller,P

.L.,andClementz,M.T.:FluvialresponsetoabruptglobalwarmingatthePalaeocene/Eoceneboundary,Nature,491,92–95,2012.10Foster,L.C.,Schmidt,D.N.,Thomas,E.,Arndt,S.,andRidgwell,A.:SurvivingrapidclimatechangeinthedeepseaduringthePaleogenehyperthermals,P.Natl.Acad.Sci.USA,110,9273–9276,2013.Fricke,H.C.andWing,S.L.:Oxygenisotopeandpaleobotanicalestimatesoftemperatureanddelta(18)O-latitudegradientsoverNorthAmericaduringtheEarlyEocene,Am.J.Sci.,304,15612–635,2004.Gibbs,S.J.,Bown,P.R.,Murphy,B.H.,Sluijs,A.,Edgar,K.M.,Pälike,H.,Bolton,C.T.,andZachos,J.C.:ScaledbioticdisruptionduringearlyEoceneglobalwarmingevents,Biogeo-sciences,9,4679–4688,doi:10.5194/bg-9-4679-2012,2012.Gingerich,P.D.:NewearliestWasatchianmammalianfaunafromtheEoceneofNorthwestern20Wyoming:compositionanddiversityinararelysampledhigh-oodplainassemblage,Univer-sityofMichiganPapersonPaleontology,AnnArbor,Michigan,1989.Gingerich,P.D.:BiostratigraphyofthecontinentalPaleocene-EoceneboundaryintervalonPolecatbenchinthenorthernBighornBasin,in:P

aleocene-EoceneStratigraphyandBioticChangeintheBighornAndClarksForkBasins,Wyoming,editedby:Gingerich,P.D.,Uni-25versityofMichiganPapersonPaleontologyVolume33,AnnArbor,Michigan,37–71,2001.Gingerich,P.D.:EnvironmentandevolutionthroughthePaleocene-EoceneThermalMaximum,TrendsEcol.Evol.,21,246–253,2006.Gingerich,P.D.andSmith,T.:Paleocene-EocenelandmammalsfromthreenewlatestClark-forkianandearliestWasatchianwashsitesatPolecatBenchinthenorthernBighornBasin,30Wyoming,ContributionsfromtheMuseumofPaleontology,UniversityofMichigan,31,245–303,2006.1393 CPD11,1371–1405,2015 MammalfaunalresponsetothePaleogenehyperthermalsETM2A.E.Chew TitlePage Abstract Introduction Conclusions References Tables Figures J I J I Back Close FullScreen/Esc Printer-friendlyVersion InteractiveDiscussion DiscussionPaper|DiscussionPaper|DiscussionPaper|DiscussionPaper| Gotelli,N.J.andColwell,R.K.:Quantifyingbiodiversity:proceduresandpitfallsinthemea-surementandcomparisonofspeciesrichness,Ecol.Lett.,4,379–391,2001.Holland,S.:A

nalyticRarefaction1.3(SoftwareandUser'sGuide),UniversityofGeorgia,Athens,GA,2003.Hurlburt,S.H.:Thenonconceptofspeciesdiversity:acritiqueandalternativeparameters,5Ecology,52,577–586,1971.Kennett,J.P.andStott,L.D.:Abruptdeep-seawarming,palaeoceanographicchangesandbenthicextinctionsattheendofthePaleocene,Nature,353,225–229,1991.Koch,P.L.,Clyde,W.C.,Hepple,R.P.,Fogel,M.L.,Wing,S.L.,andZachos,J.C.:CarbonandoxygenisotoperecordsfrompaleosolsspanningthePaleocene-Eoceneboundary,Bighorn10Basin,Wyoming,in:CausesandConsequencesofGloballyWarmClimatesintheEarlyPaleogene,editedby:Wing,S.L.,Gingerich,P.D.,Schmidtz,B.,andThomas,E.,Geol.Soc.Am.SpecialPaper369,Boulder,CO,49–64,2003.Kraus,M.J.andRiggins,S.:TransientdryingduringthePaleocene-EoceneThermalMaximum(PETM):analysisofpaleosolsintheBighornBasin,Wyoming,Palaeogeogr.Palaeocl.,245,15444–461,2007.Kraus,M.J.,McInerney,F.A.,Wing,S.L.,Secord,R.,Baczynski,A.A.,andBloch,J.I.:PaleohydrologicresponsetocontinentalwarmingduringthePaleocene-EoceneThermalMaximum,Bigho

rnBasin,Wyoming,Palaeogeogr.Palaeocl.,370,196–208,2013.Lourens,L.J.,Sluijs,A.,Kroon,D.,Zachos,J.C.,Thomas,E.,Rohl,U.,Bowles,J.,andRa,I.:20AstronomicalpacingoflatePalaeocenetoearlyEoceneglobalwarmingevents,Nature,435,1083–1087,2005.Magurran,A.E.:MeasuringBiologicalDiversity,Wiley-Blackwell,Indianapolis,IN,2004.McInerney,F.A.andWing,S.L.:ThePaleocene-EoceneThermalMaximum:aperturbationofcarboncycle,climate,andbiospherewithimplicationsforthefuture,Annu.Rev.EarthPl.25Sc.,39,489–516,2011.NationalResearchCouncil:UnderstandingEarth'sDeepPast:LessonsforOurClimateFuture,TheNationalAcademiesPress,Washington,DC,2011.Olszewski,T.D.:Auniedmathematicalframeworkforthemeasurementofrichnessandeven-nesswithinandamongmultiplecommunities,Oikos,104,377–387,2004.30Pachauri,R.K.andReisinger,A.:ClimateChange2007–SynthesisReport,IPCCFourthAssessment,Geneva,Switzerland,2007.1394 CPD11,1371–1405,2015 MammalfaunalresponsetothePaleogenehyperthermalsETM2A.E.Chew TitlePage Abstract Introduction Conclusio

ns References Tables Figures J I J I Back Close FullScreen/Esc Printer-friendlyVersion InteractiveDiscussion DiscussionPaper|DiscussionPaper|DiscussionPaper|DiscussionPaper| Sluijs,A.,Schouten,S.,Donders,T.H.,Schoon,P.L.,Rohl,U.,Reichart,G.-J.,Sangiorgi,F.,Kim,J.-H.,SinningheDamste,J.S.,andBrinkhuis,H.:WarmandwetconditionsintheArcticregionduringEoceneThermalMaximum2,Nat.Geosci.,2,777–780,2009.Smith,J.J.,Hasiotis,S.T.,Kraus,M.J.,andWoody,D.T.:TransientdwarsmofsoilfaunaduringthePaleocene–EoceneThermalMaximum,P.Natl.Acad.Sci.USA,106,17655–517660,2009.Smith,M.E.,Singer,B.,andCarroll,A.:40Ar=39ArgeochronologyoftheEoceneGreenRiverformation,Wyoming,Reply,Geol.Soc.Am.Bull.,116,253–256,2004.Snell,K.E.,Thrasher,B.L.,Eiler,J.M.,Koch,P.L.,Sloan,L.C.,andTabor,N.J.:HotsummersintheBighornBasinduringtheearlyPaleogene,Geology,41,55–58,2013.10Stap,L.,Lourens,L.J.,Thomas,E.,Sluijs,A.,Bohaty,S.,andZachos,J.C.:High-resolutiondeep-seacarbonandoxygenisotoperecordsofEoceneThermalMaximum2andH2,Ge-ology,38,6

07–610,2010.Stassen,P.,Steurbaut,E.,Morsi,A.M.M.,Schulte,P.,andSpeijer,R.P.:BioticimpactofEoceneThermalMaximum2inashelfsetting(Dababiya,Egypt),Aust.J.EarthSci.,105,15154–160,2012.Tsukui,K.andClyde,W.C.:Fine-tuningthecalibrationoftheearlytomiddleEocenegeo-magneticpolaritytimescale:paleomagnetismofradioisotopicallydatedtusfromLaramideforelandbasins,Geol.Soc.Am.Bull.,124,870–885,2012.VanDeVelde,J.H.,Bowen,G.J.,Passey,B.H.,andBowen,B.B.:Climaticanddiagenetic20signalsinthestableisotopegeochemistryofdolomiticpaleosolsspanningthePaleocene-Eoceneboundary,Geochim.Cosmochim.Ac.,109,254–267,2013.Wilf,P.:LatePaleocene-earlyEoceneclimaticchangesinsouth-westernWyoming:paleob-otanicalanalysis,Geol.Soc.Am.Bull.,112,292–307,2000.Wing,S.L.,Bown,T.M.,andObradovich,J.D.:EarlyEocenebioticandclimaticchangein25interiorwesternNorthAmerica,Geology,9,1189–1192,1991.Wing,S.L.,Bao,H.,andKoch,P.L.:AnearlyEocenecoolperiod?EvidenceforcontinentalcoolingduringthewarmestpartoftheCenozoic,in:WarmClimatesinEarth

History,editedby:Huber,B.T.,MacLeod,K.G.,andWing,S.L.,CambridgeUniversityPress,Cambridge,197–237,2000.30Wing,S.L.,Harrington,G.J.,Smith,F.A.,Bloch,J.I.,Boyer,D.M.,andFreeman,K.H.:Tran-sientoralchangeandrapidglobalwarmingatthePaleocene-Eoceneboundary,Science,310,993–996,2005.1396 CPD11,1371–1405,2015 MammalfaunalresponsetothePaleogenehyperthermalsETM2A.E.Chew TitlePage Abstract Introduction Conclusions References Tables Figures J I J I Back Close FullScreen/Esc Printer-friendlyVersion InteractiveDiscussion DiscussionPaper|DiscussionPaper|DiscussionPaper|DiscussionPaper| Wood,H.E.:NomenclatureandcorrelationoftheNorthAmericancontinentalTertiary,Bull.Geol.Soc.Am.,52,1–48,1941.Woodburne,M.O.,Gunnell,G.F.,andStucky,R.K.:ClimatedirectlyinuencesEocenemam-malfaunaldynamicsinNorthAmerica,P.Natl.Acad.Sci.,106,13399–13403,2009.Yans,J.,Strait,S.G.,Smith,T.,Dupuis,C.,Steurbaut,E.,andGingerich,P.D.:High-resolution5carbonisotopestratigraphyandmammalianfaunalchangeatthePaleocene-Eocenebound-aryi

ntheHoneycombsareaofthesouthernBighornBasin,Wyoming,Am.J.Sci.,306,712–735,2006.Zachos,J.C.,Lohmann,K.C.,Walker,J.C.G.,andWise,S.W.:AbruptclimatechangeandtransientclimatesduringthePaleogene:amarineperspective,J.Geol.,101,191–213,1993.10Zachos,J.C.,Dickens,G.R.,andZeebe,R.E.:AnearlyCenozoicperspectiveongreenhousewarmingandcarbon-cycledynamics,Nature,451,279–283,2008.1397 CPD11,1371–1405,2015 MammalfaunalresponsetothePaleogenehyperthermalsETM2A.E.Chew TitlePage Abstract Introduction Conclusions References Tables Figures J I J I Back Close FullScreen/Esc Printer-friendlyVersion InteractiveDiscussion DiscussionPaper|DiscussionPaper|DiscussionPaper|DiscussionPaper| Figure1.Fossillocalitiesinthesouth-centralpartoftheBighornBasin.ColoredlocalitieshavebeentiedbymeterleveltotheFifteenmileCreekcompositestratigraphicsectionofBownetal.(1994b).CircledlocalitiesspantheETM2andH2hyperthermallevels(290–510m)intheFifteenmileCreeksection.IsotopesectionsrecordingtheETM2andH2CIEsintheMcCulloughPeaksarefromAb